Time-dependent enhancement of fluorescence from Rhodobacter capsulatus SB1003 and its critical dependence on concentration temperature and static magnetic field

Abstract

Continuous exposure of 395 nm light increases the fluorescence emission intensity of photosynthetic purple non-sulphur bacteria, Rhodobacter capsulatus (SB1003). We show that such an increase in fluorescence emission of extracellular pigment complexes (PC) from these photosynthetic bacteria depends on the concentration of the pigment and temperature and can also be modulated by the static magnetic field. The time-dependent enhanced emission disappears either at or below 300 K or below a threshold sample concentration (0.1 mg/ml). The enhanced emission reappears at this condition (T < 278 K) if a static magnetic field (395 mT) is introduced during fluorescence measurement. The time dependence of emission is expressed in terms of a first-order rate constant, k = dF/(Fdt). The sign of k shifts from positive to negative as PC concentration is lowered than a threshold value, implying onset of fluorescence decay (k < 0) rather than amplification (k > 0). At PC concentration higher than a threshold, k becomes negative if the temperature is lowered. But, surprisingly, at low temperature, a static magnetic field reverts the k value to positive. We explain the logical nature of k-switching and photo-dynamics of the aforesaid microbial fluorescence emission by aggregation of protoporphyrin rings present in the PC. While the simultaneous presence of decay in fluorescence and susceptibility to static magnetic field suggests the dominance of triplet states at low temperatures, the process is reversed by SMF-induced removal of spin degeneracy. At higher temperatures, the optical excitability and lack of magnetic response suggest the dominance of singlet states. We propose that the restructuring of the singlet-triplet distribution by intersystem crossing may be the basis of this logical behaviour. In context with microbial function, time-dependent enhancement of fluorescence also implies relay of red photons to the neighbouring microbes not directly exposed to the incident radiation, thus serving as an indirect photosynthetic regulator.

Electronic supplementary material

The online version of this article (10.1007/s10867-020-09545-6) contains supplementary material, which is available to authorized users.

