1. Introduction

Time-resolved fluorescence microscopy and spectroscopic imaging are powerful tools for imaging biological systems [1–4]. A variety of approaches are possible including the use of time measurement techniques such as time-gated detection, streak cameras, time-correlated photon counting and frequency modulation coupled to conventional, confocal or nonlinear optical microscopes. This leads to benefits not possible with normal fluorescence microscopy based on time-averaged detection or the use of fluorescent labelling.

Picosecond fluorescence has been widely used to study energy and electron transfer in plant photosynthetic proteins in isolated particles of Photosystem II (PS II), Light-Harvesting Complexes (LHCs) and related systems [5–7]. While the distribution of fluorescence emitted from whole plant material has been imaged previously [8], so far this has involved time-averaged, video rate or static measurement of fluorescence and therefore is too slow to observe the picosecond fluorescence components which reflect the fundamental energy and electron transfer processes that occur on timescales of 1 ps – 10 ns. Time-gated picosecond fluorescence imaging should enable determination of the kinetics of these processes and monitoring of spatial variations in these kinetics within the cellular structure of the plant.

Ultraviolet-B (UV-B: 280 – 320 nm) radiation is known to cause serious photoinhibition of plant and algal photosynthesis [9–11], and can be exacerbated by enhanced UV-B due to stratospheric ozone depletion. While there have been several studies of the mechanisms of UV-B photoinhibition to crucial biomolecular components of plants, particularly PS II [12–14], little is known of the UV-induced morphological changes at the cellular level nor the changes in the energy and electron transfer kinetics associated with the collection of photosynthetic biomolecules found in intact plant tissue. In this case, time-gated picosecond fluorescence imaging can be useful because it should enable imaging of the fluorescence yield and quenching as well as fluorescence decay lifetime imaging of molecular components which are photoinhibited by UV-B.

2. Materials and Methods

Picosecond time-gated fluorescence microscopy with 100 ps time resolution was performed using a time-gated microchannel-plate intensified CCD camera system (LaVision Picostar HR) mounted on an optical microscope (Ziess Axiophot) as shown in Fig. 1. Excitation of the sample in the chlorophyll Q-band was achieved at grazing incidence using a pulsed diode laser (PDL 800B) operating at 635 nm with a ~ 60 ps pulsewidth and 80 MHz repetition rate. Fluorescence decays were obtained by sampling at 100 ps delay increments. The emitted fluorescence was collected by a 40× 0.85NA air objective and imaged onto the camera system through a lens and aperture. Imaging quality was enhanced by the small depth of field resulting from the fact that the fluorescence originated from a thin layer near the sample surface where the excitation laser beam is strongly absorbed. A 675 nm bandpass filter allowed selective detection of chlorophyll fluorescence and rejection of stray light. Background signals were negligibly small and the signal-to-background ratio was large due to the intense chlorophyll fluorescence. The instrument response function was symmetric with a risetime of 150–200 ps and stability of ~ 25 ps limited by camera and laser jitter.

Standard preparative procedures were used to prepare PS II membrane fragments [15] and chloroplasts [16] from english spinach. Measurements on these samples were made by first suspending them in a MES buffer (pH = 6.5), dispersing the solution on a glass slide and allowing it to dry in air. The reaction centres in these samples were prepared in the open or closed states by either adding or not adding, respectively, the electron acceptor potassium ferricyanide which allows electron transfer through the PS II reaction center. Apart from washing in milli-Q water, whole spinach leaf tissue was used untreated.

For measurements of the effects of UV-B radiation, samples were irradiated with narrow linewidth tunable UV (285 – 310 nm) from an intracavity frequency-doubled Rhodamine-6G ring dye laser (Coherent 899) pumped by an argon ion laser (Coherent I90-6). Typically, leaf specimens were irradiated for 45 mins with ~ 0.6 mW of (300.0 ± 0.1) nm UV (dose of ~ 52 J cm^-2). This wavelength was chosen since it is near the center of the environmentally-significant UV-B region where various UV-induced damage processes have been studied.

 

Fig. 1. Schematic diagram of picosecond time-gated fluorescence microscope.

