DIGITONIN FRAGMENTATION OF SPINACH CHLOROPLASTS a ...

2 downloads 0 Views 1008KB Size Report
625-watt quartz-iodine lamp anda Bausch and Lomb 500-mm diffraction ... was analyzed at right angles to the exciting beam by a second Bausch and Lomb.
FLUORESCENCE PROPERTIES OF PARTICLES OBTAINED BY DIGITONIN FRAGMENTATION OF SPINACH CHLOROPLASTS BY N. K. BOARDMAN, S. W. THORNE, AND JAN AI. ANDERSON DIVISION OF PLANT INDUSTRY, C.S.I.R.O., CANBERRA, AUSTRALIA

Communicated by R. N. Robertson, May 31, 1966

Previous studies1-3 have shown that spinach chloroplasts may be fragmented by incubation with digitonin into separable fractions with different proportions of chlorophyll a and chlorophyll b and with different ratios of the trace metals Cu, Fe, and Mn. The larger particles, sedimentable at 1,000 g and 10,000 g, had a lower ratio of chlorophyll a/chlorophyll b than intact chloroplasts, and a higher Mn content. They were active in the Hill reaction, but the rates of reduction of NADP+ were lower than the rates of reduction of either ferricyanide or 2,3',6-trichlorophenol-indophenol. The smaller particles, which either sedimented at 50,000144,000 g or remained in the 144,000 g supernatant, had substantially higher ratios of chlorophyll a/chlorophyll b than chloroplasts, and a lower content of Mn. They were inactive in the Hill reaction but they photoreduced NADP+ if provided with a suitable electron donor, and an enzyme preparation containing ferredoxin and NADP reductase. It was postulated that the digitonin treatment resulted in a physical separation of particles representative of pigment system 1 from particles which remained attached to the grana lamellae and were enriched in pigment system 2. Currently, it is thought that the two pigment systems activate different parts of the photosynthetic electron transport chain.4 The photochemical reaction which leads to NADP+ reduction is activated by the "long-wavelength system" or pigment system 1, while oxygen evolution is associated with pigment system 2 which absorbs a larger fraction of light of shorter wavelengths. In the studies reported in the present communication, the fluorescence properties of the small and the large particles, examined at both 20'C and 770K, were compared with the fluorescence properties of intact chloroplasts. The results are consistent with the hypothesis that digitonin treatment of chloroplasts gives a partial fractionation of the photochemical systems. Methods.-Preparation of fractions: Spinach chloroplasts were prepared as described previously' and fragmented by incubation with 0.5% digitonin for 30 min at 0WC. The incubation mixture was then diluted with 9 vol of 0.05 M phosphate buffer pH 7.2, and the chlorophyll-containing particles were separated by differential centrifugation at the following speeds: 1,000 g for 10 min, 10,000 g for 30 min, 50,000 g for 30 min, and 144,000 g for 60 min. The pellet from the centrifugation at 10,000 g was resuspended in 0.05 M phosphate buffer pH 7.2, re-centrifuged at 10,000 g for 30 min, and again resuspended in phosphate buffer. This fraction will be referred to as the 10,000 g fraction. The pellet from the centrifugation at 144,000 g was resuspended in phosphate buffer to give the 144,000 g fraction. For fluorescence measurements, the chloroplasts or the fractions were diluted to an absorbancy of 0.1 at 436 m,, to minimize reabsorption of the emitted light. Fluorescence measurements: Fluorescence emission and excitation spectra were recorded on a fluorescence spectrometer incorporating automatic correction for photomultiplier and monochromator responses, and variation in output of the light source. Excitation was by means of a 625-watt quartz-iodine lamp and a Bausch and Lomb 500-mm diffraction grating monochromator, with a dispersion of 33 A/mm. The exciting light was monitored by a photomultiplier type 6255. Fluorescence was analyzed at right angles to the exciting beam by a second Bausch and Lomb 500-mm monochromator and a photomultiplier type 9558. The fluorescence signal was then 586

VOL. 56, 1966

BIOCHEMISTRY: BOARDMAN ET AL.

