Báo cáo khoa học: NblA from Anabaena sp. PCC 7120 is a mostly a-helical protein undergoing reversible trimerization in solution pot

NblA from Anabaena sp. PCC 7120 is a mostly a-helical protein undergoing reversible trimerization in solution Holger Strauss 1 , Rolf Misselwitz 2 , Dirk Labudde 1 , Sabine Nicklisch 3 and Kerstin Baier 3 1 Forschungsinstitut fu ¨ r Molekulare Pharmakologie (FMP), Berlin, Germany; 2 Max-Delbru ¨ ck-Centre fu ¨ r Molekulare Medizin (MDC), Berlin, Germany; 3 Humboldt Universita ¨ t zu Berlin, Institut fu ¨ r Biologie/Biochemie der Pflanzen, Germany The nblA family of genes encodes for small proteins neces- sary for the ordered degradation of phycobilisomes under certain stress conditions, a process known as chlorosis. Genes homologous to nblA seem to occur in all phycobili- some-containing organisms. However, to date, no molecular mechanism is known for the action of NblA, nor have the gene products been characterized to understand the physical properties of the molecule and thus help elucidate the mechanism on a structural basis. In this study we report on the first characterization of an NblA-homologous gene product. The chromosomal gene from the cyanobacterium Anabaena sp. PCC 7120 was cloned, heterologously expressed in Escherichia coli and purified to apparent homogeneity. This allowed the protein to be characterized by analytical ultracentrifugation and CD spectroscopy. These experiments show that the NblA protein has a mostly a-helical structure, undergoing an association reaction of folded monomers to form trimers in solution. No dimers are detectable. Keywords: phycobilisome; chlorosis; NblA; cyanobacteria; analytical ultracentrifugation. Cyanobacteria are a widespread group of photosynthetic prokaryotes performing a plant-type oxygenic photosyn- thesis. They are very adaptable organisms that can survive in a wide variety of environmental conditions [1,2]. One limiting factor for growth is the nitrogen supply and cyanobacteria have developed various mechanisms to cope with this nutrient stress. One of the first responses exhibited by cyanobacteria when they are starved for nitrogen is the degradation of their major light-harvesting complex, the phycobilisome. Phycobilisomes (PBS), which also represent light-harvesting antennae of red algae, are large, water-soluble multiprotein complexes associated with the thylakoid membranes. PBS consist mainly of the pigmented phycobiliproteins that can constitute up to 50% of the total cellular protein, thus representing a large nitrogen store [3]. Degradation of PBS is thought to provide substrates for protein synthesis required for the acclimatization process. In addition, PBS degradation minimizes the absorption of excess excitation energy under the stress situation. Nondiazotrophic cyanobacteria such as Synechococcus sp. PCC 7942 completely degrade their PBS when starved for combined nitrogen and differentiate into nonpigmented resting cells, able to survive prolonged periods of nutrient stress [4,5]. Diazothrophic filamentous cyanobacteria such as Anabaena sp. PCC 7120 adapt to nitrogen limitation (lack of combined nitrogen) by developing differentiated cells, called heterocysts. These are specialized for fixation of N 2 in an aerobic environment [6]. In a filament, approxi- mately 5–10% of vegetative cells undergo this differenti- ation process. However, during the first hours of nitrogen starvation all cells start to degrade their PBS [7]. When heterocysts mature and nitrogenase is active, vegetative cells resynthesize their light-harvesting complexes, while in heterocysts the PBS content remains very low [8,9]. Phycobilisome degradation is an ordered proteolytic process, visible by a colour change of the cyanobacterial cell from blue-green to yellow-green, a process known as chlorosis or bleaching [10]. The small polypeptide NblA plays a central role in PBS degradation. Its gene, nblA,was first identified in Synechococcus PCC 7942 [11], but nblA homologous genes appear to be present in all PBS- containing organisms, cyanobacteria as well as red algae. In Synechococcus PCC 7942, nblA transcription is induced upon nitrogen and sulfur starvation, and, to a lesser extent, during phosphorus starvation [11]. In Synechocystis sp. PCC 6803, only nitrogen starvation leads to nblA induction [12]. Knock-out mutations of the nblA gene lead to nonbleaching phenotypes under nitrogen-limited conditions [11,13]. Several NblA homologous sequences are found in the databases. The sizes of the deduced NblA proteins range from 54 to 65 amino acids, corresponding to molar masses of about 7–7.5 kDa. Sequence identity among these NblA proteins amounts to about 30% on average, but no homology has been found to other proteins with known function. The molecular mechanism by which NblA triggers Correspondence to K. Baier, Institut fu ¨ r Biologie/Biochemie der Pflanzen, Humboldt Universita ¨ t zu Berlin, Chausseestr. 