The influence of green fluorescent protein incorporation on bacterial physiology: a note of caution

Lipophilic AIEgens as the "Trojan Horse" with Discrepant Efficacy in Tracking and Treatment of Mycobacterial Infection

The highly contagious tuberculosis is a leading infectious killer, which urgently requires effective diagnosis and treatment methods. To address these issues, three lipophilic aggregation-induced emission (AIE) photosensitizers (TTMN, TTTMN, and MeOTTMN) are selected to evaluate their labeling and antimicrobial properties in vitro and in vivo. These three lipophilic AIEgens preserve low cytotoxicity and achieve real-time and non-invasive visualization of the process of mycobacteria infection in vitro and in vivo. More importantly, these AIEgens can be triggered by white light to produce reactive oxygen species (ROS), which is a highly efficient antibacterial reagent. Among these AIEgens, the TTMN photosensitizer has an outstanding antibacterial efficacy over the clinical first-line drug rifampicin at the same therapeutic concentration. Interestingly, this study also finds that TTMN can increase the expression of pro-inflammatory cytokines in the early stage of infection after light irradiation, indicating an additional pro-inflammatory role of TTMN. This work provides some feasibility basis for developing AIEgens-based agents for effectively destroying mycobacterium.

The Probiotic Bacillus hwajinpoensis Colonizes the Digestive System of Crassostrea gigas Larvae and Protects Them from Vibrio alginolyticus Infection

The Pacific oyster Crassostrea gigas is one of the most important cultured marine species around the world. Production of Pacific oysters in China has depended primarily on hatchery produced seeds since 2016, with the successful introduction and development of triploid oysters. However, the seed supply of Pacific oysters is threatened by recurring mass mortality events in recent years. Vibriosis is the most commonly encountered disease associated with intensive oyster culture in hatcheries and nurseries. Vibrio alginolyticus and Bacillus hwajinpoensis were the two strains with pathogenic and probiotic effects, respectively, identified during the Pacific oyster larvae production. To monitor their colonization process in Pacific oyster larvae, green fluorescent protein (GFP) and red fluorescent protein (RFP) were labeled to the pathogenic V. alginolyticus and the probiotic B. hwajinpoensis stain, respectively. The pathogenic and probiotic effects of the two strains during the colonization process were then assessed. Stabile expression of GFP and RFP were observed in corresponding stains, and the capabilities of growth, biofilm formation and in vitro adhesion of GFP- and RFP- tagged stains were not significantly different from those of the wild-type strains. Usage of probiotics of 105 CFU/mL significantly inhibited the growth of pathogenic V. alginolyticus and reduced the mortality of D-sharped larvae. Both the pathogenic and probiotic strains employed a similar route to enter and colonize the oyster larvae, which indicates that competing with pathogens for binding and spreading sites were one of the mechanisms of B. hwajinpoensis to provide the probiotic effects to oyster larvae. In summary, employment of fluorescence-tagged pathogenic and probiotic strains simultaneously provides us with an excellent bioassay model to investigate the potential mechanisms of probiotics.

Can only one physiological trait determinate the adverse effect of green fluorescent protein (GFP) incorporation on Vibrio virulence?

Green fluorescent protein (GFP) has been used extensively for in situ animal studies that follow up bacterial infection under epifluorescence microscopy. It is assumed that GFP is acting as a "neutral" protein with no influence on the bacterial physiology. To verify this hypothesis, the virulence of Vibrio splendidus ME9, Vibrio anguillarum NB10, and their respective GFP-tagged strains ME9-GFP and NB10-GFP (transconjugants) was compared in vitro and tested in vivo towards blue mussel (Mytilus edulis) larvae. Results showed that the incorporation of GFP negatively impacted the growth and swimming motility of NB10 in vitro. Correspondingly, the mRNA levels of genes involved in bacterial swimming motility (flaA, flaE, and cheR) were significantly down-regulated in NB10-GFP. As for the strain ME9 on the other hand, GFP incorporation only had a negative effect on swimming motility. However, both the strains NB10-GFP and ME9-GFP showed almost the same virulence as their respective parental strain towards mussel larvae in vivo. Overall, the data presented here demonstrated that incorporation of GFP may cause modifications in cell physiology and highlight the importance of preliminary physiological tests to minimize the negative influence of GFP tagging when it is used to monitor the target localization. The study also supports the idea that the virulence of Vibrio species is determined by complex regulatory networks. Notwithstanding the change of a single physiological trait, especially growth or swimming motility, the GFP-tagged Vibrio strain can thus still be considered usable in studies mainly focusing on the virulence of the strain. KEY POINTS: • The effect of GFP incorporation on physiological trait of Vibrio strains. • The virulence in vibrios could be multifactorial. • The stable virulence of Vibrio strains after GFP incorporation.

