Photosynthesis Research 30: 139-143, 1991. © 1991 Kluwer Academic Publishers. Printed in the Netherlands.

Technical communication

The use of non-denaturing Deriphat-polyacrylamide gel electropboresis to fractionate pigment-protein complexes of purple bacteria

Linda Ferguson, Evelyn Halloran, Anna M. Hawthornthwaite, Richard Cogdell, Cheryl Kerfeld ~, Gary F. Peter ~ & J. Philip Thornber 1. Department of Botany, University of Glasgow, Glasgow G12 8QQ, UK; 1Department of Biology, University of California, Los Angeles, California 90024-1606, USA; *Author for correspondence: J. Philip Thornber, Dept. of Biology, University of California, Los Angeles, California 90024-1606, USA

Received 1 April 1991; accepted in revised form 18 July 1991

Key words: Purple bacteria, pigment-protein, gel electrophoresis, detergents

Abstract

The suitability of Deriphat-polyacrylamide gel electrophoresis as a method for separating purple bacterial pigment-protein complexes has been tested. When appropriate non-denaturing detergents are used to solubilize chromatophores, this method provides a rapid, easy and microscale procedure for analyzing the composition of the bacterial photosynthetic apparatus with minimal disruption of in- dividual pigment-proteins. Its usefulness is further illustrated by employing it to test for suitable detergents with which to solubilize purple bacterial chromatophores, and as an assay to study variation in the composition of the photosynthetic unit of bacterial cultures grown under different conditions.

Introduction

Electrophoresis of detergent extracts of higher plant and algal photosynthetic membranes through molecular sieving materials (e.g., poly- acrylamide) has been used since the early 1960s to obtain individual pigment-proteins or pigmen- ted multiprotein complexes (Thornber et al. 1991). Such procedures have been continually improved as a wider selection of detergents, particularly glycosidic surfactants, has become available; components can now be isolated in a state that better reflects their in situ form and composition. Rather surprisingly, electrophoretic methods have only been used occasionally for equivalent studies on purple bacteria (Broglie et al. 1980, Thornber 1970). Instead, separation of chromatophore extracts by column chromatog- raphy has been the preferred method (e.g., Cog- dell 1986). Column-based methods allow a larger scale production of the complexes; however,

they are often harsher than the electrophoretic methods. Since the bacterial complexes are gen- erally more stable to detergents than their plant counterparts, this has not presented much of a problem in the past. Nevertheless, the 'core' B870-type antenna complexes of some organisms are rather unstable and are often not readily obtained from column-based methods. Conse- quently, characterization of them has lagged be- hind studies of the B800-850 components. Therefore, alternative purification procedures are needed.

It was found (Markwell et al. 1979) in higher plants that the use of the mild detergent, De- riphat-160, effectively stabilized pigment-protein complexes during gel electrophoresis, preventing dissociation of the pigments from their apopro- teins. We therefore decided to see whether the Deriphat-PAGE method (Peter and Thornber 1991) would also be useful for fractionation of pigment-protein complexes from a range of put-

140

pie bacteria. This method involves solubilization of the photosynthetic membrane with a mixture of two glycosidic detergents (e.g., laurylmalto- side and octyl-glucoside) and then separating them electrophoretically in the presence of De- riphat-160.

Materials and methods

The purple bacteria, Rhodobacter sphaeroides, strain 2.4.1, Rhodopseudomonas acidophila, strain 7750 and 7050, Rhodopseudomonas palus- tris, strain 2.1.6, and Chromatium purpuratum, were grown anaerobically in the light as previ- ously described (Cogdell et al. 1983, Cogdell et al. 1990, Evans et al. 1990). After growth the ceils were harvested, resuspended in 20mM MES, pH6.5, 100 mM KCI and distributed by passage through a French pressure cell at - 150MPa , in the presence of MgC12 and DNAase. The photosynthetic membranes were then isolated by centrifugation (Cogdell et al. 1983) and resuspended in 20mM Tris-HCl, pH 8.0. The membranes were either used imme- diately or stored frozen at - 20 °C until required.