Introduction

Bacterial communication by chemical cue is well-known, and the phenomenon of quorum sensing and biofilm formation is largely based on such inter-microbial chemical signalling-driven unification of a population. For photosynthetic bacteria, the signal has an additional optical component; as in most cases, the photosynthetic pigments are known to absorb as well as emit light. This is in addition to the near-infrared (NIR) absorbance and fluorescence exhibited by R. capsulatus [1]. The present report shows how continuous exposure by UV light leads to increasing emission levels and how this process shows a critical concentration dependence similar to the quorum sensing phenomenon. Thermodynamically, fluorescence can be described largely as a non-thermal energy release process, in which excitation at a higher energy of photons is followed by emission at lower energy, the only thermal part being the vibrational relaxation [2]. For photosynthetic bacteria like R. capsulatus, the presence of a visible domain fluorescence in response to UV excitation would imply an input of higher photonic energy and release a part of its available photosynthetic energy in the form of light with lower energy. As the conversion of photonic energy to proton motive force is a part of metabolic energy utilization, the question remains why the cell loses the efficiency of photosynthetic machinery by releasing such photonic energy in an amplified manner. The answer may lie in a queer thermodynamic aspect; the sacrifice of energy conversion enables the bacterial species to achieve a few logical attributes, as reported in this paper. What makes this question more relevant is that fluorescence from the photosynthetic bacteria R. capsulatus SB1003 becomes stronger upon prolonged UV excitation, the fluorescence intensity following the first-order kinetics dF(t)/dt = − kF, whereas normally one would expect the photobleaching to occur in the absence of any diffusional recovery mechanism [3]. In the present work, the conversion from amplification to bleaching is dictated by some parameters, e.g. fluorophore density, temperature and the presence or absence of the static magnetic field. The onset of fluorescence amplification occurs once a threshold density is reached. Interestingly, what follows is the logical behaviour of the cells, downregulating the photosynthetic activity by showing a higher rate of energy release in the form of fluorescence, when the photon capture centres have a high density (and vice versa). What we report here is naturally occurring biomaterials produced by photosynthetic bacteria can play an implicit role not only in harvesting solar radiation [4] but also in a dynamic though controlled light emission process. This biomaterial produced by photosynthetic bacteria comprises of extracellular pigment complexes or PC in the form of extracellular light-harvesting complexes that absorbs excess high-intensity photon energy and emits low energy photons to maintain the equilibrium of the metabolic process at the cellular level. What we show in this paper, for microbe-produced PC that absorbs incoming UV radiation, mainly consists of porphyrin derivatives as pigments that can be characterized by the Soret band and Q band [5], show time-dependent amplified emission (depending on intrinsic and extrinsic parameters) in the red region. All the porphyrin rings containing systems including chlorophyll and haemoglobin show an intense absorption peak around 400 nm (known as Soret band) and few less intense multiple peaks ranging from ~ 500−600 nm (known as Q band). Time-dependent amplified fluorescence emission upon continuous excitation though studied extensively for artificial systems like polymers or synthetic organic molecules merely reported in natural systems like bacteria and bacteria-produced biomaterials. The photosynthetic activity of R. capsulatus is critically controlled by light intensity, bright light suppressing the photosynthetic activity without altering the transcription from the bchA, bchC, bchE or bchF genes [6]. The mechanism of such a critical light intensity-dependent effect was however unclear. This work reports such criticality in fluorescence enhancement concerning variations in the concentration of pigment complex (PC), temperature and static magnetic field, in the presence of a UV excitation at 395 nm. We then highlight the plausible mechanism of the amplification process and how the amplification can be switched to ‘ON’ and ‘OFF’ mode by the stated concentration and temperature factors. How the molecular crowding mechanism [7] can come into play is also discussed. The overview of the studied system is depicted in Fig. 1. The enhanced emission by the extracellular photopigment complexes at room temperature is discussed in terms of aggregation-induced enhancement of emission (AIEE). The effect of the static magnetic field on fluorescence emission has been previously reported [8]. How radical pair formation [9] and enrichment of excited singlet species may result in an increased quantum yield is discussed here in context with a proposed porphyrin aggregation mechanism. A possible mechanism of the effect of the static magnetic field in modulating the quantum yield, particularly at the lower temperature, is discussed. Lastly, we must mention that the enhancement effect reported for R. capsulatus is also observed for other bacterial species (data are not shown). The existence of this rather universal trait in photosynthetic bacteria suggests that there may be a functional implication of emission enhancement in the photosynthetic microbial community (see “Discussion”).

Fig. 1.

The overview of the system being studied, representing accumulation of pigment complexes (PC) by photosynthetic bacteria and their fluorescence amplification in different conditions

Materials and method

Bacterial growth

The bacteria R. capsulatus SB 1003 were grown photo-heterotrophically in yeast extract supplemented RCV medium (4 g malic acid, 1 g NH₄Cl, 120 mg MgCl₂, 75 mg CaCl₂, 20 mg sodium EDTA (Ethylenediaminetetraacetic acid) salt, 1 mg nicotinic acid, 10 mM KPO₄ buffer, 1000 ml dH₂O supplemented with 0.3 g yeast extract powder, _PH 6.8 ± 0.2). The bacterial culture was incubated at room temperature under continuous illumination for 6–8 days [10] in screw cap bottles. The excreted pigment complexes were obtained from the supernatant after precipitating the cells by centrifugation at 4000 rpm for 6 min followed by further centrifugation to remove cell debris and other particulate matter. The full-grown bacterial culture and extracellular pigment complex (PC) were used for monitoring amplified fluorescence emission. All chemical reagents used in this study were of analytical grade Sigma-Aldrich (USA) or SRL (India) products.

Spectroscopic studies

Spectroscopic measurements of samples were performed by scan method setting in Thermo Scientific™ Evolution 300 UV-Vis spectrophotometer; wavelength range was set from 250 nm to 1000 nm, at 1-nm bandwidth and 600 nm/min scan speed was used. The steady-state fluorescence properties of samples were studied with Photon Technology International (PTI) fluorimetric setup (QuantaMaster™ 40). A 72-W Xenon lamp was used as a light illumination source, and the detection was preceded by bypassing the emitted beam through an optical chopper and emission monochromator. The slit width was set at 4 nm for both excitation and emission. Temperature-dependent fluorescence measurement was guided through a software controlled Peltier. The magnetic field application is illustrated in Fig. S1 of supplementary. A small square-shaped magnet (area 4 mm × 4 mm, field strength 395mT) was placed under the quartz cuvette, in the cuvette holder of above-mentioned fluorimetric set-up to study the effect of the static magnetic field (SMF). The temperature variation experiment in the presence and absence of SMF was performed using this setup. The effect of varying field strength was studied by placing non-magnetic small square pieces (4 mm × 4 mm) of plastic plates between the magnet and the cuvette in the cuvette holder. Time-resolved fluorescence measurements were performed using 370-nm sub-nanosecond-pulsed LED source from HORIBA Scientific single photon counting controller (FluoroHub).

