Rapid microspectrofluorimetry for biochemical and metabolic studies in single living cells

A rapid multichannel microspectrofluorometer (e.g., to NAD(P)H, fluorescent probes) can be operated on a topographic mode for the evaluation of intracellular metabolic topography or on a spectral mode for the individual or simultaneous intracellular spectral analysis of various fluorochromes. The fluorescence emission spectra of the living cells, as well as difference spectra (spectra after intracellular microelectrophoretic addition of substrate minus before)_are analyzed under various conditions, and provide a direct proof that the fluorescence observed is that of NAD(P)H. The spectral changes which accompany treatment with substrate (e.g., glucose-6-P) can be further followed in cells incubated with other probes (e.g., acridine orange). Repeated and quite reversible transients of NAD(P) reduction—reoxidation may be observed in cells having absorbed acridine orange following repetitive additions of substrate. The spectral response to substrate is also comparatively studied in cells grown in presence of agents affecting the cell cycle (e.g., dibutyryl cyclic AMP, bleomycin).     

Use of 1,4 diacetoxy-2,3 dicyanobenzol in microspectrofluorometry for determining the intracellular pH of isolated living cells

Microspectrofluorometry allows to obtain the fluorescence spectrum of an isolated living cell. When cells are preincubated with 1,4 diacetoxy-2,3 dicyanobenzol the cellular fluorescence spectrum can be resolved in its components i.e. the characteristic fluorescence spectrum of each ionized forms of the probe and the intrinsic cell fluorescence spectrum due to NAD(P)H. This allows the determination of the intracellular pH with good accuracy. Furthermore, comparison between the intensity of the intrinsic cell fluorescence and the probe fluorescence intensity offers us an opportunity to monitor the intracellular amount of the drug.     

Rapid microspectrofluorometric studies in EL2 cells following intracellular accumulation of dibenzocarbazoles.

Microspectrofluorometric observations were carried out in EL2 ascites cancer cells and dibenzo(a,e)fluoranthene (diB(a,e)F)-grown EL2 cells, following treatment (5 min) with three dibenzocarbazoles (1,2,7,8; 1,2,5,6 and 3,4,5,6). After microinjection of glucose-6-P leading to reduction of NAD(P), a sequence of difference spectra (after substrate minus before) is recorded. In dibenzocarbazole-untreated cells, maximum (NAD(P) reduction (emission maximum at 465-475 nm) is attained within 5 s, followed by a gradual return to initial fluorescence within 20 to 200 s (faster in the diB(a,e)F-grown). In dibenzocarbazole-treated cells there is a rather regular increase in the intensity of the difference spectrum up to approximately 300-500 s. Initially the increase is more predominant in the region around 460-470 nm, but it gains later prominence in the shorter wavelength region (420-430 nm) characteristic of the hydrocarbon (higher and steadier increase in the 3,4,5,6, dibenzocarbazole-treated diB(a,e)F-grown). Subsequently there is a gradual decrease of fluorescence which may or may or not return to initial level. The observed increase spectra require evaluation in terms of possible components (e.g. a mixture of NAD(P)H and hydrocarbon, binding changes, succession of fluorescent metabolites).     

Rapid microspectrofluorometric studies in EL2 cells following intracellular accumulation of dibenzocarbazoles

Summary Microspectrofluorometric observations were carried out in EL2 ascites cancer cells and dibenzo(a,e)fluoranthene (diB(a,e)F)-grown EL2 cells, following treatment (5 min) with three dibenzocarbazoles (1,2,7,8; 1,2,5,6 and 3,4,5,6). After microinjection of glucose-6-P leading to reduction of NAD(P), a sequence of difference spectra (after substrate minus before) is recorded. In dibenzocarbazole-untreated cells, maximum NAD(P) reduction (emission maximum at 465–475 nm) is attained within 5 s, followed by a gradual return to initial fluorescence within 20 to 200 s (faster in the diB(a,e)F-grown). In dibenzocarbazole-treated cells there is a rather regular increase in the intensity of the difference spectrum up to 300–500 s. Initially the increase is more predominant in the region around 460–470 nm, but it gains later prominence in the shorter wavelength region (420–430 nm) characteristic of the hydrocarbon (higher and steadier increase in the 3,4,5,6, dibenzocarbazole-treated diB(a,e)F-grown). Subsequently there is a gradual decrease of fluorescence which may or may not return to initial level. The observed increase spectra require evaluation in terms of possible components (e.g. a mixture of NAD(P)H and hydrocarbon, binding changes, succession of fluorescent metabolites).     