The membranes were fractionated by the De- riphat-PAGE method essentially as described by Peter and Thornber (1991). In each case the following protocol was used: The concentration of bacteriochlorophyll a (Bchl a) in the mem- brane sample was determined by extraction into 7 : 2 (v/v) acetone : methanol (Clayton 1963), using a miUimolar extinction coefficient at 772 nm of 76 cm-Z. A sample of the membranes was solubilized with a 1 :1 (w/w) mixture of lauryl-maltoside and octyl-glucoside. The ratio used of Bchl a : detergent (between 1 : 25-1 : 40) is stated in each figure, and the final detergent concentration was 1% (w/v). Unsolubilized ma- terial was removed by a brief (2-3 min) centrifu- gation in a microcentrifuge and the supernatant was loaded onto an 8.5% polyacrylamide gel (Peter and Thornber 1991). The gel was elec- trophoresed at 4 °C at 20 mA and stopped when discrete pigment-protein complexes could be seen (45-75 min). This procedure can be scaled up by increasing the thickness of the gel from the 2 mm used here to 1 cm. Deriphat-160 can be obtained from Henkel Corp., Hoboken, New Jersey.

The absorption spectra of the complexes in the gel were determined by excising the pigmented bands and recording their spectra in a Shimadzu UV-160 spectrophotometer.

The polypeptide composition of the fraction- ated complexes was determined by excising the requisite band and packing it into a well of a fully denaturing SDS-polyacrylamide gel (Laemmli 1970). The gel pieces were preincu- bated in a buffer containing 4% SDS, 100mM dithiothreitol, 10% glycerol, 20 mM Tris-HC1, pH 8.0, for 30 min at room temperature.

Results and discussion

When glycosidic surfactant extracts of Chr. pur- puratum (Fig. IA) were separated by Deriphat- PAGE, two spectrally distinct major pigment- protein complexes were observed. Figure 2 shows the absorption spectra of the two major bands for Chr. purpuratum. The top band (Fig. 2A) shows an absorption spectrum typical of a core complex, the B875-reaction center complex (Cogdell et al. 1990, Thornber 1970). The main absorption band in the near-infrared at 871 nm arises from the antenna complex and the smaller bands at 800 nm and 760 nm reflect the presence of some of the Bchl a and bacteriopheophytin a in the reaction center. The bottom band (Fig. 2B) shows the absorption spectrum typical of the B800-830-complex of this organism (Cogdell et al. 1990). The middle band is almost certainly an aggregate of the fastest migrating band (since its spectrum is like that of the B800-830 complex). Importantly, there is almost no free pigment, even at the high solubilizing ratios of Bchl a : detergent used (1 :4) . This is in contrast to the results of other electrophoretic methods, such as the one using lithium dodecyl-sulphate (Broglie et al. 1980).

We have tested whether this method of frac- tionating purple bacterial photosynthetic mem- branes is generally useful by applying it to other purple bacteria, Rb. sphaeroides, Rps. palustris and Rps. acidophila (Fig. 1B). In each case the method worked well and maintains the in vivo spectrum of the pigment-proteins. As shown for Chr. purpuratum (Fig. 2) the top band is the B875-reaction center core component (cp. Fig. 1A and 1B). The major bottom band, which

