Characterizing the Host and Symbiont Proteomes in the Association between the Bobtail Squid, Euprymna scolopes, and the Bacterium, Vibrio fischeri

The beneficial symbiosis between the Hawaiian bobtail squid, Euprymna scolopes, and the bioluminescent bacterium, Vibrio fischeri, provides a unique opportunity to study host/microbe interactions within a natural microenvironment. Colonization of the squid light organ by V. fischeri begins a lifelong association with a regulated daily rhythm. Each morning the host expels an exudate from the light organ consisting of 95% of the symbiont population in addition to host hemocytes and shed epithelial cells. We analyzed the host and symbiont proteomes of adult squid exudate and surrounding light organ epithelial tissue using 1D- and 2D-polyacrylamide gel electrophoresis and multidimensional protein identification technology (MudPIT) in an effort to understand the contribution of both partners to the maintenance of this association. These proteomic analyses putatively identified 1581 unique proteins, 870 proteins originating from the symbiont and 711 from the host. Identified host proteins indicate a role of the innate immune system and reactive oxygen species (ROS) in regulating the symbiosis. Symbiont proteins detected enhance our understanding of the role of quorum sensing, two-component signaling, motility, and detoxification of ROS and reactive nitrogen species (RNS) inside the light organ. This study offers the first proteomic analysis of the symbiotic microenvironment of the adult light organ and provides the identification of proteins important to the regulation of this beneficial association.


Introduction
The light organ symbiosis between the Hawaiian bobtail squid, Euprymna scolopes, and the bioluminescent bacterium, Vibrio fischeri, is used as a model association for understanding host/microbe interactions [1][2][3]. Hours after hatching from its egg case, the host is colonized when environmental V. fischeri take up residence in epithelia-lined crypt spaces located within a specialized light organ [1]. V. fischeri is the sole bacterium that colonizes the light organ and prior research has focused on understanding the mechanisms for establishing and maintaining the high degree of specificity between the partners [1][2][3][4]. While in the light organ, the bacteria are connected directly to the external environment through ciliated ducts and pores (Fig. 1). This conduit is important as it serves as an interface between the host and the environment and is used in a daily venting of the symbionts. The venting behavior is linked to the nocturnal foraging activities of the host. At night the light organ crypt spaces contain the highest densities of bacteria (10 9 /adult squid; [5]), and the light provided by these symbionts is used to avoid predation [6]. At dawn the host expels 95% of its symbionts from the light organ, while entering a quiescent state in which it buries in the substrate [5,7]. The remaining bacteria repopulate the crypts ensuring a full complement of symbionts by the following nightfall. This venting mechanism helps regulate the symbiont population in the light organ as well as increases the concentration of V. fischeri in the immediate squid habitat, allowing future generations to be colonized [1,8].
The exudate of adult hosts emerges from the light organ pores as a thick paste-like substance that can be easily collected for experimental analyses (Fig. 1). This material represents the immediate microenvironment of the light organ crypts and is comprised of symbiont cells and a mixed population of host cells (macrophage-like hemoctyes and shed epithelial cells), all surrounded by an acellular matrix [5]. In order to understand the host and symbiont contributions to this microenvironment, previous studies have focused on the cellular and biochemical components of the exudate [5,9]. Recent work has focused on changes in host and symbiont gene expression during the daily rhythm within the light organ [10]. Transcriptome analyses at different time points during the day/night revealed dynamic changes both metabolically and physiologically for the host and symbiont, and identified a large number of differentially expressed genes [10]. In addition, microscopy at these time points revealed that the crypt epithelium also undergoes morphological changes whereby apical surfaces are blebbed into the crypt spaces [10]. Many of these gene expression and cellular changes were most dramatic in the hours just before and after dawn, reflecting the dynamic turnover that occurs in the light organ upon venting.
In this study, we employed a number of techniques to characterize the host and symbiont proteomes of the adult light organ microenvironment at dawn when the association undergoes a dramatic reduction in symbiont population. To date, proteomic analyses of the squid/Vibrio association are limited. A previous study used two-dimensional polyacrylamide gel electrophoresis (2D-PAGE) to reveal numerous differences in the soluble proteins present in the light organs of juvenile aposymbiotic (uncolonized) and symbiotic (colonized) squid during the development of the symbiosis, however no proteins were identified [11]. Recent advances in proteomics, including multidimensional protein identification technology (MudPIT), have provided the tools to allow the identification of a large number of host and symbiont proteins in the squid/Vibrio association for the first time [12,13]. MudPIT utilizes strong cation exchange chromatography (SCX) to separate peptides by charge prior to liquid chromatography tandem mass spectrometry (LC MS/MS), thus increasing the number of identified peptides. In this study we utilized MudPIT, in addition to 1D-and 2D-PAGE, to describe both the host and symbiont proteomes in the light organ exudate and the surrounding host epithelial tissue. These analyses identified components of the host's innate immune system as well as numerous proteins involved in the detoxification of reactive oxygen species (ROS). Symbiont proteins detected were involved in stress responses, quorum sensing, motility, and two-component signaling pathways. Our data also highlight many proteins that are presently uncharacterized with regard to the squid/Vibrio symbiosis. Identifying the host and symbiont proteins present in the light organ represents a first step to understanding key functional aspects of the association's molecular dialogue that is responsible for maintaining this highly specific relationship and complements a number of other molecular and genetic techniques that have been applied to this symbiosis.

Ethics statement
Euprymna scolopes is an invertebrate and is not regulated by animal care regulations in the United States. All field collection of research animals was done in accordance with state and federal regulations. The State of Hawaii does not require collection permits for this species outside of marine reserves. None of the animals collected for this study were caught/collected within a marine reserve or regulated area.