Microspectrofluorometry of autofluorescence emission from human leukemic living cells under oxidative stress.

Image cytometry was applied to study the intracellular localization of autofluorescence and the influence of an oxidative stress on this emission. K562 erythroleukemia cancer cells were analyzed with a microspectrofluorometer, coupled with a Argon laser (Ar+) (363 nm). From each cell, 15 x 15 emission spectra were recorded in the 400-600 nm spectral range to generate a spectral image of autofluorescence. The intracellular locations of the autofluorescence emission and of the specific mitochondrial probe rhodamine 123 (R123) were matched. Under a 363 nm excitation, all spectra from K562 cells show equivalent profiles with a 455 nm maximum emission, near of reduced nicotinamide adenine dinucleotide-(Phosphate) solution (NAD(P)H) (465 nm maximum emission). The spatial distribution of autofluorescence is homogeneous and different from the one of R123. Hydrogen peroxide (H2O2) (200 microM) and menadione (Men) (5 microM) induce a weak spectral change and a decrease in autofluorescence intensity, down to 40% of the initial emission. Doxorubicin (Dox) induces a dose-dependent decrease in autofluorescence emission and a release of intracellular free radicals. When cells were pre-treated 1 h with 1 mM glutathione (GSH), Dox induces a lower free radicals release, no significant variation of autofluorescence intensity and a lower growth inhibitory effect. Images cytometry of autofluorescence suggest that the intracellular NAD(P)H would not be restricted to mitochondrial compartments. The release of free radicals was associated with a decrease in autofluorescence intensity, mainly attributed to NAD(P)H oxidation both inside and outside mitochondria.     

Plant cell culture monitoring using an in situ multiwavelength fluorescence probe

A multiwavelength fluorescence probe is proposed for in situ monitoring of Eschscholtzia californica and Catharanthus roseus plant cell cultures. The potential of the probe as a tool for real-time estimation of biomass and production in secondary metabolites has been studied. The probe excitation range is 270-550 nm and the emission range is 310-590 nm, with a step of 20 nm for both excitation and emission filters. Many endogenous fluorophores such as NAD(P)H, riboflavins (riboflavin and derivatives such as FMN, FAD), tryptamine and tryptophan, and fluorescent secondary metabolites were analyzed simultaneously. NAD(P)H fluorescence signal (350/450 nm) showed to be an adequate signal for estimating cells activity. Riboflavins fluorescence signal (450/530 nm) followed C. roseus cell concentration both for the growth phase and after elicitation with jasmonic acid. Fluorescence from the alkaloids interfered with NAD(P)H signal during the production phase. For C. roseus, tryptophan, tryptamine, ajmalicine and serpentine were monitored by the probe. For E. californica, fluorescence from alkaloids overlapped with riboflavins preventing from using the probe to follow cell growth but global alkaloids production could be followed using the probe.     

Two-photon fluorescence spectroscopy and microscopy of NAD(P)H and flavoprotein

Two-photon (2P) ratiometric redox fluorometry and microscopy of pyridine nucleotide (NAD(P)H) and flavoprotein (FP) fluorescence, at 800-nm excitation, has been demonstrated as a function of mitochondrial metabolic states in isolated adult dog cardiomyocytes. We have measured the 2P-excitation spectra of NAD(P)H, flavin adenine dinucleotide (FAD), and lipoamide dehydrogenase (LipDH) over the wavelength range of 720-1000 nm. The 2P-excitation action cross sections (sigma2P) increase rapidly at wavelengths below 800 nm, and the maximum sigma2P of LipDH is approximately 5 and 12 times larger than those of FAD and NAD(P)H, respectively. Only FAD and LipDH can be efficiently excited at wavelengths above 800 nm with a broad 2P-excitation band around 900 nm. Two autofluorescence spectral regions (i.e., approximately 410-490 nm and approximately 510-650 nm) of isolated cardiomyocytes were imaged using 2P-laser scanning microscopy. At 750-nm excitation, fluorescence of both regions is dominated by NAD(P)H emission, as indicated by fluorescence intensity changes induced by mitochondrial inhibitor NaCN and mitochondria uncoupler carbonyl cyanide p-(trifluoromethoxy) phenyl hydrazone (FCCP). In contrast, 2P-FP fluorescence dominates at 900-nm excitation, which is in agreement with the sigma2P measurements. Finally, 2P-autofluorescence emission spectra of single cardiac cells have been obtained, with results suggesting potential for substantial improvement of the proposed 2P-ratiometric technique.     