141

Fig. 1. Use of the Der ipha t -PAGE to fractionate the pigment-protein complexes from a range of purple bacteria. Part A: Chr. purpuratum, at a Bchl a : detergent of 1 : 40 (w/w). Part B: Lane 1 - Rb. sphaeroides, at a Bchl a : detergent of 1 : 30 (w/w); Lane 2 - Rps. palustris, at a Bchl a : detergent of 1 : 40 (w/w); Lane 3 - Rps acidophila, at a Bchl a : detergent of 1 : 25 (w/w). Part C: The use of Der ipha t -PAGE to test the efficacy of different detergents to solubilize chromatophores. Membranes were from Rps. palustris, and all detergents were used at a Bchl a : detergent of 1 : 40 (w/w); Lane 1 - lauryl-maltoside and octyl-glucoside; Lane 2 - sodium cholate; Lane 3 - lauryldimethylamine-N-oxide (LDAO); Lane 4 - s o d i u m lauryl-sulphate (SDS); Lane 5 - Triton X-100. Part D: Fractionation of membranes from high and low light grown cells of Rps. acidophila strain 7050 with the Der ipha t -PAGE. Left - high light grown cells; Right - low light grown cells. In both cases the membranes were solubilized at a

Bchl a : detergent ratio of 1 : 25 (w/w).

142

I I !

A ! 871 I

4°° A

I i i I t 0 B 826

I I I I " ~ 400 500 600 700 800 900

WAVELENGTH (nm) Fig. 2. The absorption spectra of the two major bands from a Deriphat-PAGE of Chr. purpuratum. A - the core B875- RC complex; B - the B800-830 complex. The spectra were recorded of the pigmented bands in the gel slice.

occasionally separates into two bands, represents the peripheral B800-850 complex. A minor Bchl a-free band containing carotenoids and proteins migrates the most rapidly.

In order to confirm that the photosynthetic unit had been fractionated into its core and peripheral components by Deriphat-PAGE, the upper and lowermost major bands of Rps. palus- tris were cut out of the gel (Fig. 1B), and their subunit composition analyzed by fully denaturing SDS-PAGE (Fig. 3). It is clear that a single elctrophoretic step does not yield pure antenna complexes, since each band has more than the expected number of polypeptides; although, some of the protein bands may be aggregates of lower molecular weight proteins (C. Kerfeld and J.P. Thornber, unpublished). Note that in the size region of the antenna apoproteins ( -5 kDa) there is a distinction between the two lanes: In lane 1 the B875-apoproteins have a different mobility than those of the B800-850-complex (lane 2). Deriphat-PAGE, therefore, produces spectrally pure complexes, but further purifica- tion is needed to produce a homogeneous corn-

Fig. 3. Analysis of the polypeptide composition of the two pigment-protein complexes obtained from Rps. palustris and separated by Deriphat-PAGE followed by a fully denaturing SDS-PAGE on a 16.5% polyacrylamide gel (Laemmli 1970). Lane 1 - The core B875-RC complex; Lane 2 - The B800- 850 complex. The position of marker protein are shown on the right-hand side.

plex with respect to its constituent polypeptides. For Chr. purpuraturn we have found that FPLC or ammonium sulfate fractionation of material electroeluted from Deriphat-PAGE provides such purification (C. Keffeld and J.P. Thornber, unpublished).

The Deriphat-PAGE system is also a useful tool for testing the suitability of different de- tergents for the preparation of antenna complex- es (Fig. 1C). In Fig. 1C Rps. palustris chromato- phore membranes have been solubilized with a range of commonly used detergents, and then fractionated by Deriphat-PAGE. Lane 1 is a control with laury-maltoside and octyl-glucoside as the solubilizing agents; sodium cholate was used in lane 2. Clearly sodium cholate solubilizes the membranes poorly, and some very large molecular weight species do not enter the gel.

143

Lane 3 shows that LDAO effectively solubilizes the membranes and lane 4 shows that SDS com- pletely denatures the core complex. Triton X-100 gives promising results (lane 5); however, the core complex is often contaminated by some of the B800-850 complex (data not shown).