Dynamic Immune Response to Vibriosis in Pacific Oyster Crassostrea gigas Larvae during the Infection Process as Supported by Accurate Positioning of GFP-Tagged Vibrio Strains

As the immune system is not fully developed during the larval stage, hatchery culture of bivalve larvae is characterized by frequent mass mortality caused by bacterial pathogens, especially Vibrio spp. However, the knowledge is limited to the pathogenesis of vibriosis in oyster larvae, while the immune response to pathogenic microorganisms in this early life stage is still far from being fully elucidated. In this study, we combined green fluorescent protein (GFP)-tagging, histological and transcriptomic analyses to clarify the pathogenesis of experimental vibriosis and the mechanisms used by the host Pacific oyster Crassostrea gigas larvae to resist infection. The Vibrio strains first colonized the digestive system and rapidly proliferated, while only the transcription level of IκB kinase (IKK) and nuclear factor κB (NF-κB) associated with signaling transduction were up-regulated in oyster at 18 h post challenge (hpc). The mRNA levels for integrin β-1, peroxinectin, and heat shock protein 70 (HSP70), which are associated with phagocytosis, cell adhesion, and cytoprotection, were not upregulated until 30 hpc when the necrosis already happened in the larval digestive system. This suggested that the immunity in the early stages of C. gigas is not strong enough to prevent vibriosis and future research may focus on the strengthening of the gastrointestinal immune ability to defend vibriosis in bivalve larvae.

Designing a rigorous microscopy experiment: Validating methods and avoiding bias

Images generated by a microscope are never a perfect representation of the biological specimen. Microscopes and specimen preparation methods are prone to error and can impart images with unintended attributes that might be misconstrued as belonging to the biological specimen. In addition, our brains are wired to quickly interpret what we see, and with an unconscious bias toward that which makes the most sense to us based on our current understanding. Unaddressed errors in microscopy images combined with the bias we bring to visual interpretation of images can lead to false conclusions and irreproducible imaging data. Here we review important aspects of designing a rigorous light microscopy experiment: validation of methods used to prepare samples and of imaging system performance, identification and correction of errors, and strategies for avoiding bias in the acquisition and analysis of images.

Effects of Package Atmosphere and Storage Conditions on Minimizing Risk of Escherichia coli O157:H7 in Packaged Fresh Baby Spinach

Packaged fresh spinach has been associated with outbreaks of illness caused by Escherichia coli O157:H7. The purpose of this study was to assess the behavior of E. coli O157:H7 in packaged baby spinach in response to storage conditions of temperature and package atmosphere and including effects of inoculation level, spinach leaf damage (cut leaves), internalized or leaf surface contamination, exposure to hypochlorite sanitizer, and package size. Behavior of E. coli O157:H7 inoculated at 2 and 4 log CFU/g on spinach packaged in polymer bags composed of a two-layer laminate (polypropylene and polyethylene) and stored under atmospheres of 20% O₂-3% CO₂ and 0% O₂-15% CO₂ (aerobic and anaerobic, respectively) was assessed at 5, 7, 12, and 15°C for up to 14 days. Growth kinetics were calculated using DMFit software. Temperature decreases progressively diminished growth or survival of the pathogen, and an aerobic package atmosphere resulted in longer lag times (4 to 6 days) and lower population levels (0.2 to 1.4 log CFU/g) compared with the anaerobic atmosphere at 15°C. Internalized contamination, leaf cuts, or exposure to 100 ppm of hypochlorite did not result in changes in pathogen behavior compared with controls; however, a growth minimization trend consisting of longer lag times and lower population levels was repeatedly observed in the aerobic compared with the anaerobic package atmospheres. In contrast, growth of indigenous mesophiles and Enterobacteriaceae was unaffected by package atmosphere. Spinach stored at 5 to 7°C in two sizes (5 and 16 oz) of polyethylene terephthalate clamshell packages with ambient air atmospheres was more likely to progress to lower-oxygen conditions in 16-oz compared with 5-oz packages after 7 days of storage (P < 0.05). Practices to maintain aerobic conditions within the package, as well as storage of the package at low temperature, are ways to limit growth of E. coli O157:H7 in packaged spinach.