587

automatically divided by the excitation signal to provide a true record of the fluorescence signal. This referenced output was applied to a potentiometric recorder synchronized either to the excitation monochromator wavelength drive at a fixed fluorescence wavelength to produce an excitation spectrum, or to the fluorescence monochromator wavelength drive to produce an emission spectrum at a fixed excitation wavelength. Excitation from 600 to 720 my was outside the calibrated correction range, and excitation spectra in this range were partly uncorrected and hence of reduced accuracy. Fluorescence measurements at low temperature were carried out in 60% glycerol in a cylindrical glass cell, cooled to the temperature of liquid nitrogen by a cold finger. Fluorescence quantum efficiency (0) was determined by an extension of the relative method.' Fluorescein in 0.1 N NaOH at 10-6 M was taken as a convenient standard, with an absolute value of quantum efficiency of 0.92 at 200C.6 Absorption spectra at 770K were recorded in 60% glycerol with a Cary model 14R spectrophotometer, as described elsewhere.7

Results.-Fluorescence yields of chloroplasts and of chloroplast fractions at 200C: The solid line in Figure 1 shows the fluorescence emission spectrum at 20'C of intact chloroplasts excited at 436 mMA. Addition of digitonin to a final concentration of 0.5 per cent increased the fluorescence yield by a factor of 1.7 without changing the shape of the fluorescence spectrum. The fluorescence emission spectra of the 10,000 g and 144,000 g fractions are shown in Figures 2 and 3. At 20'C, the peak positions of the two fractions did not differ significantly, either from one another or from that of the chloroplasts; all maxima were at 683 m1A. There was a striking difference, however, in the quantum yields of fluorescence (Table 1). The 144,000 g fraction gave an apparent quantum yield of only 0.3 per cent, which was one half the quantum yield of chloroplasts and less than one third the yield of digitonintreated chloroplasts. In contrast, the quantum yield of the 10,000 g fraction was 1.6 per cent, fivefold greater than that of the 144,000 g fraction and 50 per cent more than the yield of the digitonin-treated chloroplasts. It is apparent, therefore, that digitonin frag8 mentation of spinach chloroplasts results in the physical separation of "weakly fluorescent" particles from particles which are more fluorescent than chloroplasts. 6 Effect of additions on fluorescence yields at 200C: z Addition of 3- (p- chlorophenyl) - 1,1 -dimethylurea 77°K (CMU) to chloroplasts markedly increased their 4 A fluorescence yield (Table 2), which suggests that the i fluorescence of the control chloroplasts was quenched by electron flow, even in the absence of an added Hill 3 2 ; \ / oxidant. The fluorescence of chloroplasts was higher x3\J \ in 5 X 10-4 M CMU than in 1 X 10-4 M CMU. / 201C 0 660 70 0 7 Ferricyanide slightly reduced the fluorescence yield of 780 AmX chloroplasts, but the further addition of CMU inhibited electron flow and restored the fluorescence to FIG. 1-Fluorescence emisthat observed with chloroplasts and CMU alone. In sion spectra of chloroplasts at 20C0 (solid line, X3) and agreement with the observation of Duysens and Buffer, 0.05 M phosphate770K. pH Sweers,8 dithionite increased the fluorescence yield of 7.2. Excitation wavelength, 436 2 X 1013 mjA. Light intensity, chloroplasts more than threefold. ConcentraThe second and third columns in Table 2 show a quanta/cm2/sec. tioll, 0.1 absorbancy units at 436 my. Measurements at 77 K comparison between the fluorescence yields of chloro- were plasts, and of chloroplasts incubated with digitonin erol. carried out in 60% glycU

>

PROC. N. A. S.

BIOCHEMISTRY: BOARDMAN ET AL.

588

6

6

FIG.