The photosynthetic unit of most purple bac- teria is a rather plastic structure, its composition being modulated by environmental factors such as light-intensity and temperature (Cogdell et al. 1983). Deriphat-PAGE provides a very rapid, microscale method for studying pigment dis- tribution in the photosynthetic unit under chang- ing environmental conditions (e.g., Fig. 1D). For example, cells of Rps. acidophila strain 7050 were grown at high and low light intensities (Cogdell et al. 1983). The cells were broken and the photosynthetic membranes prepared and then solubilized. The composition of the photo- synthetic unit, in each case, was examined on Deriphat-PAGE (Fig. 1D). The low light grown cells (right) have much more peripheral antenna complex (the bottom purple band) than the core B875-RC complex (the top orange/brown band) whereas the relative amounts are much more nearly equal in the high light grown cells (left). This figure also illustrates how the color of the major band differs between the two growth con- ditions. This is due to a change in the type of carotenoid present, a change that is induced by altering the light intensity.

Since only small amounts of membranes are needed for these gels, it is also feasible with Deriphat-PAGE to follow changes directly in the composition and the amounts of the pigment- protein complexes during development. Further- more, characterization of mutants with altered pigment-protein content should also be straight- forward with the Deriphat-PAGE system

Acknowledgements

This research was supported by funds from the S.E.R.C. (R.J.C.), the E.E.C. (R.J.C.), N.S.F. (J.P.T.), and the U.S.D.A. Higher Education Program (C.K).

References

Broglie RM, Hunter CN, Delepelaire P, Niederman RA, Chua N-H and Clayton RK (1980) Isolation and charac- terization of the pigment-protein complexes of Rhodo- pseudomonas sphaeroides by lithium dodecyl sulfate/poly- acrylamide gel electrophoresis. Proc Natl Acad Sci USA 77:86-91

Clayton RK (1963) Absorption spectra photosynthetic bac- teria and their chlorophylls. In: Gest H, San Pietro A and Vernon LP (eds) Bacterial Phostosynthesis, pp. 495-500. Antioch Press, Yellow Springs, OH

Cogdell RJ (1986) Light-harvesting complexes in the purple bacteria. Enc Plant Physiol, New Series 19:252-259

Cogdell RJ, Durant 1, Valentine J, Lindsay JG and Schmidt K (1983) The isolation and partial characterization of the light-harvesting pigment-protein complexes of Rhodo- pseudornonas acidophilia. Biochim Biophys Acta 722: 427- 435

Cogdell R J, Hawthornthwaite AM, Evans MB, Ferguson LA, Kerfeld C, Thornber JP, van Mourik F and van Grondelle R (1990) Isolation and characterization of an unusual antenna complex from the marine purple sulfur bacterium Chromatium purpuratum BN 5500. Biochim Biophys Acta 1019:239-244

Evans MB, Hawthornthwaite AM and Cogdell RJ (1990) Isolation and characterization of the different B800-850 light-harvesting complexes from low- and high-light grown cells of Rhodopseudomonas palustris, strain 2.1.6. Biochim Biophys Acta 1019:71-76

Laemmli UK (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 227: 680-685

Markwell JP, Thornber JP and Boggs RT (1979) Higher plant chloroplasts: Evidence that all the chlorophyll exists as chlorophyll-protein complexes. Proc Natl Acad Sci USA 76:1233-1235

Peter GF and Thornber JP (1991) Electrophoretic proce- dures for fractionation of Photosystems I and II pigment- proteins of higher plants and for determination of their subunit composition. In: Rogers LJ (ed) Methods in Plant Biochemistry, Vol 5, pp 195-210. Academic Press, San Diego

Thornber JP (1970) Photochemical reactions of purple bac- teria as revealed by studies of three spectrally different caroteno-bacteriochlorophyll-protein complexes isolated from Chromatium vinosum. Biochemistry 9:2688-2698

Thornber JP, Morishige DT, Anandan S and Peter GF (1991) Chlorophyll-carotenoid proteins of higher plant thy- lakoids. In: Scheer H (ed) Chlorophylls, pp 549-585. CRC Press, Boca Raton