Rapid loss of a green fluorescent plasmid in Escherichia coli O157:H7

Plasmids encoding green fluorescent protein (GFP) are frequently used to label bacteria, allowing the identification and differentiation from background flora during experimental studies. Because of its common use in survival studies of the foodborne pathogen Escherichia coli O157:H7, it is important to know the extent to which the plasmid is retained in this host system. Herein, the stability of a pGFPuv (Clontech Laboratories Inc) plasmid in six Escherichia coli O157:H7 isolates was assessed in an oligotrophic environment (phosphate buffered saline, PBS) without antibiotic selective pressure. The six test isolates were recovered from a variety of animal and human sources (cattle, sheep, starlings, water buffalo, and human feces). GFP labeling of the bacteria was accomplished via transfer electroporation. The stability of the GFP plasmid in the different E. coli O157:H7 isolates was variable: in one strain, GFP plasmid loss was rapid, as early as one day and complete plasmid loss was exhibited by four of the six strains within 19 days. In one of the two isolates retaining the GFP plasmid beyond 19 days, counts of GFP-labeled E. coli O157:H7 were significantly lower than the total cell population (P < 0.001). In contrast, in the other isolate after 19 days, total E. coli O157:H7 counts and GFP-labeled E. coli counts were equivalent. These results demonstrate strain-to-strain variability in plasmid stability. Consequently the use of GFP-labeled E.coli O157:H7 in prolonged survival studies may result in the underestimation of survival time due to plasmid loss.

Studies in vitro on infectivity and sensitivity to antileishmanial drugs in New World Leishmania species transfected with the green fluorescent protein [pIR3(-)-eGFP]

Current chemotherapeutic agents for leishmaniasis have several disadvantages interfering with the effective treatment and therefore more and better antileishmanial drugs are needed. Discovery of candidates for leishmaniasis treatment requires not only accurate and precise methodologies but also well-known biological system to measure infectivity of parasites and antileishmanial activity of the new compounds. Significant variation in the in vitro and in vivo infectivity and sensitivity to established and experimental drugs in Leishmania strains are reported. This work reports the in vitro biological behavior and antileishmanial drugs sensitivity of different green fluorescent protein transfectant Leishmanias strains. The in vitro growth kinetic and infectivity to U937 cells vary slightly in the Leishmania transfectant strains in comparison with their correspondant wild-type. However, the insertion of the pIR3(-)-eGFP may affect the sensitivity of the parasites to meglumine antimoniate (MA) and miltefosine but not to amphotericin B (AMB) and pentamidine isethionate. In consequence, AMB or pentamidine isethionate but not MA or miltefosine should be used as antileishmanial control drugs during in vitro assays of antileishmanial activity. Furthermore, is recommended to test compounds against more than one Leishmania strain in order to verify that the antileihmanial activity of these compound is similar among species.

The use of stable and unstable green fluorescent proteins for studies in two bacterial models: Agrobacterium tumefaciens and Xanthomonas campestris pv. campestris

Fluorescent proteins have been used to track plant pathogens to understand their host interactions. To be useful, the transgenic pathogens must present similar behaviour than the wild-type isolates. Herein, a GFP marker was used to transform two plant pathogenic bacteria, Agrobacterium and Xanthomonas, to localize and track the bacteria during infection. The transgenic bacteria were evaluated to determine whether they showed the same fitness than the wild-type strains or whether the expression of the GFP protein interfered in the bacterial activity. In Agrobacterium, the plasmid used for transformation was stable in the bacteria and the strain kept the virulence, while Xanthomonas was not able to conserve the plasmid and transformed strains showed virulence variations compared to wild-type strains. Although marking bacteria with GFP to track infection in plants is a common issue, works to validate the transgenic strains and corroborate their fitness are not usual. Results, presented here, confirm the importance of proper fitness tests on the marked strains before performing localization assays, to avoid underestimation of the microbe population or possible artificial effects in its interaction with the plant.