117, D-10115 Berlin, Germany. Fax: + 49 30 20938164, Tel.: + 49 30 20938166, E-mail: kerstin.baier@biologie.hu-berlin.de or H. Strauss, Forschungsinstitut fu ¨ r Molekulare Pharmakologie, Robert-Ro ¨ ssle Str. 10, 13125 Berlin, Germany. Fax: + 49 30 94793 169, Tel.: + 49 30 94793 223, E-mail: strauss@fmp-berlin.de Abbreviations: PBS, phycobilisomes; nbl, nonbleaching; AUC, analytical ultracentrifugation; SV, sedimentation velocity; SE, sedimentation equilibrium. (Received 22 May 2002, revised 21 July 2002, accepted 1 August 2002) Eur. J. Biochem. 269, 4617–4624 (2002) Ó FEBS 2002 doi:10.1046/j.1432-1033.2002.03161.x degradation of PBS is not clear. The following hypotheses have been discussed [11]: NblA may activate a protease degrading the PBS; alternatively, NblA could tag or disrupt the PBS, rendering it susceptible to degradation; and finally, NblA may activate other genes that are involved in the PBS degradation process. Analysis of the structural properties of NblA could help find out how this polypeptide achieves its function. The diazotrophic, filamentous cyanobacterium Anabaena sp. PCC 7120 has two nblA genes, one (ORF asr4517) on the chromosome and another (ORF asr8504) on plasmid Delta [14]. We have cloned and overexpressed nblA from ORF asr 4517 in Escherichia coli. This allowed the NblA polypeptide to be purified to apparent homogeneity and thus to study the gene product to determine some of its structural and physical properties. We used analytical ultracentrifugation (AUC) to study the hydrodynamic properties of the protein and to determine its state of association as a function of concentration. The stoichiometry of the association reac- tion, as well as the extent, in terms of an association constant at three different temperatures (10, 18 and 26 °C), were determined. CD, together with fluorescent measurements, was used to probe the gross secondary structure and any changes observable with temperature and concentration. MATERIALS AND METHODS Construction of the expression plasmid, protein expression and purification The chromosomal gene (asr4517) (Table 1) coding for NblA from Anabaena sp. PCC 7120 [14] was amplified, with total DNA isolated from that strain as template, by PCR using the following oligonucleotides: 5¢-GTCTTTT AGGAGTCTCATATGAACC-3¢, complementary to a DNA region upstream of the N-terminus, with NdeIsite inserted (in italics) and 5¢-GTTGACGCCCCAGGATCCC CAGCTC-3¢, complementary to a DNA region down- stream of the C-terminus, with BamHI site inserted (in italics). The PCR product was digested with NdeIand BamHI and ligated into plasmid pET11a (Novagen) resulting in plasmid pBB8 (Fig. 1A). For expression of the NblA protein, plasmid pBB8 was cloned into host strain E. coli BL21 (DE) pLysS (Novagen). Two liters of Luria–Betrtani medium (Difco Laborator- ies) were inoculated with 100 mL of a bacterial overnight culture (50 lgÆmL )1 ampicillin, 34 lgÆmL )1 chlorampheni- col) and grown at 30 °CtoaD 600 1 of 0.5–0.8. Isopropyl thio- b- D -galactoside was added to a final concentration of 1 m M and incubation continued for 3 h. Cells were harvested by centrifugation (5000 g for 10 min at 4 °C) and washed with Tris/NaCl/EDTA (50 m M Tris/HCl pH 7.5, 150 m M NaCl, 1m M EDTA). Cells were disrupted by sonication and centrifuged (14 000 g for 10 min at 4 °C). (NH 4 ) 2 SO 4 was added to the supernatant (30% saturation, 10 min centri- fugation at 14 000 g at 4 °C), the pellet dissolved in 25 mL of Tris/EDTA (50 m M Tris/HCl pH 7.5, 1 m M EDTA) and applied to a column of Superdex 75 (HiLoad 16/60, Pharmacia Biotech), run in Tris/NaCl/EDTA (1 mLÆmin )1 ). Fractions of 2 mL were collected and assayed for NblA in a discontinuous tricine/SDS/PAGE system [15]. The protein eluted at a volume corresponding to 19–24 kDa. The column was calibrated with BSA, myoglobin and cytochrome c (67, 17.8 and 12.3 kDa, respectively). Fractions containing NblA were pooled and concentrated by precipitation as above. After desalting on a PD-10 column (Amersham Biosciences Europe), the preci- pitated protein was applied to a column of Q-Sepharose (1 mL bed volume), equilibrated with Tris/EDTA buffer and eluted with a linear gradient of NaCl (0–300 m M in 30 min) in Tris/EDTA at a flow rate of 1 mLÆmin )1 .NblA eluted at 150 m M NaCl (Figs 1B.C). All purification steps were performed at 0–8 °Cwiththe exception of chromatography on Q-Sepharose, which was carried out at room temperature. MALDI-TOF mass spectrometry Mass spectrometry measurements