Download Full Size | PDF

3. Results and discussion

Before applying these techniques to whole-tissue, it is valuable to first characterize purified components of the tissue material such as PS II and chloroplasts. It is also important to determine the typical chlorophyll fluorescence decay kinetics for these basic photosynthetic units using time-gated fluorescence imaging (TGFI) in order to interpret the fluorescence lifetime data for whole tissue. The fluorescence image for PS II membrane fragments (Fig. 2) indicates that the fragments have consistent and uniform shape and size distributions and that there is some aggregation of these fragments. It should be remembered that TGFI is primarily a high-resolution imaging tool and that the picosecond time resolution (100 ps), which is less than that achievable by time-correlated photon-counting (TCPC), appears to realistically allow decays to be fitted with up to 2 exponentials. The picosecond fluorescence decay curves in Fig. 2 were analyzed and gave the amplitudes and decay times shown in Table 1. The same analysis for each individual PS II particle in the image in Fig. 2 gave very similar amplitude and kinetic parameters to those presented in Table 1 for decays obtained after integrating the intensities over the whole image. These kinetics are consistent with the picosecond kinetics obtained previously by Roelofs et al. [5,6] from TCPC measurements of bulk PS II suspensions. This shows that the fluorescence decay properties of individual PS II particles and bulk PS II suspensions are similar and that these properties are stable, consistent and reproducible for different time measurement techniques and specimen preparations. The conservation of the fluorescence decay kinetics between single-molecule, bulk and in-vivo systems indicates that the energy and electron transfer kinetics in these proteins are largely independent of their aggregation state. For chloroplasts, the resulting fluorescence images resembled those obtained using conventional and confocal light microscopies, and the picosecond decay curves were similar to those obtained for PS II fragments. The decay for chloroplasts is intermediate between that for open and closed PS II centers.

 

Fig. 2. Fluorescence images for the PS II membrane fragments (open centers) and fluorescence decay curves for various chloroplast and PS II preparations. The inset in the image shows a 2x zoom of a region in which aggregation occurs.

Download Full Size | PDF

Table 1. Fluorescence decay amplitudes and time constants for photosynthetic systems^a.

View Table

Picosecond chlorophyll fluorescence imaging experiments were performed for scans of 5 ns on precisely the same region of whole spinach leaf tissue both before and after UV irradiation. The movie (Fig. 3) shows the time evolution of the chlorophyll fluorescence starting from before the excitation laser pulse arrives and extending for 1.6 ns after the fluorescence peak at which point the signal has decayed to 11 % of its maximum value. The images near where the fluorescence peaks show localized and well-defined cellular structure in the case of the control but relatively diffuse and non-specific structures in the UV-irradiated case indicating significant UV-induced rupture of the cellular structure. The total fluorescence signal integrated over the whole image for the fluorescence peak is similar for the control and UV-irradiated cases because, although the fluorescence decay times change, the chlorophyll emission cross-section is not significantly altered by UV irradiation. These UV-induced effects are also reflected in the images for later times. We have analyzed the fluorescence time decay profiles for each pixel in both the control and UV-irradiated experiments as well as for the corresponding data integrated over the whole image (Table 1). The UV-irradiated samples show faster fluorescence decay and a more random spatial distribution of fluorescence compared with untreated samples.

 

Fig. 3. Picosecond time-resolved imaging of chlorophyll fluorescence from untreated (left) and 300 nm UV-irradiated (right) whole spinach leaf tissue (100 ps per s). Animation.mov (1.3 Mb).

Download Full Size | PDF

These effects are only seen on the timescale of 100 ps – 3 ns and cannot be observed using slower imaging rate methods. Neither conventional nor confocal fluorescence imaging of chloroplasts or leaf tissue provide clear evidence for significant morphological change on UV exposure. While PS II fluorescence is sensitive to the dose and wavelength of UV, the resulting UV-induced kinetic changes, which can be seen on the millisecond timescale in PS II preparations, do not appear to be spatially resolved in confocal studies of whole leaf tissue. Overall, picosecond fluorescence imaging is required to observe the changes reported here.