2

?_~~~~I

/ Fl-77Ksc I~ ~ ~ ~ ~ ~

II

0

1' 660700

racioa 20C 660' 700 0

0

78

?40

70

560

80

74

780K

70

780

70

I 20aC

i l 740

I 70

{>0

70

FIG. 2. - Fluorescence emission spectra of the 10,000 g fraction at 20'C (solid line, X3) and 77'K. Conditions as for Fig. 1.

c~~~~~~~~00 700 660 780 740

FIG. 3.-Fluorescence emission spectra of the 144,000 g fraction at 2000 (solid line, X3) and 770K. Conditions as for Fig. 1.

660

700

FIG. 4. - Fluorescence emission spectra of the 144,000 g supernatant at 200C (solid line) and 770K. Conditions as for Fig. 1.

for 30 min at 00C. As mentioned earlier, incubation with digitonin increased the fluorescence of chloroplasts, but under strongly reducing conditions the fluorescence yields were comparable. Ferricyanide quenched the fluorescence of digitonintreated chloroplasts and chloroplasts by different amounts, with the result that the yields under oxidizing conditions did not differ significantly. The diminished effect of CMU in the presence of ferricyanide is puzzling. A comparison between columns 3 and 4 in Table 2 shows that the various additions had similar effects on the fluorescence yields of the 10,000 g fraction and the digitonin-treated chloroplasts. The fluorescence yield of the 10,000 g fraction was increased by dithionite and by CMU, and decreased by ferricyanide. However, the fluorescence yields of the 10,000 g fraction were about 50 per cent higher than the corresponding values for the digitonin-treated chloroplasts, a result which indicates that the 10,000 g fraction is intrinsically more fluorescent than are the chloroplasts. In contrast, the 144,000 g fraction exhibited a much lower fluorescence yield and, furthermore, the yield was not influTABLE 1 QUANTUM YIELDS OF FLUORESCENCE AT 2000

Chloroplasts

Chloroplasts in 0.5% digitonin

Fraction at 10,000 g

enced by dithionite or CMU.

Fern-

Quantum yield

cyanide appeared to cause a small but reproducible decrease in fluorescence

0.007

yield. Previously,3 we had found that

of fluorescence, 0.011

0.016

the 144,000 g fraction is capable of re-

ducing NADP+ at high rates if provided 0. 003 Fraction at 144,000 g with the electron donor couple, sodium 0.027 Supernatant at 144,000 g 0.19 Chloroplasts in 4% Triton X-100 ascorbate + dichiorophenolindophenol, 0.235 Chlorophyll a in ethanol and both ferredoxin and NADP reducin a chlorophyll for except for 1, Fig. Conditions as ethanol, which was excited at 429 mp. Chlorophyll a was purified by repeated chromatography on sucrose columns. tase. However, Table 2 shows that

VOL. 56, 1966

BIOCHEMISTRY: BOARDMAN ET AL.

589

TABLE 2 EFFECT OF ADDITIONS ON FLUORESCENCE YIELDS AT 20'C Additions

Chloroplasts

Chloroplasts in 0.5% digitonin

Fraction at 10,000 g

Fraction at 144,000 g

12 42 28 37

20 45 30 37

29 62 36

6 6 7

13

14 16 -

26 -

5

27 37

None Dithionite CMU (1 X 10-4 M) CMU (5 X 10-4 M)

Ferricyanide (5 X 10-5 M) Ferricyanide + CMU (1 X 10-4 M) Ferricyanide + CMU (5 X 10-4 M) Ascorbate + DCIP* Ascorbate + DCIP + enzymes*t + NADP+

9

10

6 6

Figures represent amplitudes of fluorescence emission at 683 mp at the same electrical gain. Conditions as for Fig. 1. * Reaction mixture contained in 3 ml (in jsmoles): Tris-HCI buffer (pH 8.0), 40; NaCl, 70; MgCh2, 10; sodium ascorbate, 6.0; DCIP, 0.02. t Contained 0.6 pmole NADP+, and ferredoxin and NADP+ reductase.

