Identification of secreted proteins of the cyanobacterium Synechocystis sp. strain PCC 6803

Abstract

We investigated the spectrum of secreted proteins in the cyanobacterium Synechocystis, and identified these proteins by amino-terminal sequencing. In total, seven sequences have been determined that corresponded to the proteins Sll0044, Sll1694, Sll1891, Slr0924, Slr0841, Slr0168, and Slr1855. The protein Sll1694 of 18 kDa that formed one of two major bands on SDS–PAGE was identified as cyanobacterial pilin, PilA. The amino-terminal sequence of another protein that formed a second major band was blocked. The analysis of the data revealed that five of seven proteins had distinct putative leader sequences for secretion.

Cyanobacterium, Leader sequence, Secreted protein

1 Introduction

Cyanobacterial strains secrete various polypeptides into the culturing media. These proteins are extracellular nucleases, proteases, and structural components of peptidoglycan layers [1–5]. Some of such polypeptides have medical and pharmacological importance [5,6].

The cyanobacterium Synechocystis sp. PCC 6803 has been serving as a useful model system to study photosynthesis, stress responses, and signal transduction pathways [7–9]. The complete genome of this organism has been sequenced and the annotated information is available at the Cyanobase site via internet [10].

Up to now nothing has been known about the spectrum of extracellular proteins of this strain of cyanobacterium. We took the advantage to determine the amino-terminal sequences of proteins found in the culturing medium of Synechocystis, and to identify their complete and leader sequences.

2 Materials and methods

2.1 Culture conditions

Synechocystis sp. PCC 6803 was obtained from Dr. Williams (Du Pont de Nemours and Company, Inc., Wilmington, DE, USA). Cells were grown at 34°C in BG11 medium [11] buffered with 25 mM HEPES-NaOH pH 7.5 under continuous illumination as described previously [12]. Cells were grown until they reached OD₇₉₀ 0.5, 1 or 2, as indicated.

2.2 Isolation of proteins from the culture medium

Cells were collected by centrifugation of 500 ml of suspension at 10,000×g for 20 min at 20°C. Supernatant was collected, centrifuged again, and filtered through a 0.22 micron Stericup filter (Millipore). Proteins were precipitated from the medium by ammonium sulfate at 80% saturation on ice during 1–2 h, and centrifuged for 40 min at 10 000×g at 4°C. Pellets were resuspended in 1–2 ml of 0.5 M phosphate buffer and dialyzed against the same buffer for 6–12 h at 4°C using Slyde-A-Lyzer (10 kDa cut off) units (Pierce, USA). Dialyzed proteins were concentrated in Centricon YM-10 units (Amicon, USA).

2.3 Protein separation, blotting and identification

Proteins were separated on 12% SDS–PAGE at room temperature and blotted onto PVDF membrane Immobilon-P (Millipore) using a semi-dry transfer unit. Proteins were visualized by staining the membrane with 0.1% Coomassie brilliant blue (CBB) in 50% methanol for 5 min at room temperature, and the location of the bands was marked. The membrane was then destained with a solution of 50% methanol+10% acetic acid, and dried in air, whereafter destained bands were cut out and stored at −20°C in Eppendorf tubes. Amino-terminal sequences were determined by the Edman degradation method [13] using an automated protein sequencing machine (model 477A, Applied Biosystems, USA).

3 Results and discussion

The proteins from the culture medium of Synechocystis were concentrated and resolved in SDS–PAGE (Fig. 1). Two major bands were observed on the gel irrespective of the stage of culture growth. The absence of blue-colored phycocyanins in the medium and on the gels allows to assume that no contaminating proteins from broken cyanobacterial cells were present in our preparations of extracellular proteins.

1

Extracellular proteins of Synechocystis sp. PCC 6803. Proteins were concentrated from the culture media collected when cells reached OD₇₉₀ 0.5 (1), 1 (2), and 2 (3) and resolved in 12% SDS–PAGE. Arrows indicate the bands that were subjected to N-terminal sequencing. From bottom to top – Sll0044, Sll1694 (PilA), Sll1891, Tic22, unidentified protein with a blocked N-terminus, Sll0841, Slr1855, Slr0168. MWM – molecular mass marker.

The resolved proteins were transferred onto PVDF membrane, stained with CBB, and eight visible bands of different intensities were cut out and used for N-terminal sequencing (Fig. 1). Among the proteins that were subjected to the analysis, one of 35 kDa could not be sequenced, probably, because its N-terminus was blocked. The sequences of seven other proteins have been successfully determined. These proteins and their known and putative functions are listed in Table 1.

1

Some characteristics of the identified extracellular proteins of Synechocystis

	Amino acid sequence	Protein	Function	AA	MW (kDa)
1	37PIALVVKILPH	Sll0044	Chemotaxis?	161	18
2	22FTLIELLVVVIIIGVLAAIALP	Sll1694	Pilin (PilA)	168	18
3	44LKFPQAGDRGTTQSSGGGGTR	Sll1891	Unknown	252	26
4	77PTEEVVKILQG	Slr0924	Tic22	286	31
5	37METSKSQSLQGTTWYISGWPS	Slr0841	Unknown	291	32
6	2TIQYTPLADRLLAYLAADRLNL	Slr0168	Glycoprotein?	643	68
7	1MSPKINFPFSDLIAGYVTSYDQ	Slr1855	Porin?	607	70
The determined amino acid sequences of the proteins are shown. Numbers preceding the sequences indicate positions of amino acid in the whole protein as referred to translations from the corresponding nucleotide sequences. Putative functions are depicted by homology. AA – number of amino acids in the complete sequence. MW – calculated molecular mass of the whole polypeptide. Modified phenylalanine of the mature PilA is underlined.