were performed on a Voyager-DE STR BioSpectrometry Workstation MALDI- TOF mass spectrometer (Perseptive Biosystems, Inc., Fra- mingham, MA, USA), using a standard protocol as described [16]. After analytical ultracentrifugation, sample solutions were taken from the centrifugation cells, pooled, and subjected to the sample preparation procedure without further dilution. Table 1. Results from SV experiments of NblA Parameter estimates were obtained from fits over the whole boundary to the monomer-trimer model. The molecular mass parameter was kept constant at the value calculated from the sequence and obtained with MALDI-TOF MS. m, molecular mass. Parameter Concentration of NblA 14 (mgÆL )1 /l M ) 300/40 400/53 500/66 Number of datapoints 15461 12241 12542 rmsd 0.003467 0.003901 0.005989 S monomer [S )13 ] (fitted) 0.84 0.83 0.82 S trimer [S )13 ] (fitted) 2.24 2.23 2.21 D monomer [10 7 cm 2 Æs )1 ] (calculated) 10.07 10.28 10.09 D trimer [10 7 cm 2 Æs )1 ] (calculated) 8.96 9.18 9.08 f/f 0 monomer (calculated from m and s) 1.58 1.59 1.62 f/f 0 trimer (calculated from m and s) 1.23 1.24 1.25 K a [ M )2 ] 2.06 · 10 10 2.11 · 10 10 1.99 · 10 10 4618 H. Strauss et al. (Eur. J. Biochem. 269) Ó FEBS 2002 Protein concentrations were determined spectrophoto- metrically using the extinction coefficient at 280 nm as calculated from the sequence [17]. Values at all other wavelengths used were calculated relative to that value from wavelength spectra recorded with appropriate concentra- tions of NblA on a JASCO V-550 spectrometer. Absorb- ance measurements at different wavelengths were transformed to molar concentrations using the law of Lambert and Beer. All optical measurements were carried out with buffer in dialysis equilibrium with the solution. Circular dichroism and fluorescence measurements CD studies in the far UV region were performed with a Jasco J720 spectropolarimeter equipped with a Neslab temperature control system using 0.01–1.0 cm path length quartz cuvettes and protein concentrations in the range 0.53–53 l M (4–400 mgÆL )1 ). Measurements were performed at (10 ± 0.2) °C. Molar mean residue ellipticities [Q] (degÆcm 2 Ædmol )1 ) were calculated using a mean residue molecularmassof116.0Da. The content of secondary structure was determined from the far-ultraviolet CD spectra using the variable selection method (program VARSLC 1) starting with a set of 33 reference proteins [18]. Fluorescence spectra were measured with a Shimadzu RF 5001 PC spectrofluorimeter at excitation wave- lengths of 295 nm and 280 nm with bandwidths of 5 nm for both excitation and emission monochro- mator. Concentration of protein solutions were adjusted to 5.3 l M (40 mgÆL )1 ) and were measured in cuvettes of 0.3 cm path length at (10 ± 0.2) °C. The intensityoftheRamanpeakofwaterwasusedasan internal standard. Thermal-induced unfolding measured by circular dichroism Thermal unfolding of NblA was carried out in 20 m M sodium phosphate buffer, pH 7.5 monitoring changes in the ellipticity at 222 nm at protein concentrations in the range 0.54–43.7 l M (4.1–330 mgÆL )1 ) and at a heating rate of 20 °CÆh )1 . The reversibility of unfolding of the protein was checked by slow cooling down to 20 °C. The transition curves were normalized to the fraction of folded protein f, where f ¼ ([Q] ) [Q u ](T))/([Q n ](T) ) [Q u ](T)) where [Q n ] and [Q u ] are the mean residue ellipticities of the folded and unfolded protein, respectively, and were corrected for their temperature dependence by linear extrapolation of the pre- and post-melting range. [Q] is the observed mean residue ellipticity. Analytical ultracentrifugation Both sedimentation velocity (SV) and sedimentation equi- librium (SE) experiments were performed on a Beckman XL-I analytical ultracentrifuge (Beckman-Coulter, Fuller- ton CA, USA) in a four-hole AN Ti60 rotor, using the absorption optics of the instrument. The partial specific volume (  vv) of NblA was calculated from the sequence [19,20]. Values for the density (q) and viscosity (g)ofthe buffer used were calculated from composition using the options implemented in ULTRASCAN 5.0 (B. Demeler, University of Texas, Health Science Center at San Antonio, TX, USA) 2 . Values are corrected for the temperatures used [21–23]. SV experiments were performed at 20 °C and 250 000 g in double-sector, charcoal filled epon centerpieces capped with quartz windows over a concentration range of 300– 1000 mgÆmL )1 (40 l M to 133 l M ). Sedimentation patterns were acquired at a single wavelength for a single experiment (286 nm and 278 nm, depending on loading concentration) in continuous mode every 90 s with a Dr of 0.003 cm. Data were analysed by fitting the sedimentation patterns to the