TABLE 3 QUANTUM YIELDS OF FLUORESCENCE AT 770K Material

4 (683 + 693 mp&)

Chloroplasts Fraction at 10,000 g Fraction at 144,000 g Supernatant at 144,00 g Chlorophyll a in ethanol

0.049 0.055 0.004 0.027 0.19 (4 678)

4 735 mp

0.15

0.09 0.126 0.051 0. 072

(q 738)

4 Total

, 735 4, Total

0.20 0.15 0.13 0.078 0.26

0.75 0.60 0.97 0.65 0.27

Conditions as for Fig. 1. Excitation wavelength for chlorophyll a, 449 my.

the addition of these components to the 144,000 g fraction had no effect on the fluorescence yield. Fluorescence emission at 770K: Lowering the temperature of chloroplasts to 770K drastically altered the shape of the fluorescence emission spectrum (Fig. 1). The strongest emission was at 735 mj. The 683-mM band also showed intensification, and a new band was seen at 693 ma. These results are in agreement with those of previous investigators.9' 10 The 10,000 g fraction gave a low-temperature spectrum which resembled that obtained with chloroplasts, but the contributions of the 683-mu and 735-mA bands to the total fluorescence were smaller (Fig. 2 and Table 3). In contrast to the chloroplasts and the 10,000 g fraction, the 144,000 g fraction showed a very different behavior on lowering the temperature (Fig. 3). The fluorescence yield at 683 m1s was increased only slightly, as compared with the yield at 200C, and there was no significant splitting of the band. The intensification at 735 m,4 was very great indeed, and at 770K this band accounted for 97 per cent of the total fluorescence emission (Table 3). The corresponding figures for chloroplasts and the 10,000 9 fractions were 75 per cent and 60 per cent, respectively. Lowering the concentration of the 144,000 g fraction from 0.1 absorbancy units at 436 m~uto 0.033 units did not change the relative amounts of fluorescence emitted at 735 niA and at 683 miA. This result shows that reabsorption of the 683mu fluorescence emission was not a signifiealt factor in our experiments. Fluorescence emission of the supernatant fraction: The 144,000 g supernatant gave a relatively high fluorescence yield (2.7% at 200C), and the maximum was located at 680 m1A (Fig. 4), compared with 683 mA for chloroplasts and the fractions. These results suggest that part of the chlorophyll in the 144,000 g supernatant had been split from the native chlorophyll-protein complex and solubilized by the

590

BIOCHEMISTRY: BOARDMAN ET AL.

PROC. N. A. S.