Role of calcium in the reaction between pyrroloquinoline quinone and pyridine nucleotides monomers and dimers.

Redox reactions were carried out in aerobiosis and anaerobiosis between NAD(P) dimers or NAD(P)H and pyrroloquinoline quinone (PQQ) in different buffers. The buffer system and pH significantly affected the oxidation rates of nucleotides and the ESR signal intensity of the PQQ(*) radical formed in anaerobiosis by comproportion between the quinone and quinol forms. The relative reactivity of the four nucleotides toward PQQ was affected by pH and buffer nature. PQQ, which behaves as an electron shuttle from nucleotides to oxygen, was first converted to PQQH(2) and then rapidly reoxidized by oxygen, with formation of hydrogen peroxide. Both NAD(P) dimers and NAD(P)H consumed 1 mol of oxygen per mole of reacted molecule of pyridine nucleotide, yielding 1 or 2 mol of NAD(P)(+) from NAD(P)H or from NAD(P) dimers, respectively. Chelating agents such as EDTA and phytate strongly decreased the reaction rate and the PQQ(*) radical signal intensity. Kinetics carried out in the presence of metal ions showed instead an increased reaction rate in the order Ca(2+) > Mg(2+) > Na(+) > K(+). Spectrofluorimetric measurements of PQQ with increasing concentrations of Ca(2+) showed a fluorescence quenching and shift of the maximum emission toward lower wavelengths, while other metal ions showed minor effects, if any. Therefore, it is demonstrated that Ca(2+) binds to PQQ, probably forming a complex which is more reactive with both one-electron (NAD(P) dimers) or two-electron donors (NAD(P)H) in nonenzymic reactions. It is important to recall that Ca(2+) was already found to play active role in PQQ-containing enzymes.     

The effect of atebrine and an acridine analog (BCMA) on the coenzyme fluorescence spectra of cultured melanoma and Ehrlich ascites (EL2) cells.

Coenzyme fluorescence spectra of single living cells are due to free pyridine nucleotides (folded configuration), bound pyridine nucleotides (unfolded configuration) and a third component, possibly a mixture or flavins. Such spectra can be used to recognize possible differences in coenzyme composition between cell lines or changes of metabolic pathways due to chemicals acting at levels below or above cytotoxicity, by high resolution spectrofluorometry. A study of spectra recorded from cultured Ehrlich ascites (EL2), and Harding Passey melanoma cells (HPM-67 and HPM-73 line) grown under comparable conditions, shows that free NAD(P)H predominates in HPM-67 and EL2, while this coenzyme is bound in HPM-73. The free/bound ratio may be profoundly modifed by chemicals, e.g. in the HPM-73 increase of free and decrease of bound NAD(P)H occurred upon treatment with 10(-6) oligomycin. When atebrine at levels (10(-6) M) below cytotoxicity was added, there was a decrease of the free NAD(P)H spectrum possibly through energy transfer from NAD(P)H to atebrine. Consideration of long range energy transfer i.e., excitation of atebrine by fluorescence of NAD(P)H vs. short range transfer of excitation energy from free NAD(P)H to atebrine, favors the latter mechanism. A transient (reversible) increase in atebrine fluorescence is seen following intracellular microinjection of substrate (e.g. glucose-6-P) leading to an increase in free NAD(P)H. At cytotoxic levels of atebrine (e.g 2 x 10(-5) M) an irreversible increase of atebrine fluorescence is seen. The microspectrofluorometric technique appears therefore well suited to study physiological processes at the level of intracellular coenzymes, as well as possible processes of intermolecular energy transfer in the microenvironment.     

Rapid multichannel microfluorometry for metabolic and spectral studies in single living cells

A multichannel microspectrofluorometer has been developed for operation on two modes, a ‘morphological’ mode for the assay of intracellular fluorochromes (e.g. NAD(P)H) in correlation with topography, and a ‘spectral’ mode for wavelength analysis of natural cell fluorescence. This instrument is based on an electron bombardment silicon camera tube (EBS) operated in conjunction with a multiscaling computer. The total NAD(P)H emission from 2 × 30 μm cell strips can be analysed in real time (32.8 msec frame scan) with a signal-to-noise ratio over 100:1. The metabolic changes in cytoplasmic regions are compared with those in regions comprising cytoplasm + nucleus, where the major contribution may be nuclear (cf earlier studies). The observation of a ‘multilocalized’ and asynchronous metabolic response is facilitated with substrates such as glucose-1-phosphate, associated with a longer lag period before the initiation of fluorescence changes. The latter largely occur in the 440–480 nm region. Fluorescence spectra recorded from intracellular regions are nearly super-posable to the spectrum obtained from NAD(P)H crystals.     