Lamm equation [24–28] and by the method of van Holde-Weischet [29,30]. All sedimentation (S) 4 and diffusion coefficients (D) as reported here have been corrected for water at 20 °C [21–23]. Fig. 1. Preparation and characterization of NblA from E. coli. (A) Restriction map of the T7 lac promoter-nblA region of plasmid pBB8. For details of construction, see Materials and methods. (B) Purifica- tion of NblA from recombinant E. coli cells, harbouring plasmid pBB8. Discontinuous tricine/SDS/PAGE of different stages of purifi- cation. Lane 1, noninduced cells; lane 2, induced cells; lane 3, soluble lysate; lane 4, after Superdex 75 and lane 5, after Q-sepharose chro- matography. The positions of standard marker proteins are indicated on the left. (C) MALDI-TOF MS of NblA, after the final purification step. The peak at 3772.05 (m/z) represents the doubly charged monomer, the peak at 7749.74 (m/z) corresponds to the singly charged matrix adduct. Ó FEBS 2002 Characterization of NblA (Eur. J. Biochem. 269) 4619 SE experiments were performed in 12 mm, six-sector charcoal filled epon 5 centerpieces loaded with 75 lLof protein solution in each sector at different concentrations for each of the nine sectors, in the range 20–1000 mgÆL )1 (2.7–133 l M ) in repeated experiments. Detection wave- lengths ranging from 225 to 290 nm were chosen so that A initial was 0.1–0.4 6 for the respective cell and three different wavelengths were used for detection of each concentration gradient. Data were acquired in step mode with a Dr of 0.001 cm and 20 replicate absorption measurements were performed at every step point. After overspeeding the solution for 20–30 min at 30 000–38 000 r.p.m. 7 (depending on the rotor speed which was later on used for attainment of equilibrium [31,32]), the samples were spun in repeated experiments at various speeds, in the range 18 000– 32 000 r.p.m. 8 as indicated. Equilibrium was judged to be reached when a fit to the concentration gradients of a single molecular species model of the form: 9 c r ¼ c 0 e m app F þ d ð1Þ where F ¼½ð1 À q  vvÞx 2 ðr 2 À r 2 0 Þ=2RT ð2Þ didn’t show any 10 systematic deviations in the residuals; c r is the concentration of the solute at position r of the cell, c 0 the concentration at an arbitrarily selected reference position, m app is the apparent molecular mass 11 ,  vv 11 is the partial specific volume of the solute, x the angular velocity and d the baseline offset. Temperatures were kept constant during one experiment. In repeated experiments, we chose different temperatures (10, 18 and 26 °C) to understand in more detail the nature of the association reaction. Datasets were globally fitted using the general nonlinear least-squares procedures as described previously [33] and the extensions of Eqn (1) for multiple species in reversibly associating equilibrium [34], taking into account the associ- ation constants. Data were analyzed using the programs LAMM [25,26], SEDFIT 8.3 [27] and ULTRASCAN 5.0. RESULTS Analytical ultracentrifugation The chromosomal nblA gene from Anabaena sp. PCC 7120, ORF asr4517, encodes a polypeptide of 65 amino acids with a predicted molecular mass of 7542 Da. During purification of recombinant NblA from E. coli, the protein eluted from size exclusion chomatography columns with an apparent molar mass of 19–24 kDa. However, SDS/PAGE and MALDI-TOF MS confirmed the purity and identity of the sample (Fig. 1B,C), thus suggesting that NblA was multi- meric or highly elongated in solution. We routinely checked NblA by MALDI-TOF MS to test the stability of the protein under the experimental conditions used and no degradation was detected. A Van-Holde Weischet analysis of the sedimenting bundaries at different loading concen- trations indicated mass-dependent heterogeneitity (not shown). Based on these results and the information obtained from the SE experiments (see below), we used direct boundary modeling [27] to a monomer-trimer system to gain insights into the hydrodynamic parameters of the monomer and the trimer. Sedimentation patterns of the 300, 400 and 500 mgÆL )1 (40, 53 and 66 l M ) loading concentrations were fitted over the whole boundary (Fig. 2). Apparent values of S, determined at the lowest and highest loading concentra- tion (for S monomer and S trimer , respectively), and the value of K a at 18 °C from the SE experiments were used as starting estimates. The monomer-trimer model yields random residuals over the whole range of fitted data points, as judged from the conventional presentation and the recently proposed two dimensional bitmap-presentation [28]. The hydrodynamic and some statistical parameters obtained from the best-fit values for the three different concentrations