Determination of kinetic parameters: Spectroscopic data analysis

Graphs were plotted considering the initial fluorescence amplification rate along the Y-axis and other variables such as temperature or concentration along the X-axis. Change in fluorescence intensity (count) with per unit time was expressed by a first-order kinetic parameter k (s⁻¹). In the case of amplification, the k value is positive. A negative k value could be obtained if only bleaching occurs. To obtain k, only the initial rate was evaluated using the first 20 s of data, the long-term trend being more complex. The initial rate of changes in F could be expressed by as

$$\frac{\mathit{dF}}{\mathit{dt}}=k\cdot F$$ 1

where k represents the rate of fluorescence intensity (F) change. The MATLAB curve-fitting tool CFTOOL (MathWorks, USA) was used to fit log (F) values for t < 20 to obtain k. The error in k calculations was obtained from triplicate k evaluations from independent experiments.

Fluorescence correlation spectroscopy

In the confocal microscope [11] (MicroTime 200, PicoQuant GmbH), a picosecond diode laser with a base frequency of 40 MHz was used as the excitation source. A water immersion objective (60X, 1.2 NA) was used to focus the excitation light (405 nm) from a pulsed diode laser (PDL 828-S ‘SEPIA II’, PicoQuant) onto the sample kept on a cover-slip. A dichroic mirror (Z405RDC, Chroma) and a filter (HQ430lp) were used to block the excitation light (405 nm) along the fluorescence emission optical path. The fluorescence was then focused through a pinhole (75 μM) onto the MPD detectors (Micro Photon Device). The emission spectra under a confocal microscope were recorded using an electron-multiplying charge-coupled device (EMCCD, Andor Technology) attached to a spectrograph (Andor Technology, Shamrock series). The spectrograph is attached to one of the ports of the PicoQuant Microtime 200 apparatus. In Fluorescence correlation spectroscopy (FCS), the autocorrelation function G(τ) of fluorescence intensity is defined as

$$G\left(\tau \right)=\frac{<\mathit{\delta F}\left(0\right)\mathit{\delta F}\left(\tau \right)>}{<F{>}^2}$$ 2

where F denotes the intensity of fluorescence, so F(0) and F(τ) is the fluorescence intensity at time 0 and at a lag time τ, respectively. Here, < F >² denotes square of mean fluorescence intensity, as a normalization factor for fluctuating fluorescence signal. The autocorrelation traces are fitted to a 3-D diffusion model [12, 13] having triplet contribution

$$G\left(\tau \right)=\left[\frac{1}{N}\right]{\left[1+\frac{\tau }{{\tau }_D}\right]}^{-1}{\left[1+1+\frac{\tau }{S^2{\tau }_D})\right]}^{-\frac{1}{2}}\left[\frac{1-T+\mathit{Tex}{p}^{-\frac{\tau }{{\tau }_{\mathit{tr}}}}}{1-T}\right]$$ 3

In the above equation, τ_D denotes the diffusion time in the confocal volume, τ_tr is the triplet lifetime of a dye molecule, T is the fraction of molecules in the triplet state, τ is the delay time and N represents the average number of molecules in the confocal volume. S (= ω_z/ω_xy) is the structure parameter of the excitation volume, ω_z and ω_xy being the longitudinal and transverse radii, respectively. S is calibrated using Rhodamine 6G (R6G) in water and is found to be 5. The diffusion constant D_t of a dye molecule was determined using the following equation:

$$D_t=\frac{{\omega }^2\mathit{xy}}{4{\tau }_D}$$ 4

where ω_xy is the transverse radius (∼ 320 nm) of the confocal volume (∼ 0.9 fL). The value of ω_xy for our microscope was determined using R6G in water whose D_t value (426 μm²s⁻¹) is known [14]. From D_t, we estimated the hydrodynamic radius r_H of the fluorescent molecule applying Stokes-Einstein equation given by:

$$D_t=\frac{k_{B}T}{6\mathit{\pi \eta }{r}_H}$$ 5

where η indicate the viscosity of the system, k_B is Boltzmann’s constant, r_H is the hydrodynamic radius of the diffusing species and T is the temperature.

