Applicability of tetrazolium salts for the measurement of respiratory activity and viability of groundwater bacteria

A study was undertaken to measure aerobic respiration by indigenous bacteria in a sand and gravel aquifer on western Cape Cod, MA using tetrazolium salts and by direct oxygen consumption using gas chromatography (GC). In groundwater and aquifer slurries, the rate of aerobic respiration calculated from the direct GC assay was more than 600 times greater than that using the tetrazolium salt 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyl tetrazolium chloride (INT). To explain this discrepancy, the toxicity of INT and two additional tetrazolium salts, sodium 3'-[1-(phenylamino)-carbonyl]-3,4-tetrazolium]-bis(4-methoxy-6-nitro) benzenesulfonic acid hydrate (XTT) and 5-cyano-2,3-ditolyl tetrazolium chloride (CTC), to bacterial isolates from the aquifer was investigated. Each of the three tetrazolium salts was observed to be toxic to some of the groundwater isolates at concentrations normally used in electron transport system (ETS) and viability assays. For example, incubation of cells with XTT (3 mM) caused the density of four of the five groundwater strains tested to decline by more than four orders of magnitude. A reasonable percentage (>57%) of cells killed by CTC and INT contained visible formazan crystals (the insoluble, reduced form of the salts) after 4 h of incubation. Thus, many of the cells reduced enough CTC or INT prior to dying to be considered viable by microscopic evaluation. However, one bacterium (Pseudomonas fluorescens) that remained viable and culturable in the presence of INT and CTC, did not incorporate formazan crystals into more than a few percent of cells, even after 24 h of incubation. This strain would be considered nonviable based on traditional tetrazolium salt reduction assays. The data show that tetrazolium salt assays are likely to dramatically underestimate total ETS activity in groundwater and, although they may provide a reasonable overall estimate of viable cell numbers in a community of groundwater bacteria, some specific strains may be falsely considered nonviable by this assay due to poor uptake or reduction of the salts.     

An investigation of staining methods to determine total cell numbers and the number of respiring micro-organisms in samples obtained from the field and the laboratory

Abstract Dyes were evaluated in combination with 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyl tetrazolium chloride (INT) to enable total cell numbers and the numbers of respiring cells to be determined on the same preparation. Malachite green and 4',6-diamidino-2-phenylindole (DAPI) were unsuitable counter-stains. Cells which contained INT formazan crystals could be stained with ethidium bromide or auramine. At high concentrations of INT formazan, auramine fluorescence was reduced, although this effect was partially rectified by prior fixation with glutaraldehyde. Staining with ethidium bromide produced a strong fluorescence in cells containing crystals of INT formazan. This observation was developed into a procedure which allowed total cells to be determined and provided a useful estimate of the number of respiring cells in samples obtained from the laboratory and the field.     

Metabolic Status of Bacteria and Fungi in the Rhizosphere of Ponderosa Pine Seedlings

We determined the quantity and metabolic status of bacteria and fungi in rhizosphere and nonrhizosphere soil from microcosms containing ponderosa pine seedlings. Rhizosphere soil was sampled adjacent to coarse, fine, or young roots. The biovolume and metabolic status of bacterial and fungal cells was determined microscopically and converted to total and active biomass values. Cells were considered active if they possessed the ability to reduce the artificial electron acceptor 2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyltetrazolium chloride (INT) to visible intracellular deposits of INT formazan. A colorimetric assay of INT formazan production was also used to assess dehydrogenase activity. INT-active microorganisms made up 44 to 55% of the microbial biomass in the soils studied. The proportion of fungal biomass that exhibited INT-reducing activity (40 to 50%) was higher than previous estimates of the active proportion of soil fungi determined by using fluorescein diacetate. Comparison between soils from different root zones revealed that the highest total and INT-active fungal biomass was adjacent to fine mycorrhizal roots, whereas the highest total and active bacterial biomass was adjacent to the young growing root tips. These observations suggest that fungi are enhanced adjacent to the fine roots compared with the nonrhizosphere soil, whereas bacteria are more responsive than fungi to labile carbon inputs in the young root zone. Colorimetric dehydrogenase assays detected gross differences between bulk and rhizosphere soil activity but were unable to detect more subtle differences due to root types. Determination of total and INT-active biomass has increased our understanding of the role of spatial compartmentalization of bacteria and fungi in rhizosphere carbon flow.     