The protein that formed a major band of about 18 kDa was identified as pilin, PilA (sll1694). The first sequenced amino acid of this polypeptide was modified. Based on sequence analysis of the Synechocystis genome database we found that this amino acid is phenylalanine. Pilins are typically synthesized as precursors with unique short basic amino-terminal leader peptides, and processed pilins have an unusual N-methylphenylalanine at the amino-terminus [14–17]. Therefore we assume that the first modified amino acid of the mature cyanobacterial PilA is methylphenylalanine. This observation implies that cells of Synechocystis might have a prepilin leader peptidase that is similar to the eubacterial one (PilD), which possesses two activities, of signal peptidase and methylase [16–18]. The homology search in Cyanobase [10] revealed an open reading frame Slr1120 which has up to 85% identity to the known bacterial prepilin leader peptidases.

The N-terminus of the mature PilA showed strong similarity to some other proteins of Synechocystis – Sll1695, Slr1456, Slr1929, Slr1930, Slr2015, Slr2016, and Slr2017. One can see that, except slr1456, all other genes for the proteins with similar N-termini form gene clusters in the Synechocystis genome. The pilA and sll1695 are organized as tandemly repeated genes, and the amino acid sequence of Sll1695 resembles the sequence of pilin PilA, although having a rather different sequence of the putative leader peptide. Other proteins display a significant similarity to PilA only in short N-terminal hydrophobic regions that supposedly correspond to the N-termini of the putative mature proteins (Fig. 2).

2

A comparison of the partial amino acid sequences of the polypeptides of Synechocystis that show homology to PilA. Putative leader peptides followed by the homologous hydrophobic regions are underlined. An arrow shows the predicted sites of cleavage by the processing aminopeptidase.

The putative leader sequences of the above mentioned polypeptides demonstrate rather high variability (Fig. 2). The common feature of these sequences is that they are positively charged. This feature is characteristic for the leader peptides of bacterial pilins [16–18]. All these oligopeptides have glycine at the putative site of processing. This observation suggests that these proteins could be secreted into the culture medium by Synechocystis similarly to PilA. However, we could not detect and identify any of the PilA-homologous proteins in the culture media.

Another protein with a similar calculated molecular mass (18 kDa) was identified as Sll0044, a protein which is similar to that involved in chemotaxis. The corresponding gene is located in the cluster of genes sll0041, sll0042 and sll0043. Sll0041 (CheD or PilJ) and Sll0042 (CheM) are homologous to bacterial methyl-accepting chemotaxis proteins I and II [19,20]. Sll0043 is homologous to chemotaxis regulatory protein CheA [21] and represents a putative signal transduction histidine kinase Hik18.

The functions of Sll1891 and Slr0841 are unknown. Sll1891 is similar to the translated polypeptide of sll0645, which forms a cluster with slr0559. The latter encodes a periplasmic protein of an ABC transporter for amino acids.

Slr0924 corresponds to a plant chloroplast protein Tic22 that is involved in protein import into thylakoid membranes [22]. The subcellular localization of some cyanobacterial analogs of chloroplast protein importers, located at the outer membrane of the organelles (Toc complex), revealed that the cyanobacterial polypeptides are located at the surface of the cell [23]. Tic22 belongs to the Tic complex that is a translocator of the inner membrane of higher plant chloroplasts. However, Slr0924 was found among the secreted proteins of Synechocystis. It could imply that the function of this protein may differ from that of the analog in higher plants, and that Slr0924 could be located at the surface of the cyanobacterial cells.

N-terminal sequencing of two proteins, Slr0168 and Slr1855, showed that they do not have putative leader peptides. Slr0168 resembles a glycoprotein, and the corresponding gene is part of a cluster consisting of six genes with unknown function. Slr1855 shows some similarity to bacterial porins (Table 1), and the corresponding gene is a terminal member of a cluster consisting of slr1852, slr1853, and slr1854. The function of slr1852 is unknown, slr1853 encodes a carboxymuconolactone decarboxylase, and slr1854 encodes a putative intracellular protease.

Putative leader sequences of the identified proteins are shown in Fig. 3. Their length ranges from 21 to 76 amino acids, and they do not exhibit any significant similarity between each other. Moreover, these sequences do not show any similarity to other known proteins as determined by Blast search. It is possible that cyanobacterial leader sequences are quite different from those of other bacteria. The observation that the leader sequences of the PilA family of proteins in Synechocystis differ extensively allows us to suggest that the amino acid sequence of the leaders is not essential for secretion. It is possible that the overall charge of the leader sequence is the major factor that determines the ability of the protein for secretion. It was suggested that a positive charge of leaders is essential for the electrostatic interaction of the precursors of the secreted proteins with the membrane before they pass through [17]. Although the mechanisms of protein secretion in cyanobacteria are unknown, it could be that the positive charge of leader peptides facilitates interactions of these proteins with the major component of the secretion pathway SecA, as found in Escherichia coli [18]. Homologues of members of the SecA secretion machinery are present in the genome of Synechocystis.

3

Amino acid sequences of the putative leader peptides of the identified extracellular proteins of Synechocystis. Arrows show the suggested sites of processing. The first amino acids after the arrows correspond to the first amino acids of the experimentally identified N-terminal sequences.

Acknowledgements

This work was supported in part by a Grant from ASGL-Research Laboratories, St. Petersburg, Russia (23-1/SNT-99).