are given in Table 1. To understand the stochiometry and the nature of the association reaction in more detail, we have performed SE experiments over a range of concentrations, temperatures and speeds. Multiple datasets are best described by a monomer-trimer model, which results in random residuals over the whole region included in the fit (Fig. 3) and a monomer molecular mass in good agreement with the theoretical value deduced from the sequence and confirmed with MALDI-TOF MS. Using a monomer-dimer-trimer model to deconvolute the data showed that no detectable portion of dimer was present. Increasing the temperature increased the fraction of monomer present in solution. Higher order associates can be excluded for the concentra- tion range and conditions used in this study, because the average molecular mass level off at 20–22 kDa at the highest loading concentrations when fitted to Eqn (1). Detailed information for the values obtained is given in Table 2. Robustness of the parameter estimates was ascer- tained by Monte-Carlo simulations of the fitted data, using 10 000 iterations for each dataset. From this, the 95% confidence intervals were obtained and are reported for the molecular mass and the association constant parameters. CD experiments Under native conditions NblA is well folded with spectral characteristics of proteins with predominantly a-helical Fig. 2. Direct boundary modeling to a monomer-trimer model of the sedimentation patterns obtained with 300 mgÆL )1 loading concentration. (A) Raw data (points) and best fit (solid lines); (B) residuals of the fit; (C) residuals bitmap presentation of the fit. 4620 H. Strauss et al. (Eur. J. Biochem. 269) Ó FEBS 2002 secondary structure elements (Fig. 4A). The maximum at 192 nm and double minima at 222 and 208 nm, respect- ively, point to largely alpha helical structures. For reference, the CD spectrum in the presence of 7 M guanidine hydrochloride, which is typical for unfolded proteins, is also shown. Evaluation of the CD spectrum with the variable selection method starting with 33 reference proteins [18] supports this conclusion. About 61% a-helix, 10% b-sheet and 12% turn structures were calculated. Secondary structure prediction with the PHD program [35,36] classified NblA as an all a-helical protein and predicts about 71% a-helix, 2% b-structure and 28% loop structure. The prediction for residues with a reliability 12 index ‡ 5resultsin 63% a-helix, no b-structure and 28% loop structure, which is in good agreement with the secondary structure content estimated by CD measurements. PHD predicts two a-helical regions in the sequence of NblA, a shorter N-terminal helix from L9 to M25 and a longer C-terminal helix from H27 to Q55. The N-terminal eight and the C-terminal 10 amino acids are less structured (Table 3). The predicted helical character and the position of the helical elements are corroborated by homologous sequences in the PDB protein structure database. A FASTA search identified for the NblA sequence (from aa 2–28 and 24–65) highly similar sequences with helical structures for gene regulation XRCC4-DNA ligase, PDB entry 1IK9 (Y177– D208) and for synapse-enriched clathrin adaptor protein, PDB entry 1HX8 (D239–P280), respectively. No coiled-coil motives in the sequence of NblA were detected by either bioinformatic sequence analysis tools, such as COIL SCAN [ WISCONSIN Package Version 10.2, Genetics Computer Group (GCG), Madison, WI, USA], or homology searches. As shown by AUC experiments, the trimeric NblA dissociates at low protein concentrations to monomers. To understand the impact of association on secondary struc- ture, we performed CD measurements in the concentration range 0.53–53 l M (4–400 mgÆL )1 ), which corresponds to 98% to 17% of monomers, respectively. The spectra of NblA measured in Tris/NaCl/EDTA are largely identical and the ratio of [Q] 222 /[Q] 208 nm is % 1 and does not change considerably in its dependence on protein concentration (Fig. 4B). Thus, the folding of NblA in the trimeric and monomeric state is very similar with a high content of a-helical structures. Fluorescence experiments The emission position of tryptophan at an excitation wavelength of 295 nm depends on whether it is localized in a hydrophobic or hydrophilic surrounding and varies between in the range 320–350 nm. For NblA, we found an emission maximum at 346 nm, which correlates with a hydrophilic environment of the single tryptophan residue at position 56. Unfolding of NblA in 7 M guanidine hydro- chloride results in a small red shift of the emission maximum to about 350 nm and a decrease in the fluorescence intensity (not shown). Thermal-induced unfolding To measure the thermal stability and to investigate the dissocation/unfolding behavior of NblA, the thermal transition was measured at protein concentrations in the range 0.54–43.7 l M (4.1–330 mgÆL )1 )in20m M sodium Table 2. Results from SE experiments and from global fits to a mono- mer-trimer model. m,molecularmass. Temperature (°C) 10 18 26 Number of datapoints 5390 5044 4362 m (fitted) [Da] 7595 7576 7594 95% confidence limits Upper 7669 7609 7783 Lower 7522 7543 7399 K a [ M )2 ] 5.88 · 10 10 2.17 · 10 10 1.1 · 10 10 95% confidence limits Upper 7.07 · 10 10 2.31 · 10 10 1.91 · 10 10 Lower 4.88 · 10 10 2.05 · 10 10 0.64 · 10 10 Fig. 3. 