Particle size measurement for extra cellular pigments

Four hundred microlitre of the sample was used for DLS measurement (Zetasizer Nano Series; Malvern Instruments, Malvern, UK) at 278 K, 288 K, 298 K, 303 K and 313 K, respectively. The effect of a weak magnetic field on the sample at 278 K was studied by placing the magnet just below the cuvette during the measurement.

Live cell imaging using confocal microscopy

Microscopic fluorescence study was carried out with small biofilm [15] and cellular clusters (collected from bacterial culture tube) in an inverted confocal microscope (Olympus, FV 1000). A 405-nm laser was used as an excitation source, and fluorescence emission was observed between 570 and 670 nm. Time-lapse imaging of the samples was acquired using the continuous laser at a 1-min time interval for a period of 10 min. Additionally, time trace emission from a single bacterium and cell clusters were measured using the confocal microscope (MicroTime 200, PicoQuant GmbH).

Results

The study reports the biophysical properties of amplified fluorescence emission from the pigment complexes or PC from the photosynthetic bacteria R. capsulatus SB1003. These bacteria are found to accumulate extracellular pigment complexes (PC) in the culture medium. In the extracellular medium, in the optimum temperature range when PC is present in an adequate amount, it undergoes aggregation upon physical changes induced by ultraviolet (UV) or blue light (360–420 nm, λ_max 395 nm) portion of the spectrum. Figure 1 depicts an overview of the studying system and conditions. The aggregation-induced optical changes have been studied spectroscopically in different conditions, and results are obtained from the absorption and fluorescence spectra. Further information about aggregation dynamics and excited state photochemistry are obtained from dynamic light scattering (DLS) and fluorescence correlation spectroscopy (FCS), respectively. The cellular level changes for this aggregation-induced fluorescence enhancement was observed by confocal microscopy.

Spectroscopic analysis

Absorbance spectra of the extracellular pigments in the medium (see Fig. 2) shows a Soret band at 395 nm and Q band at 500 nm, 535 nm and 560 nm. The pattern of absorbance spectra that is shown in inset A denotes the presence of porphyrin H and J band, respectively [16], near 400 nm and 535 nm (approximate intensity ratio 7:1). The absorbance value of the sample shows a sustained increase with time of UV exposure (main figure). The inset b represents a typical fluorescence spectrum at a given time for excitation at 395 nm. The inset c shows the log plot of absorbance with time, the calculated rate constant k_ex representing the fractional rate of increase in absorbance (i.e. the excitation). k_ex <  < k_emission = k (fractional rate of fluorescence change). Expressed in s⁻¹ and k is tenfold higher than k_ex.

Fig. 2.

a Excitation induced changes in absorbance spectra of the PC. The absorbance spectra of the extracellular pigment complex were measured at a particular time interval during UV exposure. Spectra shows gradual increase at 395 nm during prolonged excitation up to 900 s. b The figure shows the kinetics of absorbance vs time, implying time-dependent increase in excited state population. c Figure shows an example of change in fluorescence spectrum over excitation light exposure for a period of 200 s

The bacterial cells and the excreted pigment complex show fluorescence emission at 613 nm when excited at 395 nm. Fluorescence time scan reveals a systematic increase in fluorescence emission intensity for both whole cells and the pigment complex excreted by cells. The rate of amplification in the two cases are however different. Organic extraction of the sample in acetone-methanol exhibited two emission peaks at 594 nm and 635 nm. Notably, the fluorescence excitation maxima are red shifted to 411 nm and 403 nm for the respective emissions. The deconvolution of original emission peak through organic solvent perturbation (see Fig. S2 in supplementary information) into two peaks indicates the presence of two porphyrin derivatives that is coproporphyrin and Mg-protoporphyrin monoethyl ester fluorescence [6]. The amplification in the reconstituted suspension can be realized only by using a mixed solvent having both polar and nonpolar components (data not shown).