Ultraviolet-induced autofluorescence characterization of normal and tumoral esophageal epithelium cells with quantitation of NAD(P)H.

Cellular autofluorescence was characterized in normal human esophageal cells and in malignant esophageal epithelial cells. The study was performed under excitation at 351 nm where the cell fluorescence is mainly due to the reduced pyridine nucleotides (NAD(P)H) with a very small contribution from the oxidized flavins (FMN, FAD) or lipopigments. The autofluorescence emission of squamous cell carcinoma, adenocarcinoma on Barrett's mucosa and normal cells was characterized by microspectrofluorimetry on monolayers and by spectrofluorimetry on cell suspensions. The relative contribution of each fluorophore to the fluorescence emission of the different cell types was evaluated by a curve-fitting analysis. A statistically highly significant difference was observed between the average intensity of the raw spectra of the different cell types. Tumoral cells had a fluorescence intensity approximately twice as high as that of normal cells. The results of the NAD(P)H quantitation analyzed by microspectrofluorimetry on single living cells and spectrofluorimetry on cell suspensions were consistent with those obtained by biochemical cycling assays, showing that the amount of intracellular NAD(P)H is higher in tumoral cells than in normal cells. Bound NAD(P)H concentration was found to be quite stable whatever the cell type while the amount of free NAD(P)H showed a very important increase in tumoral cells.     

Skeletal muscle NAD(P)H two-photon fluorescence microscopy in vivo: topology and optical inner filters

Two-photon excitation fluorescence microscopy (TPEFM) permits the investigation of the topology of intercellular events within living animals. TPEFM was used to monitor the distribution of mitochondrial reduced nicotinamide adenine dinucleotide (NAD(P)H) in murine skeletal muscle in vivo. NAD(P)H fluorescence emission was monitored (~460 nm) using 710–720 nm excitation. High-resolution TPEFM images were collected up to a depth of 150 μm from the surface of the tibialis anterior muscle. The NAD(P)H fluorescence images revealed subcellular structures consistent with subsarcolemmal, perivascular, intersarcomeric, and paranuclear mitochondria. In vivo fiber typing between IIB and IIA/D fibers was possible using the distribution and content of mitochondria from the NAD(P)H fluorescence signal. The intersarcomeric mitochondria concentrated at the Z-line in the IIB fiber types resulting in a periodic pattern with a spacing of one sarcomere (2.34 ± 0.17 μm). The primary inner filter effects were nearly equivalent to water, however, the secondary inner filter effects were highly significant and dynamically affected the observed emission frequency and amplitude of the NAD(P)H fluorescence signal. These data demonstrate the feasibility, and highlight the complexity, of using NAD(P)H TPEFM in skeletal muscle to characterize the topology and metabolic function of mitochondria within the living mouse.     

The effect of atebrine and an acridine analog (BCMA) on the coenzyme fluorescence spectra of cultured melanoma and Ehrlich ascites (EL2) cells

Summary Coenzyme fluorescence spectra of single living cells are due to free pyridine nucleotides (folded configuration), bound pyridine nucleotides (unfolded configuration) and a third component, possibly a mixture of flavins. Such spectra can be used to recognize possible differences in coenzyme composition between cell lines or changes of metabolic pathways due to chemicals acting at levels below or above cytotoxicity, by high resolution spectrofluorometry.A study of spectra recorded from cultured Ehrlich ascites (EL2), and Harding Passey melanom a cells (HPM-67 and HPM-73 line) grown under comparable conditions, shows that free NAD(P)H predominates in HPM-67 and EL2, while this coenzyme is bound in HPM-73. The free/bound ratio may be profoundly modified by chemicals, e.g. in the HPM-73 increase of free and decrease of bound NAD(P)H occurred upon treatment with 10^–6 oligomycin.When atebrine at levels (10^–6 M) below cytotoxicity was added, there was a decrease of the free NAD(P)H spectrum possibly through energy transfer from NAD(P)H to atebrine. Consideration of long range energy transfer i.e., excitation of atebrine by fluorescence of NAD(P)H vs. short range transfer of excitation energy from free NAD(P)H to atebrine, favors the latter mechanism. A transient (reversible) increase in atebrine fluorescence is seen following intracellular microinjection of substrate (e.g. glucose-6-P) leading to an increase in free NAD(P)H. At cytotoxic levels of atebrine (e.g. 2×10^–5 M) an irreversible increase of atebrine fluorescence is seen.The microspectrofluorometric technique appears therefore well suited to study physiological processes at the level of intracellular coenzymes, as well as possible processes of intermolecular energy transfer in the microenvironment.     