15 Global fits of the monomer-trimer model to the equilibrium gradients obtained at various loading concentrations, speeds and tempera- tures. Points represent the raw data, solid lines the best fits. On top of the fits are shown the respective residuals. (A) Data obtained at 10 °C, 22 000 r.p.m. and 28 000 r.p.m. (B)Dataobtainedat18 °C, 22 000 r.p.m. and 28 000 r.p.m. (C) Data obtained at 26 °C, 18 000 r.p.m. and 27 000 r.p.m. Ó FEBS 2002 Characterization of NblA (Eur. J. Biochem. 269) 4621 phosphate, pH 7.5 (Fig. 4C). Transition curves were monitored by changes of ellipticity [Q]at222nmata heating rate of 20 °CÆh )1 . NblA unfolds under these conditions in a one step manner with an isodichroic point at about 203.5 nm and the unfolding is largely reversible (Fig. 4D). As expected for the unfolding of noncovalently associated NblA molecules the melting temperatures varied with the protein concentration ranging between about 53 °Cat0.54l M (4.1 mgÆL )1 )and66°Cat 43.7 l M (330 mgÆL )1 ). DISCUSSION To our knowledge, this is the first report on properties of an NblA-homologous gene product since the gene was first identified in 1994 in Synechococcus sp. strain PCC 7942 [11]. We have cloned the gene from Anabaena sp. PCC 7120 and purified the protein without tags. We show that NblA from Anabaena sp. PCC 7120 is a mostly a-helical protein. The results from the thermally induced folding-unfolding experiments indicate the presence of only a single domain Fig. 4. Far-UV-CD spectra of NblA (A), protein concentration dependence of far-UV-CD spectra of NblA (B), thermal denaturation of NblA (C) and CD spectra at various temperatures and after recooling (D). (A) The spectra were measured in 20 m M sodium phosphate buffer, pH 7.5 (thin line) andinthepresenceof7 M guanidine hydrochloride (thick line). Experiments were carried out at (20 ± 0.2) °C. (B) The spectra were recorded at (10 ± 0.2) °C in Tris/NaCl/EDTA buffer and at protein concentrations of 54.0 l M (solid line), 5.4 l M (dashed line), 1.08 l M (dashed-dotted line) and 0.54 l M (dotted line) (407, 40.7, 8.1 and 4.1 mgÆL )1 , respectively). The inset shows values of the ratio [Q 222 ]/[Q 208 ] as a function of concen- tration. (C) The temperature-induced unfolding was monitored by changes of ellipticity at 222 nm in 20 m M sodium phosphate buffer, pH 7.5 and protein concentrations of 0.54, 5.4, 15.0 and 43.7 l M (4.1, 40.7, 113 and 330 mgÆL )1 , respectively). Unfolding curves from left to right. (D) CD spectra in the peptide region at a concentration of 43.7 l M and temperatures of 20, 40, 60, 80, and 90 °C (solid lines, from bottom to top) and after recooling to 20 °C (dotted line). The unfolding was measured with a heating rate of 20 °CÆh )1 . Table 3. Sequence and PHD secondary structure prediction of NblA. The sequence of NblA is shown in bold figures and directly below is given the secondarystructureprediction.H,helix;E,extended;L,loop. 10 20 30 40 50 60 MNQPIELSLE QQFSIRSFAT QVQNMSHDQA KDFLVKLYEQ MVVREATYQE LLKHQWGLDS GSTPA LLLLE. .HH HHHHHHHHHH HHHHH HHHH HHHHHHHHHH HHHHHHHHHH HHHHH L LLLLL 4622 H. Strauss et al. (Eur. J. Biochem. 269) Ó FEBS 2002 [37], not unexpected for such a small protein. NblA undergoes trimerization of stable monomers in solution in a mass-dependent manner. In an equilibrium situation, no dimers are present and no higher associates could be found. Upon association, the gross secondary structure of the protein doesn’t change in an observable manner. The high frictional ratio f/f 0 of the monomer could indicate either a globular structure with a roughed surface or a highly elongated structure. Upon trimerization, f/f 0 decreases moderately. Based on purely geometrical considerations and the hypothesis of elongated monomers, these findings can be rationalized by a symmetrical arrangement of the monomers to a trimer with a threefold symmetry. The monomeric building blocks of such a structure would have two different interaction surfaces along their major axis, one being identical to the surface of the trimer and one that is buried within, forming the scaffold for their interaction. At