Permanent Presence of Grazing-Resistant Bacteria in a Hypertrophic Lake

The size structure of planktonic bacteria from a hypertrophic lake was investigated at 5- to 15-day intervals by means of a semiautomatic image analysis system during 1 year. Characteristic of this bacterial assemblage was the permanent presence of large filamentous bacteria and small cocci with cell sizes of <0.01 (mu)m(sup3). These filamentous bacteria, sometimes longer than 200 (mu)m and with cell volumes of up to 276 (mu)m(sup3), are larger than nanoflagellates (<20 (mu)m) and, even, metazoans living in the lake. Although they account for only 4 to 16% of bacterial abundance, their contribution to total bacterial biovolume was between 45 and 86%. An analysis of the food web structure indicates that this particular bacterial size structure may be the consequence of a strong bacterivory pressure by nanoflagellates and the absence of other larger bacterivores. The persistence of bacterial forms resistant to grazing has important consequences for the carbon flow within the microbial food web.     

The tetrazolium reduction method for assessing the viability of individual bacterial cells in aquatic environments: improvements,performance and applications

The electron transport system of respiring organisms reduces 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl tetrazolium chloride (INT) to INT-formazan. Active bacterial cells may be recognized under the microscope by epifluorescence and by the simultaneous presence, seen under bright light field of optically dense intracellular deposits of INT-formazan. An improved procedure that leads to a sharp definition of cells and formazan deposits is presented here. Cells are concentrated on cellulose membrane filters of 0.1 μm porosity which are rendered further transparent prior to immersion of the cells in a layer of 4′, 6-diaminidino-2-phenylindole (DAPI) s′ fluorochrome. This process leads to two significant improvements: (1) the fluorochrome is not trapped inside the membrane, which decreases the background fluorescence and leads to a better detection of the small cells; (2) the cells are immersed in an aqueous solution, which prevents rapid dissolution of the formazan crystals which would be expected if they were in contact with oily clearing agents. Tests on formazan labelling and on storage of INT-processed samples suggest other precautions for reliable use. Improved in this way, the method is simple, rapid and has numerous applications in environmental studies, ecophysiology and ecotoxicology. Some examples are given, with 2 to 98% of INT reducing cells observed, depending on different environmental conditions.     

Determination of microbial respiratory and redox activity in activated sludge

The tetrazolium salt 5-cyano-2,3-ditolyltetrazolium chloride (CTC) was used for the determination of metabolically active bacteria in active sludge. The method was adapted and optimized to the conditions of activated sludge. The colorless and nonfluorescent tetrazolium salt is readily reduced to a water-insoluble fluorescent formazan product via the microbial electron transport system and indicates mainly dehydrogenase activity. After more than 2 h incubation, no further formation of new formazan crystals was observed, although the existing crystals in active cells continued to grow at the optimal CTC-concentration of 4 mM. The dehydrogenase activity determined by direct epifluorescence microscopic enumeration did not correlate with cumulative measured activity as determined by formazan extraction. The addition of nutrients did not lead to an increase of CTC-active cells. Sample storage conditions such as low temperature or aeration resulted in a significant decrease in dehydrogenase activity within 30 min. The rapid and sensitive method is well suited for the detection and enumeration of metabolically active microorganisms in activated sludge. Extracellular redox activity was measured with the tetrazolium salt 3′-{1-[phenylamino-) carbonyl]-3,4-tetrazolium}-bis(4-methoxy-6-nitro)benzene-sulfonic acid hydrate (XTT), which remains soluble in its reduced state, after extraction of extracellular polymeric substances (EPS) with a cation exchange resin. Received 12 August 1996/ Accepted in revised form 29 May 1997     