Concentration-dependent enhanced fluorescence emission

The extracellular pigment complex obtained from bacterial culture shows a transition in emitted fluorescence intensity over a threshold concentration. The upper panel of Fig. 2 reveals concentration-dependent enhanced emission during a fluorescence time scan. Figure 2 indicates that not only the fluorescence intensity but also the amplification has a systematic dependence on the concentration; the rate of emission (k) changes from positive to negative with decreasing fluorophore concentration. One can find fluorescence emission intensity either increases or decreases with time depending on the concentration of the fluorophore. Further insight of such concentration dependence is provided by a sigmoid signature in Fig. 3, where the initial (first 20 s) fluorescence emission rate has been studied.

Fig. 3.

a Kinetics of concentration-dependent fluorescence amplification are shown. The ordinate represents the kinetic parameter (k) defined by the equation dF/dt = kF. The sigmoidal profile illustrates the change of k upon excitation at 395 nm with increasing PC concentration. b Figure shows the linear fit of ln(F) against time for each concentration, the slope representing k in reciprocal time unit.. The concentration-dependent reversal of the temporal profile may be noted

Temperature and static magnetic field dependence

The PC shows an increase in the rate of amplified emission with increasing temperature, at a fixed sample density (see Fig. 4). The rate of enhancement is found negative at low temperature (5 °C), which reverts to positive with increasing temperature, whereas a magnetic field-driven amplification switch is present that operates at low temperature. Hence, Fig. 4 shows a clear temperature-driven lowering of the amplification rate at 278 K whereas a magnetic field-driven elevation of the amplification rate at the same temperature illustrate the effect of static magnetic field (SMF) on fluorescence emission of the system, (see Fig. 5). The system shows magnetic field strength dependent amplification at low temperature, the rate of amplification at lower temperatures is found maximum for the highest magnetic field exposure (see Fig. S3 in supplementary information).

Fig. 4.

The rate of fluorescence enhancement is plotted with increasing temperature ranging from 5 to 25 °C or 278 to 303 K. A kinetic switch for the temperature induced fluorescence amplification. The figure shows fluorescence intensity increases with temperature and decreases with decreasing temperature

Fig. 5.

Magnetic field-induced fluorescence amplification rate of emission k s⁻¹ is plotted in presence and absence of 395 mT field. a Figure shows amplification response at 278 K, (t = 2.558, df = 8,p value = 0.03375). b Figure shows amplification response at 298 K, (t = − 2.5781, df = 8,p value = 0.03271)

Fluorescence lifetime

Alteration of sample concentration shows that the lifetime value decreases with decreasing order of concentration, (see Table 1). The evaluation of lifetime data for the excreted fluorophore revealed that two components are on the order of 2 ns and 15 ns. This indicates the presence of two species in excited states. So, the notable point is that during fluorescence emission, only one species shows varying lifetime and another is not.

Table 1.

Life time variation with respect to fluorophore concentration

Concentration	Life time (ns)
	τ1	τ2
0.039	1.69 ± 0.0766*10−3	13.81 ± 0.1754*10−3
0.113	1.88 ± 0.0862*10−3	14.11 ± 0.1465*10−3
0.590	2.18 ± 0.2653*10−3	14.25 ± 0.1451*10−3
1.206	2.36 ± 0.3426*10−3	14.29 ± 0.1575*10−3
2.984	2.6493 ± 0.4107*10−3	14.32 ± 0.1537*10−3

Fluorescence correlation spectroscopy

Fluorescence correlation spectroscopy (FCS) with a cell-free fluorophore (PC) has been performed under such a condition where there is minor bleaching and also an amplification phase (see Fig. 6). In Fig. 6, the upper panel shows fluorescence intensity time trace of the bacterial fluorophore, and the lower panels show FCS decay patterns in two successive periods of time, one in which the bleaching dominates and the other in which there is a small gain over the steady state. Notably, the FCS pattern and particularly the G (0) value remains identical in different time courses. The parameter values described in Table 2 which additionally compares the FCS parameters for the sample and a control dye clearly indicates that the FCS parameters (explained in Eqs. 1, 2, 3 and,4) practically remain unchanged in the two temporal phases.