present, it is not clear whether the monomer or the trimer is the biologically relevant species. However, several arguments favor the trimer to be the species responsible for NblA action: NblA proteins that we have investigated from cyanobacterial crude extracts (NblA1 and NblA2 from Synechocystis sp. PCC 6803) or purified in recombinant form from E. coli (NblA1 and NblA2 from Synechocystis sp. PCC 6803 and NblA protein from Anabaena sp. PCC 7120 encoded by plasmid Delta) all behave similarly on size exclusion chromatography columns, eluting at positions which correspond to the size of a trimer (data not shown). This finding suggests that trimerization is indeed an important prerequisite for NblA action in vivo. Moreover, although the sequence homology among the known NblA proteins is not very high, PHD predicts a similar helical arrangement for all of the analysed NblA sequences from cyanobacteria as well as red algae. Thus, we propose that all NblA-homologous molecules identified so far share a common overall structure and behave similarly to Anabaena sp. PCC 7120 as reported here. As mentioned in the introduction, one of the proposed modes of action of NblA is the destabilization of the structure of PBS, which facilitates attack by proteolytic enzymes already present in the cell. If so, it is tempting to speculate that for destabilization of PBS, the NblA-trimer might directly fit into the central channel of the hexamers of phycobilisome rods (reviewed in [3,38]), thereby displacing the so-called linker peptides that normally reside there and thus changing the structure of the phycobilisome rods, making them amenable to proteolysis. However, up to now, there has been no evidence for a direct interaction between NblA and PBS. Clearly, experimental evidence for the mechanism of NblA action is needed. Determination of the three-dimen- sional structure of the protein could yield insights into the action of this small protein family. ACKNOWLEDGEMENTS The authors which to thank E. Krause and H. Lerch, both FMP, for performing the MALDI-TOF MS measurements and G. Krause (FMP) for bioinformatics, and Prof J. Behlke (MDC) for helpful suggestions on the AUC. Prof H. Oschkinat (FMP), Prof H. Welfle (MDC) and Prof W. Lockau (HUB) are acknowledged for their continuing support. The authors wish to thank the FMP for financial support. REFERENCES 1. Bhaya, D., Schwarz, R. & Grossman, A.R. (2000) Molecular responses to environmental stress. In The Ecology of Cyano- bacteria (Whitton, B.A. & Potts, M., eds), pp. 397–442. Kluwer Academic Publishers, Dordrecht. 2. Tandeau de Marsac, N. & Houmard, J. (1993) Adaptation of cyanobacteria to environmental stimuli: new steps towards mole- cular mechanisms. FEMS Microbiol. Rev. 104, 119–190. 3. Grossman, A.R., Schaefer, M.R., Chang, G.G. & Collier, J.L. (1993) The phycobilisome, a light-harvesting complex responsive to environmental conditions. Microbiol. Rev. 57, 725–749. 4. Sauer,J.,Schreiber,U.,Schmid,R.,Vo ¨ lker,U.&Forchhammer, K. (2001) Nitrogen starvation-induced chlorosis in Synechococcus PCC 7942. Low-level photosynthesis as a mechanism of long-term survival. Plant Physiol. 126, 233–243. 5. Go ¨ rl, M., Sauer, J., Baier, T. & Forchhammer, K. (1998) Nitro- gen-starvation-induced chlorosis in Synechococcus PCC 7942: adaptation to long-term survival. Microbiology 144, 2449–2458. 6. Wolk, C.P., Ernst, A. & Elhai, J. (1994) Heterocyst metabolism and development. In The Molecular Biology of Cyanobacteria (Bryant, D.A., ed.), pp. 769–823. Kluwer Academic Publishers, Dordrecht. 7. Wood, N.B. & Haselkorn, R. (1980) Control of phycobiliprotein proteolysis and heterocyst differentiation in Anabaena. J. Bacter- iol. 141, 1375–1385. 8. Thomas, J. (1970) Absence of the pigments of photosystem II of photosynthesis in heterocysts of a blue-green alga. Nature 228, 181–183. 9. Peterson, R.B., Dolan, E., Calvert, H.E. & Ke, B. (1981) Energy transfer from phycobiliproteins to Photosystem I in vegetative cells and heterocysts of Anabaena variabilis. Biochim. Biophys. Acta 634, 237–248. 10. Allen, M.M. & Smith, A.J. (1969) Nitrogen chlorosis in blue-green algae. Arch. Mikrobiol. 69, 114–120. 11. Collier, J.L. & Grossman, A.R. (1994) A small polypeptide trig- gers complete degradation of light-harvesting phycobiliproteins in nutrient-deprived cyanobacteria. EMBO J. 13, 1039–1047. 12. Richaud, C., Zabulon, G., Joder, A. & Thomas, J.C. (2001) Nitrogen and sulfur starvation differentially affects phycobilisome degradation and expression of the NblA gene in Synechocystis strain PCC 6803. J. Bacteriol. 183, 2989–2994. 13. Baier, K., Nicklisch, S., Grundner, C., Reinecke, J. & Lockau, W. (2001) Expression of two NblA-homologous genes is required for phycobilisome degradation in nitrogen-starved Synechocystis sp. PCC 6803. FEMS Microbiol. Lett. 195, 35–39. 