digitonin. A detergent, such as Triton X-100 at a concentration of 4 per cent, which disrupts chloroplasts to the extent that the chlorophyll is nonsedimentable at centrifugal forces as high as 144,000 g for 1 hr", 12 results in a high fluorescence yield, which approaches that found for chlorophyll in organic solvents (Table 1). The fluorescence emission spectrum of chloroplasts in 4 per cent Triton X-100 has a maximum at 675 m1A. At 770K, the fluorescence emission of the 144,000 g supernatant shows two peaks, at 680 m~i and 735 mIu with almost equal amplitudes. The spectrum is very different from the low-temperature spectrum of the 144,000 g fraction, but it appears to be consistent with the suggestion of some solubilization of the chlorophyll. Fluorescence emission of fractions obtained from sonicated chloroplasts: Brody et al.13 have suggested that the amplitude ratio of the 735-mA band to the 683-mM band in chloroplast fragments at 770K is a function of the particle size of the fragments. A comparison was made, therefore, of the fluorescence properties of chloroplast fragments obtained by sonication with the properties of the particles obtained by digitonin fragmentation. Chloroplasts were disrupted by allowing them to stand in water for 10 min at 0WC, followed by sonication for 90 see at 10 kc. The fragments were separated by differential centrifugation, as described for the digitonin experiments. The fluorescence emission spectra, both of the 10,000 g and 144,000 g fractions, resembled the spectrum of chloroplasts at 20'C and 770K, in distinct contrast to the fractions produced by digitonin fragmentation. Excitation spectra of the 10,000 g and 144,000 g fractions obtained by digitonin fragmentation: Excitation spectra of the 10,000 g and 144,000 g fractions are shown in Figures 5 and 6. Above 600 mAt the spectra are only partially corrected and no quantitative significance can be attached to the relative heights of the band at 650 my due to chlorophyll b, and the main band due to chlorophyll a. For the 10,000 g fraction, excitation spectra were measured at 20'C for the fluorescence emission at 683 m/i, and at 770K for the fluorescence emissions at 683, 693, and 735 m/A. For the 144,000 g fraction, excitation spectra were measured for the 683-m/A band at 20'C, and the 735-mA band at 770K. Excitation spectra for chloroplasts were similar to those obtained with the 10,000 g fraction. Absorption spectra of the 10,000 g and 144,000 g fractions at 770K are shown in Figure 7. The 10,000 9 fraction shows enhanced absorption at 650 mjA and 470 mIA, due to its higher content of chlorophyll b.3 7 The shoulder at 490 mjA is due to carotenoid. The 144,000 g fraction shows a shoulder at 705-710 mIA, which is absent from the spectrum of the 10,000 g fraction. A comparison of the excitation and absorption spectra for both the 10,000 g and 144,000 g fractions shows that quanta absorbed at 470 and 490 mjA were more effective in promoting fluorescence than quanta absorbed at 436 m'u, the absorption maximum of chlorophyll a. But it is also apparent that quanta absorbed at 470 and 490 mMA, relative to those absorbed at 436 mj., were more effective in exciting low-temperature fluorescence in the 10,000 g fraction than in exciting 735-mM fluorescence in the 144,000 g fraction. The solid lines in Figures 5 and 6 show that 470-mn light, relative to 436-ma light, was also more effective in exciting 683-mg fluorescence in the 10,000 g fraction at 20'C than in the 144,000 g fraction. The higher effectiveness of 470-miu light with the 10,000 g fraction apparently reflects the higher content of chlorophyll b in this fraction. The pronounced band at 650

VOL. 56, 1966

BIOCHEMISTRY: BOARDMAN ET AL.

591

6my in the excitation spectrum of the 10,000 g fraction of 200 supports this conclu- ! sion. 4 '7 //o , Discussion.-The fluorescence properties of chloro ,phyll a in vivo have been re- i 2 lated to the two photochem- 0 I' ical systems of photosyntheK sis. Duysens and Sweers8 500 540 580 620 70C 420 660 460 A mu concluded that chlorophyll a in system 1 (chl a,) is weakly FIG. 5.-Fluorescence excitation spectra of the 10,000 g or nonfluorescent, while fraction. Conditions as for Fig. 1. , 20'C, F683 mju chlorophyll a in system 2 (X3); ...-. 77OKp F683 mA; -. -----X-*----, 770K, F693 (chl a2) is capable of a rela8 tively strong fluorescence. Light absorbed by system 2 increased the fluorescence 2 6 yield of chl a2, and light ab_z sorbed by system 1 lowered / the enhanced yield. The \ x 4 fluorescence A, light-induced U were changes attributed to chemical changes in a subO 2 201C stance Q, present in the (6 neighborhood of chl a2. Q E m , 0 0

/

was oxidized by system 1

420

540

500

460

light, and the oxidized form

580

6 60

620

700

AMg

quenched the fluorescence of FIG. 6.-Fluorescence excitation spectra of the 144,000 g fraction. Conditions as for Fig. 1. chl a2 by allowing electron , 20'C, F683 flow. Light absorbed by 3 system 2 reduced Q, which led to an increased fluorescence yield. The enhanced | fluorescence in the presence U 6 of dithionite or CMU was I \ IQ000 fraction 0 attributed 1 to the1 accumula-1