Individual cell heterogeneity in Predictive Food Microbiology: Challenges in predicting a "noisy" world

Gene expression is a fundamentally noisy process giving rise to a significant cell to cell variability at the phenotype level. The phenotypic noise is manifested in a wide range of microbial traits. Heterogeneous behavior of individual cells is observed at the growth, survival and inactivation responses and should be taken into account in the context of Predictive Food Microbiology (PMF). Recent methodological advances can be employed for the study and modeling of single cell dynamics leading to a new generation of mechanistic models which can provide insight into the link between phenotype, gene-expression, protein and metabolic functional units at the single cell level. Such models however, need to deal with an enormous amount of interactions and processes that influence each other, forming an extremely complex system. In this review paper, we discuss the importance of noise and present the future challenges in predicting the "noisy" microbial responses in foods.

Colonization of the Intestinal Tract of the Polyphagous Pest Spodoptera littoralis with the GFP-Tagged Indigenous Gut Bacterium Enterococcus mundtii

The alkaline gut of Lepidopterans plays a crucial role in shaping communities of bacteria. Enterococcus mundtii has emerged as one of the predominant gut microorganisms in the gastrointestinal tract of the major agricultural pest, Spodoptera littoralis. Therefore, it was selected as a model bacterium to study its adaptation to harsh alkaline gut conditions in its host insect throughout different stages of development (larvae, pupae, adults, and eggs). To date, the mechanism of bacterial survival in insects' intestinal tract has been unknown. Therefore, we have engineered a GFP-tagged species of bacteria, E. mundtii, to track how it colonizes the intestine of S. littoralis. Three promoters of different strengths were used to control the expression of GFP in E. mundtii. The promoter ermB was the most effective, exhibiting the highest GFP fluorescence intensity, and hence was chosen as our main construct. Our data show that the engineered fluorescent bacteria survived and proliferated in the intestinal tract of the insect at all life stages for up to the second generation following ingestion.

First evidence for a Vibrio strain pathogenic to Mytilus edulis altering hemocyte immune capacities

Bacterial isolates were obtained from mortality events affecting Mytilus edulis and reported by professionals in 2010-2013 or from mussel microflora. Experimental infections allowed the selection of two isolates affiliated to Vibrio splendidus/Vibrio hemicentroti type strains: a virulent 10/068 1T1 (76.6% and 90% mortalities in 24 h and 96 h) and an innocuous 12/056 M24T1 (0% and 23.3% in 24 h and 96 h). These two strains were GFP-tagged and validated for their growth characteristics and virulence as genuine models for exposure. Then, host cellular immune responses to the microbial invaders were assessed. In the presence of the virulent strain, hemocyte motility was instantaneously enhanced but markedly slowed down after 2 h exposure. By contrast, hemocyte velocity increased in the presence of the innocuous 12/056 M24T1. At the same time interval, 10/068 1T1 invaded hemocytes and was more rapidly internalized than the innocuous strain. Extracellular products (ECPs) prepared from 10/068 1T1 cultures significantly inhibited phagocytic activity while 12/056 M24T1 ECPs had no effect. Furthermore, the pathogenic strain and its ECPs inhibited oxidative burst unlike 12/056 M24T1 strain/ECPs which enhanced ROS production. Taken together, our results suggest that the mussel pathogen 10/068 1T1 may escape immune response by altering hemocytes functions.

Construction and characterization of outbreak Escherichia coli O157:H7 surrogate strains for use in field studies

Escherichia coli O157:H7 has been the causative agent of many outbreaks associated with leafy green produce consumption. Elucidating the mechanism by which contamination occurs requires monitoring interactions between the pathogen and the plant under typical production conditions. Intentional introduction of virulent strains into fields is not an acceptable practice. As an alternative, attenuated strains of natural isolates have been used as surrogates of the virulent strains; however, the attachment properties and environmental stabilities of these attenuated isolates may differ from the unattenuated outbreak strains. In this study, the Shiga toxin (stx1, stx2, and/or stx2c) genes as well as the eae gene encoding intimin of two E. coli O157:H7 outbreak isolates, F4546 (1997 alfalfa sprout) and K4492 (2006 lettuce), were deleted. Individual gene deletions were confirmed by polymerase chain reaction (PCR) and DNA sequencing. The mutant strains did not produce Shiga toxin. The growth kinetics of these mutant strains under nutrient-rich and minimal conditions were identical to those of their wild-type strains. Attachment to the surface of lettuce leaves was comparable between wild-type/mutant pairs F4546/MD46 and K4492/MD47. Adherence to soil particles was also comparable between the virulent and surrogate pairs, although the F4546/MD46 pair exhibited statistically greater attachment than the K4492/MD47 pair (p≤0.05). Wild-type and mutant pairs F4546/MD46 and K4492/MD47 inoculated into wet or dry soils had statistically similar survival rates over the 7-day storage period at 20°C. A plasmid, pGFPuv, containing green fluorescent protein was transformed into each of the mutant strains, allowing for ease of identification and detection of surrogate strains on plant material or soil. These pGFPuv-containing surrogate strains will enable the investigation of pathogen interaction with plants and soil in the farm production environment where the virulent pathogen cannot be used.