Activity Measurements of Planktonic Microbial and Microfouling Communities in a Eutrophic Estuary

[³H]thymidine incorporation, the rate of reduction of iodonitrotetrazolium violet (INT) to INT formazan normalized to DNA, and the ratio of ATP to DNA were adapted to measure the activity of attached and unattached microbial assemblages of Bayboro Harbor, Fla. Activity measurements by [³H]thymidine incorporation were made of cells attached to polystyrene culture dishes, in unfiltered water samples, and in the <1-μm-filtered fraction. In most cases, the activity of attached cells was greater than that of unattached cells either in unfiltered water samples or in the <1-μm fraction. The calculated thymidine incorporation rates for cells in the >1-μm fraction were higher than those for cells either in unfiltered water or in the <1-μm-filtered fraction. By the rate of reduction of INT to INT formazan normalized to DNA and by ATP-to-DNA ratios, attached cells were also more active than cells in unfiltered water samples. These results indicate that the microenvironment afforded by attachment is a more beneficial habitat for microbial growth. Reasons for greater activity by natural populations of attached bacteria are discussed.     

Reduction of tetrazolium salts by sulfate-reducing bacteria

Abstract The reduction of tetrazolium salts by the sulfate-reducing bacteria, Desulfovibrio desulfuricans and Desulfotomaculum orientis , was examined. D. desulfuricans and D. orientis reduced triphenyltetrazolium chloride (TTC) and 2-( p -iodophenyl)-3-( p -nitrophenyl)-5-phenyltetrazolium chloride (INT) forming intracellular formazan deposits. The reduction rate of INT was higher than that of TTC. INT reduction was not inhibited by the addition of sulfate or molybdate, and sulfate uptake was inhibited by the addition of both INT and molybdate. The ratio of intracellular formazan forming cells to acridine orange direct counts in both strains decreased with culture age and starvation time.     

Tetrazolium reduction in Saccharomyces cerevisiae

Summary The tetrazolium salt, 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl tetrazolium chloride (INT) was used to determine viable respiring cells in batch cultures of Saccharomyces cerevisiae. Respiring cells reduce INT to water insoluble iodonitrotetrazolium formazan (INT-formazan) which is deposited within the respiring cell. The INT-formazan granules can be observed by brightfield microscopy. This allows a rapid quantitative determination of the percentage of respiring cells and total cells within the same microscopic field.In actively growing batch cultures of S. cerevisiae, the respiring cell count was equal to the total cell count for the first 72 h of the growth cycle. After 144 h of incubation only 22.7% of the total cell numbers were actively respiring.     

Assay of succinate dehydrogenase activity by the tetrazolium method: evaluation of an improved technique in skeletal muscle fractions

An improved spectrophotometric method for measuring succinate dehydrogenase (EC 1.3.99.1) activity with the use of 2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyltetrazolium chloride (INT) is described. The procedure has been evaluated in mitochondrial fractions and homogenates of frog skeletal muscle. For mitochondrial suspensions, extraction of formazan with alcohol was found to be superior to extraction with ethyl acetate. For homogenates, complete extraction of formazan required sequential treatment with alcohol and ethyl acetate; the generally employed procedure of extracting once with ethyl acetate alone led to serious underestimation of the amount of formazan in the tissue. Observations of mitochondrial suspension incubated with various concentrations of INT led to the selection of 0.8 mM INT for optimal results. Higher concentrations, although commonly used, can exert undesirable inhibitory effects on succinate dehydrogenase activity, especially at low concentrations of mitochondria and after longer periods of incubation. The problem of instability of succinate dehydrogenase was solved by the addition of buffer at pH 7.5.     