The change in the decay kinetics seen with UV-exposed material is more explicity shown in Fig. 4 and Table 1. The decay times for the untreated leaf are similar to those for both PS II membrane fragments and isolated chloroplasts indicating that the whole-leaf material is photosynthetically intact. In normal tissue, the chlorophyll fluorescence decay time depends strongly on several factors including excitonic coupling of chlorophylls in the various chlorophyll proteins and light-harvesting systems as well as the complex kinetics associated with PS II particularly that involving the primary photoreceptor P680 which in turn is dependent on the electronic configuration of the PS II including the redox states of the water-oxidizing system. The predominant effect of UV on the kinetics is that the decay time of the faster component decreases from 450 ps in the case of the untreated tissue to 350 ps in the case of UV-irradiated tissue, while the amplitude of this component remains essentially unchanged. There is some evidence that UV also causes the longer component to increase slightly in amplitude and possibly decrease slightly in decay time. These kinetic changes indicate that on UV-B treatment, the chlorophyll molecules in the antennae become partially uncoupled from other chlorophyll molecules or from the peptide framework of the chlorophyll-containing antennae proteins. Such UV-degradation processes will result in an increased concentration of “free” chlorophylls which have shorter decay times ~ 200 ps.

In a typical photosynthetic unit there are ~ 250 – 300 antennae chlorophyll molecules per PS II reaction center. Although PS II is clearly damaged by UV, there is no evidence to suggest that this photochemical damage need be manifest as a change in chloroplast or cellular structure. On the other hand, chlorophylls associated with light-harvesting systems account for the majority of chlorophyll in the whole plant tissue and these chlorophylls are strongly excitonically-coupled. Furthermore, it has been suggested [17] that there are significant lifetime variations within and between antennae complexes and LHC II aggregates. Therefore, the UV-induced reduction in the fluorescence decay times reported here is a reflection of photochemical damage to the protein structure of the various chlorophyll proteins leading to a degree of uncoupling of the excitonic interaction between these chlorophylls. This results in an increased number of uncoupled or unbound chlorophyll molecules which have shorter fluorescence decay times than bound chlorophylls. It is likely that such a damage mechanism could lead to significant morphological changes in the tissue structure because uncoupling and even release of chlorophyll molecules from chlorophyll proteins would necessarily involve extensive damage and fragmentation of the peptide frameworks and protein complexes. This uncoupling of chlorophyll molecules will adversely effect the energy reorganization in the light-harvesting antennae systems and subsequent exciton transfer to the PS II reaction center which occurs in the intact photosynthetic unit. This damage to the antennae systems at both the molecular and cellular levels will lead to reduced photosynthetic activity of the organism when exposed to UV-B.

 

Fig. 4. Fluorescence decay curves for untreated and UV-irradiated whole spinach leaf tissue.

Download Full Size | PDF

It is important to appreciate the fundamental differences between UV and visible photoinhibition of spinach tissue. In the case of visible photoinhibition, the PS II damage sites are on the donor and acceptor sides and no discernable decoupling of the antennae system from itself or from the reaction centers is seen [18]. In contrast, UV damages the water-splitting function in PS II and leads to both uncoupling of chlorophyll molecules from the LHCs and uncoupling of these complexes from the reaction centers.

4. Conclusions

In this paper, we have demonstrated that time-gated imaging with picosecond time resolution can provide important microscopic information on whole green plant tissue that cannot be obtained using lower time resolution microscopic techniques. We also show that the effects of UV-B radiation are reflected in the fluorescence lifetime imaging data but only over times of ~ 100 ps – 3 ns following the excitation laser pulse. The observed fluorescence decay kinetics for PS II fragments and other photosynthetic chlorophyll proteins are very similar in the isolated, bulk or in-vivo states indicating that the energy and electron transfer kinetics in these proteins are largely independent of their aggregation state. The changes in the chlorophyll fluorescence decay kinetics indicate UV-induced uncoupling of chlorophyll molecules in the light-harvesting antennae system. This uncoupling then disrupts energy reorganization in the light-harvesting systems and subsequent exciton transfer to the reaction centers. This leads to UV-B photoinhibition of the organism which ultimately effects its viability.