Fig. 6.

The upper panel shows time trace emission of the PC at moderate concentration (0.1 mg/ml) where amplification occurs followed by initial bleaching. The phase of bleaching and fluorescence amplification signal is marked by the direction of the red arrows. The lower panel shows FCS decay pattern of the bacterial fluorophore at two different time course. Initial phase showing dominance of bleaching and the later one for enhancement

Table 2.

The table shows different FCS (fluorescence correlation spectroscopy) parameter values (T, N, τ_tr, τ_D, D_t) as defined in Eq. (2)

System		T	N	τtr	τD	Dt
Samples	0–4 min	0.05	300	0.015	0.17	150
	5–9 min	0.38	315	0.015	0.18	142
Control (C480)		0.05	210	0.01	0.045	570

The table compares parameter values for excreted bacterial fluorophores for FCS decay with the control dye coumarin 480 (C480). The existence of two species in PC, one with concentration independent lifetime, and the other with concentration dependent life time, should be noted

Comparison with the control dye (coumarin 480) is also depicted in Table 2. The intensity pattern showing an initial photo-bleaching followed by a recovery. This indicates that excitation of the fluorophore undergoes a photochemical transformation to form higher fluorescence quantum yield (brightness). This partially compensates for the photobleaching. The G(0) value and hence the number of molecules in the focal volume (N) of the FCS traces (Fig. 4) recorded at two different time courses (0–4 min and 5–9 min) are thus found to be identical in spite of bleaching or amplification. However further insight with samples containing varying PC concentrations shows a differential lifetime value of the exciting triplets. The triplet lifetime value (standard deviation of 10%) is decreased with an increasing degree of dilution, whereas the extracellular pigments and the membrane fraction isolated from bacterial cells show a more or less similar number of molecules in the excited triplet state (see Table 3).

Table 3.

Characterization of extracellular pigments and membrane fraction through FCS

Fluorophore concentration (nM)	T	N	τtr (ms)	τD (ms)	Dt (μm2s-1)
1184 (free fluorophore)	0.54	642	0.015	0.182	140
895 (free fluorophore)	0.49	485	0.020	0.157	162
647 (free fluorophore)	0.51	351	0.019	0.172	148
553 (free fluorophore)	0.51	300	0.016	0.151	168
212 (membrane bound fluorophore	0.58	115	0.017	0.159	160

Dynamic light scattering (DLS) measurements

At low temperature (278 K), the pigment assembly shows an abundance of small aggregates which is shifted to larger population size when exposed to a static magnetic field (see Fig. 7). DLS measurement of the sample shows temperature-induced aggregation (see Fig. S4 in supplementary information). With increasing temperature, there is a tendency to form a larger population of large size with the heterogeneous distribution.

Fig. 7.

Dynamical light scattering measurement showing intensity histogram distribution of hydrodynamic size in presence and absence of static magnetic field, respectively, at 278 K. The sample tends to form larger aggregates in presence of SMF. Here ‘+ SMF’ denotes the size distributions of the sample in presence of magnetic field and ‘– SMF’ indicates the size distribution in absence of magnetic field or control system

Enhanced fluorescence emission from live bacterial cells

The microscopic data described in Fig. 8 reveals that whole bacterial cells showed enhanced fluorescence emission over a longer time of 10 min. The panels numbered 0, 1,…9 corresponds to minutes of excitation light exposure to the sample. When a FRAP (fluorescence recovery after photobleaching) experiment was carried out with bacterial cells, the fluorescence emission intensity gradually increased instead of photobleaching. For further study, the intensity of the incident laser (405 nm) beam was increased from 25 to 100%, and the sample was exposed for 1 min. The result is shown in the lower panel of Fig. 8. The indices a and b refer to pre-excitation and post-excitation states of the sample. This time there was a dual effect, the central part showing a bleaching (see Fig. 8c) and the peripheral region showing an intense fluorescence. The data is also a scaled-up version of what we obtained in Fig. 8a which implies that as we increase the concentration, the amplification gain mode dominates.

Fig. 8.