NAD(P)H fluorescence imaging of mitochondrial metabolism in contracting Xenopus skeletal muscle fibers: effect of oxygen availability.

The blue autofluorescence (351 nm excitation, 450 nm emission) of single skeletal muscle fibers from Xenopus was characterized to be originating from mitochondrial NAD(P)H on the basis of morphological and functional correlations. This fluorescence signal was used to estimate the oxygen availability to isolated single Xenopus muscle fibers during work level transitions by confocal microscopy. Fibers were stimulated to generate two contractile periods that were only different in the PO2 of the solution perfusing the single fibers (PO2 of 30 or 0-2 Torr; pH = 7.2). During contractions, mean cellular NAD(P)H increased significantly from rest in the low PO2 condition with the core (inner 10%) increasing to a greater extent than the periphery (outer 10%). After the cessation of work, NAD(P)H decreased in a manner consistent with oxygen tensions sufficient to oxidize the surplus NAD(P)H. In contrast, NAD(P)H decreased significantly with work in 30 Torr PO2. However, the rate of NAD(P)H oxidation was slower and significantly increased with the cessation of work in the core of the fiber compared with the peripheral region, consistent with a remaining limitation in oxygen availability. These results suggest that the blue autofluorescence signal in Xenopus skeletal muscle fibers is from mitochondrial NAD(P)H and that the rate of NAD(P)H oxidation within the cell is influenced by extracellular PO2 even at high extracellular PO2 during the contraction cycle. These results also demonstrate that although oxygen availability influences the rate of NAD(P)H oxidation, it does not prevent NAD(P)H from being oxidized through the process of oxidative phosphorylation at the onset of contractions.     

NADH fluorescence and oxygen uptake responses of hybridoma cultures to substrate pulse and step changes

A fiber-optic probe was interfaced to an analytical spectrofluorophotometeru and used to measure NAD(P)H fluorescence of hybridoma cells in a bioreactor. NAD(P)H fluorescence was found to qualitatively represent metabolic state during various induced metabolic transitions. NAD(P)H fluorescence increased immediately following aerobic-anaerobic transitions, and decreased immediately upon transition back to aerobic metabolism. Pulsing of glucose to glucose-depleted cultures caused NAD(P)H fluorescence to first increase immediately after the pulse, and then decrease gradually before reaching a constant level. Pulsing of glutamine to glutamine-depleted cultures resulted in a gradual increase in NAD(P)H fluorescence which lagged a simultaneous increase in oxygen uptake. ATP production and oxygen uptake also varied with metabolic state. The decrease in oxidative phosphorylation following transition from aerobic to anaerobic metabolism was found to be only partially compensated by the concomitant increase in substrate-level phosphorylation, as shown by decreases of 35-52% in calculated total specific ATP production rates. The specific oxygen uptake rate decreased by 6-38% following glucose pulses of between 0.2 and 0.5 g/L, respectively, and by 50% following glutamine depletion. Subsequent pulsing of glutamine after depletion caused oxygen uptake to increase by 50%.     

The UV fading of hydrocarbon fluorescence and its prevention for observations in single living cells.

Excitation intensities used for standard microspectrofluorometric observations of natural cell fluorescence, i.e. NAD(P)H, lead to fading of hydrocarbon (polycyclic aromatic, heterocyclic) fluorescence in EL2 cells incubated with such compounds. The disappearance of hydrocarbon fluorescence under excitation at 366 nm seems to be an exponential function of time. The fading prevents studies on hydrocarbon metabolization in correlation with intracellular microelectrophoretic injection of substrate, e.g. glucose-6-P. A return to 8-10 times less intense excitation conditions used in an earlier prototype microspectrofluorometer, has allowed the observation of sequential changes in the difference spectra (after glucose-6-P minus before) of hydrocarbon-treated cells (e.g. benzo(a)pyrene, dibenzocarbazols). The possible relative contributions of NAD(P)H and hydrocarbon metabolites (or alterations) to such sequential spectra are still under consideration, but the main obstacle to their observation, fading, is removed by less intense excitation.     