Activity of an Attached and Free-Living Vibrio sp. as Measured by Thymidine Incorporation, p-Iodonitrotetrazolium Reduction, and ATP/DNA Ratios

Three independent techniques, [³H]thymidine incorporation, the reduction rate of p-iodonitrotetrazolium violet (INT) to INT formazan normalized to DNA, and the ratio of ATP to DNA, were adapted to measure the activity of attached and unattached estuarine bacteria. In experiments employing the estuarine isolate Vibrio proteolytica, nutrient concentrations were manipulated by varying the concentration of peptone-yeast extract. In the presence of exogenous nutrients, the activity of free-living cells was greater than that of attached cells as measured by [³H]thymidine incorporation and ATP/DNA ratios. In the absence of peptone-yeast extract, however, the activity of attached cells surpassed that of free-living cells as determined by [³H]thymidine incorporation and INT formazan normalized to DNA. Of the three techniques, [³H]thymidine incorporation was deemed most sensitive for detecting changes in activity resulting from slight differences in nutrient concentration. By this technique, attached cells were much less sensitive to changing nutrient concentrations than were free-living cells. Below a threshold concentration, attached cell activity remained constant, while the activity of unattached cells decreased as a function of decreasing nutrient concentration. The results suggest that loss of cell surface area available for substrate uptake due to attachment may be an important factor in determining the relative activities of attached and free-living cells.     

Evaluation of the SOS/umu-test post-treatment assay for the detection of genotoxic activities of pure compounds and complex environmental mixtures

This study presents an evaluation of the SOS/umu-test after introducing an additional dilution and incubation in the post-treatment assay. This treatment reduces the influence of coloured test compounds that otherwise affect the colorimetric determination of the beta-galactosidase activity and the bacterial growth measurement during the testing of complex environmental samples. The post-treatment assay significantly increased the beta-galactosidase activity and consequently the enzyme induction ratios at higher doses of model genotoxins 4-nitroquinoline-N-oxide, N-methyl-N'-nitro-N-nitrosoguanidine, 2-aminoanthracene, benzo(a)pyrene with low or no effect on the sensitivity of the test itself. On the other hand tests of environmental extracts indicated significant increases in sensitivity after additional incubation. 4-Nitroquinoline-N-oxide treatments of bacteria in the test affected cell division and caused filamentous growth. The size of filamentous bacteria and incidence rate of the length categories was positively correlated with the concentrations of genotoxins. Presence of filamentous tester bacteria proved induction of SOS response and genotoxic activity of environment samples in SOS/umu-test.     

In situ classification and image cytometry of pelagic bacteria from a high mountain lake (gossenkollesee, austria)

We describe a procedure to measure the cell sizes of pelagic bacteria after determinative hybridization with rRNA-targeted fluorescently labeled oligonucleotide probes. Our approach is based on established image analysis techniques modified for objects simultaneously stained with two fluorescent dyes. It allows the estimation of biomass and cell size distribution and the morphological characterization of different bacterial taxa in plankton samples. The protocol was tested in a study of the bacterioplankton community of a high mountain lake during and after the ice break period. Cells that hybridized with a probe for the domain Bacteria accounted for 70% of the bacterial abundance (range, 49 to 83%) as determined by 4(prm1),6(prm1)-diamidino-2-phenylindole staining (K. G. Porter and Y. S. Feig, Limnol. Oceanogr. 25:943-948, 1980), but for >85% of the total biomass (range, 78 to 99%). The size distribution for members of the beta subclass of the Proteobacteria shifted toward larger cells and clearly distinguished this group from the total bacterial assemblage. In the surface water layer beneath the winter cover, bacteria belonging to the beta 1 subgroup constituted about one-half of the beta subclass abundance. The mean cell volume of the beta 1 subgroup bacteria was significantly less than that of the beta subclass proteobacteria, and the beta 1 subgroup accounted for less than 30% of the total beta subclass biovolume. Two weeks later, the biovolume of the beta Proteobacteria had decreased to the level of the beta 1 subgroup, and both the biovolume size distributions and cell morphologies of the beta Proteobacteria and the beta 1 subgroup were very similar. We could thus quantify the disappearance of large, morphologically distinct beta subclass proteobacteria which were not members of the beta 1 subgroup during the ice break period. Our results demonstrate that changes in biovolumes and cell size distributions of different bacterial taxa, and eventually of individual populations, reveal hitherto unknown processes within aquatic bacterial assemblages and may open new perspectives for the study of microbial food webs.     