Time-based fluorescence amplification by live bacterial cells during UV exposure under confocal microscope (scale, 10 μm). a Gradual increase in fluorescence emission of bacterial cell cluster is observed. Cells were exposed to 405 nm laser (25% intensity) up to 9′, at 1′ interval starting from 0′; b The relative increase in image intensity (with respect to 0 min, the error being evaluated using four independent ROIs); c Image of a cell cluster exposed for 60 s to 405 nm laser (100% intensity), the initial image (left) after 60 s of laser exposure assumes the image form shown at the right (of the panel c), whereas the central zone of the right image is bleached due to intense laser exposure a more illuminated zone appears at some distance

Discussion

We report time-dependent amplification of fluorescence emission (at 613 nm) intensity upon UV excitation by the excreted photosynthetic complex from R. capsulatus SB1003.In contrast, the NIR fluorescence (NIR emission at approximately 910 nm) exhibited by the bacterium does not show a gain of fluorescence with time or significant sensitivity with respect to variations in temperature (see supplementary Fig. S5). The kinetics of increase in fluorescence emission at 613 nm is a function of concentration, temperature and the applied static magnetic field. We suggest the following mechanisms behind such photoemission process: (a) aggregation-induced enhanced emission by the extracellular pigment complexes at room temperature above a critical concentration, (b) inhibition of enhanced emission at low temperature and (c) the reappearance of enhanced emission in the presence of the static magnetic field in a concentration-dependent manner.

It seems that at normal temperature range, the overall changes in fluorescence is the result of a dynamic balance between the enhancement and bleaching processes that are a reflection of kinetic equilibrium between aggregates and monomeric forms of the porphyrin derivatives; the sign of k is determined by the dominating component. To understand the mechanism and implication of the amplification process, we can have the following perspective views.

Porphyrin aggregation-induced enhancement of fluorescence and effect of the static magnetic field

When R. capsulatus cells are grown under an anaerobic or microaerophilic condition in the presence of continuous light, they do accumulate tetrapyrroles in the form of protoporphyrin and Mg-protoporphyrin derivatives [6]. Porphyrin containing tetrapyrroles easily interact non-covalently because of the presence of pi bonds, commonly known as pi stacking and lead to self-assembly [17, 18] also known as H-aggregates. At low sample concentration, the numbers of monomeric porphyrin dominate, and lesser numbers of aggregates are present. The reduction of fluorescence intensity corresponds to fluorescence bleaching of the monomeric species. But, higher concentration reverses the situation and fluorescence increases (see Fig. 3). The amplification is driven by the formation of aggregates which may have a higher extinction coefficient as well as higher quantum yield explaining the dual enhancement of both absorbance and fluorescence.

The question may now be raised of why we see lesser amplification at a lower temperature. The answer may be found in the same model; as contrary to the common self-assembly processes favoured at low temperature [19], porphyrins undergo aggregate formation with increasing temperature [20]. Porphyrins with negatively charged peripheral groups easily aggregate in aqueous solutions [21] and could lead to enhanced fluorescence emission [22]. At room temperature, porphyrins could coexist in monomeric and aggregated forms like J and H type of aggregates [16, 23] or their mixture [24]. Magnetic field-induced changes in the amplification pattern imply the role of stacked pi ring assemblies which are known to respond to the static magnetic field [25]. Porphyrin J-type aggregates are known to respond to the magnetic field [26], and when an SMF is applied, they change their orientation and are aligned in a more ordered structure. Apart from the aggregation dynamics, we should consider the excited state photochemistry of porphyrin aggregates. Both types of porphyrin aggregates should form long-lived excited triplet states [27]. Now singlet-triplet interconversion can lead to a higher lifetime of the excited states [28], so any kinetic process that leads to enrichment of the excited singlet state can favour the amplification process. As we know triplet formation is facilitated at lower temperatures through radical pair [29], it is normal to have a dominance of photobleaching at that temperature. However, a magnetic field in such a condition would destabilize the triplet state through triplet-triplet fusion (also known as triplet annihilation). Alternatively, Zeeman splitting may occur in which the spin degeneracy is removed, and singlet states can reappear resulting in photo amplification. Lastly, we illustrate the proposed model by the scheme depicted in Fig. 9. Note that the species that have non-vanishing spin are magnetically more sensitive.

Fig. 9.