Multichannel microspectrofluorometry for topographic and spectral analysis of NAD(P)H fluorescence in single living cells.

Starting from a previously described prototype microspectrofluorometer a more versatile apparatus has been developed with rapid optional operation on a topographic mode for the simultaneous multisite evaluation of NAD(P) reduction-reoxidation transients or on a spectral mode for the analysis of natural and exogenous fluorochromes, in single living cells. On the topographic mode, adetailed kinetic analysis of NAD or NAD P-linked dehydrogenases can be made from 50-100 cell points imultaneously via automatic recording of topographic scans upt to 16 times a second, in correlation with microelectrophoretic intracellular inuection of metabolites (e.g. nearly immediate response to glucose 6-phosphate, 20-25 s delay for 6-phosphogluconate). Rapid shifts from topographic to spectral operation make possible the detection of a change in fluorescence intensity at a specific intracellular site and the immediate verification of its nature (NAD(P)H or exogenous fluorochrome) by spectral observations.     

Spectral characterization of NAD(P)H fluorescence in intact isolated chloroplasts and leaves: Effect of chlorophyll concentration on reabsorption of blue-green fluorescence

In the present communication we report a spectral analysis of the blue-green fluorescence related to changes in NAD(P) redox state in chloroplasts and leaves. To assess the contribution of reabsorption and the inner filter effect, we compared transmission and fluorescence at different chloroplast concentrations, and showed that reabsorption by the photosynthetic pigments (chlorophylls and carotenoids) was at the origin of the two peaks in the emission spectrum in vivo. The absence of potential green-emitting fluorophores in chloroplasts was determined by measuring variable and time-resolved fluorescence at different wavelengths. We defined the conditions which optimize the UV-excited blue-green fluorescence signal dependent on NAD(P)H, and we present an example of monitoring of NAD(P)H fluorescence in intact leaves.     

Effect of adriamycin treatment on the lifetime of pyrene butyric acid in single living cells

We investigated the fluorescence lifetime of pyrene butyric acid (PBA) using various O₂ concentrations in cells. Both in living and freshly fixed cells, PBA lifetime decreased with oxygen concentration. We recorded decay curves in single cells and measured PBA lifetime and NAD(P)H intensity values. Under nitrogen atmosphere, the probe lifetime differences (199 and 209 ns in living and freshly fixed cells, respectively) suggest a supplemental pathway for the deactivation of the probe when the cell functions are not stopped. We propose reactive oxygen species (ROS) to be the additional quenchers that cause this decrease. We further studied the effect of drugs generating ROS the anthracycline doxorubicin (adriamycin). For living cells, PBA lifetime decreased after adriamycin (ADR) treatment (200 and 1000 ng/ml). This supports our hypothesis that under nitrogen atmosphere and for freshly fixed cells, PBA lifetimes increase to an unchanging value due to absence of quenchers.     

NAD(P)H oscillates in pollen tubes and is correlated with tip growth

The location and changes in NAD(P)H have been monitored during oscillatory growth in pollen tubes of lily (Lilium formosanum) using the endogenous fluorescence of the reduced coenzyme (excitation, 360 nm; emission, >400 nm). The strongest signal resides 20 to 40 microm behind the apex where mitochondria (stained with Mitotracker Green) accumulate. Measurements at 3-s intervals reveal that NAD(P)H-dependent fluorescence oscillates during oscillatory growth. Cross-correlation analysis indicates that the peaks follow growth maxima by 7 to 11 s or 77 degrees to 116 degrees, whereas the troughs anticipate growth maxima by 5 to 10 s or 54 degrees to 107 degrees. We have focused on the troughs because they anticipate growth and are as strongly correlated with growth as the peaks. Analysis of the signal in 10-microm increments along the length of the tube indicates that the troughs are most advanced in the extreme apex. However, this signal moves basipetally as a wave, being in phase with growth rate oscillations at 50 to 60 microm from the apex. We suggest that the changes in fluorescence are due to an oscillation between the reduced (peaks) and oxidized (troughs) states of the coenzyme and that an increase in the oxidized state [NAD(P)(+)] may be coupled to the synthesis of ATP. We also show that diphenyleneiodonium, an inhibitor of NAD(P)H dehydrogenases, causes an increase in fluorescence and a decrease in tube growth. Finally, staining with 5-(and-6)-chloromethyl-2',7'-dichlorohydrofluorescein acetate indicates that reactive oxygen species are most abundant in the region where mitochondria accumulate and where NAD(P)H fluorescence is maximal.