Biosafety assessment of GFP transplastomic tobacco to rhizosphere microbial community

Green fluorescent protein (GFP) is one of the most widely studied and exploited proteins in biochemistry, and has many applications as a marker, especially in plant transformation system. Although a number of studies have been conducted to assess the toxify of this protein to specific organisms, little is known about GFP on rhizosphere microbial community, which is regarded as good indicator for environmental risk assessment. Chloroplast genetic engineering has shown superiority over traditional nuclear genetic engineering, and has been used in many aspects of plant genetic engineering. High levels of chloroplast-based protein accumulation make this technology as an ideal strategy to evaluate biosafety of transgenes. In the present study, the effects of field-released GFP transplastomic tobacco (Nicotiana tabacum) on rhizosphere microbes over a whole growth cycle were investigated by using both culture-dependent and culture-independent methods. Compared to wild-type control, transplastomic tobacco had no significant influence on the microbial population at the seedling, vegetative, flowering and senescing stages. However, developmental stages had more influence than ecotypes (GFP-transformed and wild-type). This was confirmed by colony forming unit, Biolog Eco(TM) and PCR-DGGE analysis. Thus, these results suggest chloroplast transformation with a GFP reporter gene has no significant influence on rhizosphere microbial community, and will be potential platform for plant biotechnology in future.

Survival of GFP-tagged Rhodococcus sp. D310-1 in chlorimuron-ethyl-contaminated soil and its effects on the indigenous microbial community

The recently isolated bacterial strain Rhodococcus sp. D310-1 can degrade high concentrations of chlorimuron-ethyl (up to 1000 mg L(-1)), indicating its potential for the bioremediation of soil contaminated with high levels of chlorimuron-ethyl. In this study, Rhodococcus sp. D310-1 was tagged with green fluorescent protein gene (gfp) to track its survival in soil. Subsequently, degradation activity of the gfp-tagged strain and its effects on indigenous microbial community were analyzed. Results showed the cell numbers of Rhodococcus sp. D310-1::gfp in non-sterilized soil maintained at 8.5 × 10(4) cells g(-1) dry soil 45 days after inoculation of 7.74 × 10(6) cells g(-1) dry soil and approximately 49% of chlorimuron-ethyl was removed. However, The cell numbers of Rhodococcus sp. D310-1::gfp in sterilized samples increased gradually to 7.85 × 10(7) cells g(-1) dry soil and approximately 78% of chlorimuron-ethyl was removed. PCR-DGGE demonstrated that inoculation of this gfp-tagged strain in chlorimuron-ethyl-contaminated soil has negligible impact on the community structure of bacteria, actinomycetes and fungi. These results indicate that Rhodococcus sp. D310-1 is effective for the remediation of chlorimuron-ethyl-contaminated soil and also provides valuable information about the behavior of the inoculant population during bioremediation, which could be directly used in the risk assessment of inoculant population and optimization of bioremediation process.

Bacterial host interaction of GFP-labelled Vibrio anguillarum HI-610 with gnotobiotic sea bass, Dicentrarchus labrax (L.), larvae

The location and cell damage caused by Vibrio anguillarum, the causative agent of classical vibriosis, within the developing gut of the newly hatched sea bass, Dicentrarchus labrax (L.), is unknown. A gnotobiotic sea bass model was used to investigate the early interactions of V. anguillarum with sea bass larvae. In the present study, germ-free sea bass larvae were orally exposed to a V. anguillarum HI-610 pathogen labelled with the green fluorescent protein (GFP-HI-610) and sampled at regular intervals. Pathogenic colonization of gut enterocytes was observed 2 h post-exposure (p.e.) and onwards, whereas bacteria within the swim bladder were visualized 48 h p.e and onwards. Ultrastructural findings demonstrated direct bacterial contact with the host cell in the oesophageal mucosa and putative attachment to microvilli of mid- and hindgut enterocytes. The present findings form a starting point for studies assessing the impact of potential candidates (probiotics, prebiotics, antimicrobial peptides) to mitigate bacterial virulence.