Schematic representation of the proposed mechanism for fluorescence enhancement through aggregation. A simple model illustrating the dynamic balance between the fluorescence amplification and photobleaching by the PC through their monomeric and aggregated forms and their conversion by thermal and magnetic switch

In addition, fluorescence emission can decrease because of the singlet-triplet radiationless transition process (intersystem crossing). The singlet-triplet energy gap is inversely proportional to the energy gap between the two states. For an extended system, singlet and triplet states become almost equal in energy, and facile transitions occur between singlet and all the three sublevels of the triplet state. Application of small magnetic field lifts the degeneracy of the triplet states, and only one remains degenerate with a singlet. Thus the rate of S-T transition becomes almost one third. This enhances the fluorescence emission by singlet states [8].

A thermodynamic perspective of the energy release

The thermodynamic implications may be expressed by a simple expression for free energy changes containing both enthalpic and entropic parts. We may consider a simple three-state description of fluorescence (1 → 2 → 3). Such emission is observed in the case of coral heating by green fluorescent protein or GFP [2]. The opposite scenario, namely, cooling, is known by anti-Stokes emission [30]. The overall changes in the enthalpy in the emission process can be expressed by

$$\mathit{\Delta H}=m\cdot h{v}_{\mathit{em}}-n\cdot h{v}_{\mathit{ex}}$$ 6

where n and m are the numbers of photons involved in the excitation and emission processes and v_em and v_em are frequencies of the excitation and emission maxima, satisfying v_em < v_em. If m and n are of the same order, we expect a negative enthalpy change leading to exothermic situation, i.e. heat generation. Thus if m = n, then

$$\mathit{\Delta H}<0$$ 7

This corresponds to normal Stokes emission reported in most cases of fluorescence. One may wonder how the photosynthetic process avoids a thermal catastrophe has given the exothermic nature of the process (expressed by Eq. (7)). In a dynamic situation, e.g. AIEE, m and n may change with time, the rates being of the order of k_ex and k (see Figs. 2 and 3). As k_em >  > k, it is expected that with time, m(t) assumes higher value than n(t), and thus enthalpy may follow

$$\mathit{\Delta H}>=0$$ 8

implying thermal equilibrium or endothermic condition. Thus amplifying emission may actually bring thermal equilibrium which may be helpful for the microbial community under continuous exposure to incident radiation.

The presence of a near-infrared (NIR) absorbance peak near 800 and 900 nm for the intact microbes opens up an alternative route for thermal equilibrium between the microbial communities a fraction of which may be illuminated by incident UV light. NIR absorption of the proximate microbes not directly exposed to UV of the emission generated from one that is excited by UV. In Fig. 9, the lower right panel shows the higher illumination of the neighbouring microbes upon excitation of the central microbial layer. The possibility of a two-photon system in a microbial community needs further investigation.

Conclusion

Similar to the population inversion that leads to laser-like coherent emission, the most ancient life forms exploit a novel trick, namely, modulation of porphyrin ring aggregation, thereby achieving an efficient spin-state dependent stimulation of elevated levels of optical emission. The enhancement may occur at the extracellular level as well as in the intracellular regions. The distribution of singlet and triplets in such aggregates determines the level of optical excitability and magnetic responsiveness, respectively, and eventually, the level of aggregate formation depends on time. The stimulation is triggered by UV excitation and shift in two local environmental variables, namely, porphyrin concentration and temperature that regulates the stimulation level. The spin dependence is evident from the rescaled distribution of the singlet and triplet states when the temperature is lowered. Thus at low temperature, the amplification disappears, but in the presence of magnetic field, there is a restoration of the amplifying state. Such switching in the photosynthetic prokaryotes seems to suggest a logical attribute of the optical emission process in response to UV excitation. The aggregation of tetrapyrroles thus may serve as the basis of such Opto-chemical logic, UV excitation being the ground level inducer of such switchable optical transition.

Funding information

We thank MEITy, Govt. of India (Project Open Hardware Based Digital Communicable Biosensing Platform, 13(10)/2016-CC & BT and Department of Biotechnology, Ministry of Science and Technology (BT/PR3957/NNT/28/659/2013). We thank Ms. Boni Haldar (DBT IPLS) for technical assistance in confocal experiments. We also thank Prof. M. Bhattacharya for providing the lifetime measurement facility. Our sincere thanks to Prof. Patrick C. Hallenbeck for providing the bacterial strain.