Investigations of the localisation of bacterial activity on sandstone from ancient monuments

Methods were investigated for the determination of activity levels of bacteria on sandstone using the reduction of 2-(4-iodophenyl)-3-(4 nitrophenyl)-5-phenyl tetrazolium chloride (INT) to INT-formazan as a measure of dehydrogenase activity. A microscopy technique, based on use of acridine orange with bright-field illumination, was developed which gave a good visual image of bacterial cells, including those containing INT-formazan. Surveys at two monuments, Portchester Castle and Tintern Abbey, were carried out using this method which showed that between 20.7 and 51.9% of bacterial cells present in situ were active. Extraction of INT-formazan directly from the stone using methanol indicated that bacteria were tightly bound to stone particles and that microscopic methods would underestimate counts of bacteria. Surveys of five monuments using the extraction method showed that microbial populations on sandstone in situ were active but activity could not be related to decay state of the stone.     

Effects of substrates and phosphate on INT (2-(4-iodophenyl)-3-(4-nitrophenyl)-5-phenyl tetrazolium chloride) and CTC (5-cyano-2,3-ditolyl tetrazolium chloride) reduction in Escherichia coli

The effects of substrates of primary aerobic dehydrogenases, and inorganic phosphate on aerobic INT and CTC reduction in Escherichia coli were examined. In general, INT produced less formazan than CTC, but INT (+) cell counts remained near values of CTC (+) cells. INT and CTC (+) cell numbers were higher than plate counts on R2A medium using succinate, formate, lactate, casamino acids, glucose, glycerol (INT only) and no substrate. Formate resulted in the greatest amount of INT and CTC formazan. Reduction of both INT and CTC was inhibited above 10 mmol 1^-1 phosphate, and this appeared to be related to decreased rates of O₂ consumption. Formation of fluorescent CTC (+), but not INT (+) cells was also inhibited in a concentration dependent manner by phosphate above 10 mmol 1^-1. From light microscopic observations it appeared CTC formed increasing amounts of poorly or non-fluorescent formazan with increasing phosphate. Therefore, use of phosphate buffer in excess of 10 mmol 1^-1 may not be appropriate in CTC and INT reduction assays.     

Vital dye reaction and granule localization in periplasm of Escherichia coli

BACKGROUND: Tetrazolium salts are widely used in biology as indicators of metabolic activity - hence termed vital dyes - but their reduction site is still debated despite decades of intensive research. The prototype, 2,3,5- triphenyl tetrazolium chloride, which was first synthesized a century ago, often generates a single formazan granule at the old pole of Escherichia coli cells after reduction. So far, no explanation for their pole localization has been proposed. METHOD/PRINCIPAL FINDINGS: Here we provide evidence that the granules form in the periplasm of bacterial cells. A source of reducing power is deduced to be thiol groups destined to become disulfides, since deletion of dsbA, coding for thiol-oxidase, enhances the formation of reduced formazan. However, pervasive reduction did not result in a random distribution of formazan aggregates. In filamentous cells, large granules appear at regular intervals of about four normal cell-lengths, consistent with a diffusion-to-capture model. Computer simulations of a minimal biophysical model showed that the pole localization of granules is a spontaneous process, i.e. small granules in a normal size bacterium have lower energy at the poles. This biased their diffusion to the poles. They kept growing there and eventually became fixed. CONCLUSIONS: We observed that formazan granules formed in the periplasm after reduction of tetrazolium, which calls for re-evaluation of previous studies using cell-free systems that liberate inaccessible intracellular reductant and potentially generate artifacts. The localization of formazan granules in E. coli cells can now be understood. In living bacteria, the seeds formed at or migrated to the new pole would become visible only when that new pole already became an old pole, because of the relatively slow growth rate of granules relative to cell division.     