Green fluorescent protein labeling of Listeria, Salmonella, and Escherichia coli O157:H7 for safety-related studies

Many food safety-related studies require tracking of introduced foodborne pathogens to monitor their fate in complex environments. The green fluorescent protein (GFP) gene (gfp) provides an easily detectable phenotype so has been used to label many microorganisms for ecological studies. The objectives of this study were to label major foodborne pathogens and related bacteria, including Listeria monocytogenes, Listeria innocua, Salmonella, and Escherichia coli O157:H7 strains, with GFP and characterize the labeled strains for stability of the GFP plasmid and the plasmid's effect on bacterial growth. GFP plasmids were introduced into these strains by a CaCl₂ procedure, conjugation or electroporation. Stability of the label was determined through sequential propagation of labeled strains in the absence of selective pressure, and rates of plasmid-loss were calculated. Stability of the GFP plasmid varied among the labeled species and strains, with the most stable GFP label observed in E. coli O157:H7. When grown in nonselective media for two consecutive subcultures (ca. 20 generations), the rates of plasmid loss among labeled E. coli O157:H7, Salmonella and Listeria strains ranged from 0%–30%, 15.8%–99.9% and 8.1%–93.4%, respectively. Complete loss (>99.99%) of the plasmid occurred in some labeled strains after five consecutive subcultures in the absence of selective pressure, whereas it remained stable in others. The GFP plasmid had an insignificant effect on growth of most labeled strains. E. coli O157:H7, Salmonella and Listeria strains can be effectively labeled with the GFP plasmid which can be stable in some isolates for many generations without adversely affecting growth rates.

Interaction of Vibrio cholerae non-O1/non-O139 with copepods, cladocerans and competing bacteria in the large alkaline lake Neusiedler See, Austria

Vibrio cholerae is a human pathogen and natural inhabitant of aquatic environments. Serogroups O1/O139 have been associated with epidemic cholera, while non-O1/non-O139 serogroups usually cause human disease other than classical cholera. V. cholerae non-O1/non-O139 from the Neusiedler See, a large Central European lake, have caused ear and wound infections, including one case of fatal septicaemia. Recent investigations demonstrated rapid planktonic growth of V. cholerae non-O1/non-O139 and correlation with zooplankton biomass. The aim of this study was to elucidate the interaction of autochthonous V. cholerae with two dominant crustacean zooplankton species in the lake and investigate the influence of the natural bacterial community on this interaction. An existing data set was evaluated for statistical relationships between zooplankton species and V. cholerae and co-culture experiments were performed in the laboratory. A new fluorescence in situ hybridisation protocol was applied for quantification of V. cholerae non-O1/non-O139 cells, which significantly reduced analysis time. The experiments clearly demonstrated a significant relationship of autochthonous V. cholerae non-O1/non-O139 with cladocerans by promoting growth of V. cholerae non-O1/non-O139 in the water and on the surfaces of the cladocerans. In contrast, copepods had a negative effect on the growth of V. cholerae non-O1/non-O139 via competing bacteria from their surfaces. Thus, beside other known factors, biofilm formation by V. cholerae on crustacean zooplankton appears to be zooplankton taxon specific and may be controlled by the natural bacterial community.

Profiling of external metabolites during production of hantavirus nucleocapsid protein with recombinant Saccharomyces cerevisiae

Recombinant strains of Saccharomyces cerevisiae, producing hantavirus Puumala nucleocapsid protein for diagnostics and as a candidate vaccine were analyzed for uptake and excretion of intermediary metabolites during process optimization studies of fed-batch bioreactor cultures. Concentrations of glucose, maltose, galactose, pyruvate, acetaldehyde, ethanol, acetate, succinate and formaldehyde (used as a selection agent) were measured in the culture medium in order to find a metabolite pattern, indicative for the physiological state of the producer culture. When the inducer galactose was employed as a growth substrate, the metabolite profile of recombinant yeast cells was different from those of the non-recombinant original strain which excreted considerable amounts of metabolites with this substrate. In contrast, galactose-induced heterologous gene expression was indicated by the absence of excreted intermediary metabolites, except succinate. A model strain expressing a GFP fusion of hantavirus nucleocapsid protein differed in the excretion of metabolites from strains without GFP. In addition, the influence of alkali ions, employed for pH control is also demonstrated.