14. Kaneko, T., Nakamura, Y., Wolk, C.P., Kuritz, T., Sasamoto, S., Watanabe,A.,Iriguchi,M.,Ishikawa,A.,Kawashima,K., Kimura, T., Kishida, Y., Kohara, M., Matsumoto, M., Matsuno, A., Muraki, A., Nakazaki, N., Shimpo, S., Sugimoto, M., Takazawa, M., Yamada, M., Yasuda, M. & Tabata, S. (2001) Complete genomic sequence of the filamentous nitrogen-fixing cyanobacterium Anabaena sp. strain PCC 7120. DNA Res. 8, 205– 213. 15. Scha ¨ gger, H. & von Jagow, G. (1987) Tricine-sodium dodecyl sulfate-polyacrylamide gel electrophoresis for the separation of proteins in the range from 1 to 100 kDa. Anal Biochem. 166, 368–379. 16. Andreeva, A.Y., Krause, E., Mu ¨ ller, E., Blasig, I.E. & Utepber- genov, D.I. (2001) Protein kinase C regulates the phosphorylation and cellular localization of occludin. J. Biol. Chem 276, 38480– 38486. 17. Gill, S.C. & von Hippel, P.H. (1989) Calculation of protein extinction coefficients from amino acid sequence data. Anal. Bio- chem. 182, 319–326. 18. Johnson, W.C. Jr (1990) Protein secondary structure and circular dichroism: a practical guide. Proteins Struct. Funct Genet. 7, 205– 214. Ó FEBS 2002 Characterization of NblA (Eur. J. Biochem. 269) 4623 19. Zamyatnin, A.A. (1972) Protein volume in solution. Progr. Bio- phys. Mol Biol. 24, 109–123. 20. Zamyatnin, A.A. (1984) Amino acid, peptide, and protein volume in solution. Annu. Rev. Biophys. Bioeng 13, 145–116. 21. Laue, T.M., Shah, B.D., Ridgeway, T.M. & Pelletier, S.L. (1992) Computer-aided interpretation of analytical sedimentation data for proteins. In Analytical Ultracentrifugation in Biochemistry and Polymer Science (Harding, S.E., Rowe, A.J. & Horton, J.C., eds), pp. 90–125. Royal Society of Chemistry, Cambridge. 22. Sober, H. (1968) The Handbook of Biochemistry and Molecular Biology, Chemical Rubber Co., Cleveland, OH. 23. Lide, D.R., ed. (1974) Handbook of Chemistry and Physics.CRC Press, Boca Raton, Florida. 24. Lamm, O. (1929) Die Differentialgleichung der Ultrazentrifugie- rung. Ark Mat Astr Fys 21B, 1–4. 25. Behlke, J. & Ristau, O. (1997) Molecular mass determination by sedimentation velocity experiments and direct fitting of the con- centration profiles. Biophys. J. 72, 428–434. 26. Behlke, J. & Ristau, O. (2002) A new approximate whole boundary solution of the Lamm differential equation for the analysis of sedimentation velocity experiments. Biophys. Chem 95, 59–68. 27. Schuck, P. (1998) Sedimentation Analysis of noninteracting and self-associating solutes using numerical solutions to the Lamm equation. Biophys. J. 75, 1503–1512. 28. Schuck, P., Perugini, M.A., Gonzales, N.R., Howlett, G.J. & Schubert, D. (2002) Size-distribution analysis of proteins by ana- lytical ultracentrifugation: strategies and application to model systems. Biophys. J. 82, 1096–1111. 29. Van Holde, V.E. & Weischet, W.O. (1978) Boundary analysis of Sedimentation velocity experiments with monodisperse and pau- cidisperse solutes. Biopolymers 17, 1387–1403. 30. Demeler, B., Saber, H. & Hansen, J.C. (1997) Identification and interpretation of complexity in sedimentation velocity boundaries. Biophys. J. 72, 397–407. 31. Chatelier, R.C. (1988) A Parameterized overspeeding method for the rapid attainment of low-speed sedimentation equilibrium. Anal. Biochem. 175, 114–119. 32. Chervenka, C.H. (1969) A manual of methods for the analytical ultracentrifuge. Beckman Instruments, Inc., Palo Alto, CA. 33. Johnson, M.L., Correia, J.J., Yphantis, D.A. & Halvorson, H.R. (1981) Analysis of data from the analytical ultracentrifuge by nonlinear least squares techniques. Biophys. J. 36, 575–588. 34. McRorie, D.K. & Voelker, P.J. (1993) Self-associating systems in the analytical ultracentrifuge. Beckman Instruments, Inc., Full- erton, CA. 35. Rost, B. & Sander, C. (1993) Prediction of protein structure at better than 70% accuracy. J. Mol Biol. 232, 584–599. 36. Rost, B. (1996) PHD: predicting one-dimensional structure by profile based neural networks. Methods Enzymol 266, 525–539. 37. Jaenicke, R. (1999) Stability and folding of domain proteins. Prog. Biophys. Mol Biol. 71, 155–241. 38. Sidler, W.A. (1994) Phycobilisome and phycobiliprotein struc- tures. In The Molecular Biology of Cyanobacteria (Bryant, D.A., ed.), pp. 139–216. Kluwer, Dordrecht. 4624 H. Strauss et al. (Eur. J. Biochem. 269) Ó FEBS 2002 . NblA from Anabaena sp. PCC 7120 is a mostly a- helical protein undergoing reversible trimerization in solution Holger Strauss 1 , Rolf Misselwitz 2 ,. sp. PCC 6803) or purified in recombinant form from E. coli (NblA1 and NblA2 from Synechocystis sp. PCC 6803 and NblA protein from Anabaena sp. PCC 7120