Role of Pore Size Location in Determining Bacterial Activity during Predation by Protozoa in Soil

The predation of a luminescence-marked strain of Pseudomonas fluorescens by the soil ciliate Colpoda steinii was studied in soil microcosms. Bacterial cells were introduced in either small (neck diameter, <6 (mu)m) or intermediate-sized (neck diameter, 6 to 30 (mu)m) pores in the soil by inoculation at appropriate matric potentials, and ciliates were introduced into large pores (neck diameter, 30 to 60 (mu)m). Viable cell concentrations of bacteria introduced into intermediate-sized pores decreased at a greater rate than those in small pores, with reductions in bacterial populations being accompanied by an increase in viable cell numbers of the ciliate. The data indicate that the location of bacteria in small pores provides significant protection from predation. In the absence of C. steinii, the level of metabolic activity of the bacterial population, measured by luminometry, decreased at a greater rate than cell number, and the level of luminescence cell(sup-1) consequently decreased. The decrease in levels of luminescence indicates a loss of activity due to starvation. During predation by C. steinii, the level of the activity of cells introduced into small pores fell in a similar manner. The level of cell activity was, however, significantly greater for cells introduced into intermediate-sized pores, despite their greater susceptibility to predation. The data suggest that increased activity arises from a release of nutrients by the predator and the greater accessibility of bacteria to nutrients in larger pores. Nutrient amendment of microcosms resulted in increases in bacterial populations to sustained, higher levels, while levels of luminescence increased transiently. The predation of cells introduced into intermediate-sized pores was greater, and there was also evidence that the level of activity of surviving bacteria was greater for bacteria in intermediate-sized but not small pores.     

Estimation of Frankia growth using Bradford protein and INT reduction activity estimations: application to inoculum standardization

Abstract The growth of Frankia spp. strain ORS 020607 in BAP medium was studied by using two methods simultaneously: determination of Bradford protein content and INT (2-( p -iodophenyl-3-( p -nitrophenyl)-5-phenyl tetrazolium chloride ) reduction activity (IRA). With the latter test, red formazan crystals formed intracellularly were extracted with methanol. Colouration intensity was estimated by absorbance spectrophotometry at 490 nm. The protein content and IRA of the culture were monitored for 96 days. IRA appeared to reflect the 'metabolically active' biomass of Frankia more accurately than the Bradford protein estimations.     

Advantages of distinguishing the active fraction in bacterioplankton assemblages: some examples

The difficulty of distinguishing between active and dormant or dead bacterial cells is an important problem for the aquatic microbiologist.Active cells can be detected under the microscope by the presence of an intact electron transport system able to reduce the colourless INT [2-(p-iodophenyl)-3-(p-nitrophenyl)-5-phenyl tetrazolium chloride] to an optically dense intracellular deposit.An improvement of this method has been applied to Lake Geneva and to a fish pond in the Ivory Coast. The portion of INT-reducing bacterial cells ranged from 1 to 71%, depending on place, depth, season and time of the day. In all cases bacterial activity, determined by uptake of ³H Thymidine or ¹⁴C glucose, and frequency of dividing cells were better correlated with the number of INT-reducing cells than with the total number of cells. This means that counts of cells able to reduce INT have a better metabolic significance than total cell counts. Some examples are developed which show the advantages of applying this method in cases where it is useful to distinguish active cells in a bacterial assemblage.     

Density gradient separation of active and non-active cells from natural environments

We present a method for the selective, physical separation of active and non-active bacterial cells from natural communities. The method exploits the reduction of tetrazolium salts to form insoluble formazan crystals intracellularly in response to the addition of different oxidisable substrates. The intracellular deposition of formazan alters the bouyant density of active cells enabling them to be separated by density gradient centrifugation. The method has been successfully applied to the fractionation and collection of large whole cell sub-populations of active and non-active cells from sea-water samples. Removal of the bands from the density gradient, followed by PCR amplification and DGGE analyses showed distinct differences in the PCR amplicon diversity associated with the active and non-active cell fractions; an indication of changes in bacterial community structure in response to the addition of oxidisable substrate. Thus, based on their in situ respiration potential, the approach enables the cytochemical enrichment and molecular characterisation of mixed bacterial populations in natural environments.