General methods
Adult animals were collected in shallow sand flats of Maunalua Bay, Oahu, HI by dip net and were either maintained in the laboratory in re-circulating natural seawater aquaria at the Hawaii Institute of Marine Biology or at the University of Connecticut with artificial seawater (ASW, Instant ocean) at 23uC. All animals were acclimated at least 48 hours under laboratory conditions and kept on an approximate 12 hr light/12 hr dark cycle before sample collection. V. fischeri strain ES114 was grown in saltwater tryptone (SWT) at 28uC as previously described [14].

Exudate and central core collection
Exudate was collected as previously described [5]. Briefly, adult squid were anesthetized in a 2% ethanol/seawater solution and ventrally dissected under red light within minutes prior to dawn. A light stimulus (150 W halogen light) was used to induce venting behavior. Within 1 h, the squid had expelled the light organ contents, which were collected with a 10-ml disposable micropipette (Drummond Scientific Company) and stored on ice after the addition of a 16 protease inhibitor cocktail according to the manufacturer's protocol (Sigma Aldrich, P2714). Post-vented central cores were also dissected and removed from the light organ. All samples were flash frozen with liquid nitrogen and stored at 280uC until further analysis. No differences were detected between samples collected from animals maintained at either the Hawaii Institute of Marine Biology or at the University of Connecticut (data not shown).

Gel-based proteomic methods
Exudate sample preparation for 1D-and 2Dpolyacrylamide gel electrophoresis. For PAGE applications, symbiont cells from freshly collected light organ exudate were separated from the soluble fraction, a source of host proteins, by centrifugation (Eppendorf 5810 R, 5,000 rpm, 10 minutes, 4uC). The symbiont pellet was washed three times with 0.22 mm filtered ASW to remove additional soluble proteins. Symbiont proteins were extracted by a modified method from Ho and Hsu [15]. Briefly, 10 consecutive liquid nitrogen freeze/thaw cycles were performed in the presence of a 16 protease inhibitor cocktail (Sigma Aldrich, P2714) with 80 mM Tris, pH 8.0 for cell lysis. After separation from the bacterial pellet, soluble host proteins were quantified (see below) and stored until further analysis. For cultured V. fischeri, cells were grown to early stationary phase [14] and proteins were extracted as described for the symbiont exudate pellet. Protein concentrations of separate symbiont exudate and host soluble fractions, as well as culture-grown V. fischeri, were determined spectrophotometrically using the method of Whitaker and Granum [16] and/or a Bradford assay (Bio-Rad). Typically, protein extractions of exudate resulted in 10-20 mg of protein combined from the host soluble and symbiont pellet fractions. Comparison of 2D-PAGE gels from soluble proteins of culture-grown V. fischeri and the host soluble fraction of the exudate demonstrated that the soluble host fraction was devoid of bacterial proteins (data not shown).
1D-polyacrylamide gel electrophoresis of light organ exudate. Between 10 and 20 mg of exudate protein from either the host (soluble protein separated from bacterial pellet) or symbiont fraction (bacterial pellet) were resolved with 12.5% polyacrylamide gels (Bio-Rad). Electrophoresis was performed with a Hoeffer 250 mini-gel apparatus at 23 mA or a Bio-Rad Mini PROTEANH Tetra cell at 200 V for 30 minutes. Gels were either stained with Bio-Rad Brilliant Blue Coomassie R-250 or a Bio-Rad Silver Stain Plus Kit (Bio-Rad). 1D-PAGE of both the soluble host fractions and the bacterial pellets were shown to be reproducible (n = 3, separate and fractionated pooled exudate samples for each; data not shown).
2D-polyacrylamide gel electrophoresis of light organ exudate. 2-D PAGE was performed using the Amersham Pharmacia Biotech Multiphor II system as previously described [11]. 40 mg of pooled exudate protein from either the soluble host fraction or the bacterial pellet, originating from 2 or more adult squid or culture-grown V. fischeri cells, were denatured 1:4 in 9 M urea, 1% DTT, 2% Pharmalyte 3-10, 0.5% Triton-X-100, 0.14% phenylmethylsulfonyl fluoride, loaded onto a first dimension gel strip with an immobilized pH gradient (4-7) and focused over a 20 hour period. Samples were then separated by molecular weight on pre-cast 12% to 14% polyacrylamide gradient gels (GE Healthcare Life Sciences). Gels were silver stained as previously described [11,17]. 2D-PAGE from the soluble host fractions and the bacterial pellets or culture-grown cells were deemed to be highly reproducible (n = 3, separate and fractionated pooled exudate samples; data not shown). For comparison, 2D gels were visually aligned and similarities and differences of the molecular weights and individual protein species were noted. Five spots of interest from the 2D-PAGE gel of the exudate bacterial pellet were excised and successively washed in 50% acetonitrile, 50% acetonitrile/50 mM NH 4 HCO 3 , and 50% acetonitrile/ 10 mM NH 4 HCO 3 . The five gel spots were then dried by speed vacuum (Eppendorf Concentrator 5301) and resuspended in 10 mM NH 4

Mass spectrometry proteomic methods
Protein preparation for multidimensional protein identification technology and liquid chromatography tandem mass spectrometry. For MudPIT and LC MS/MS, pooled host and symbiont fractions from freshly collected light organ exudate were combined and quantified as described above. Additionally, central cores were homogenized in the presence of a 16 protease inhibitor cocktail (Sigma Aldrich, P2714) with 80 mM Tris, pH 8.0 using a ground-glass homogenizer. Proteins from central cores were collected from the supernatant of the homogenate after centrifugation (Eppendorf 5810 R, 14,000 rpm, 30 minutes, 4uC) and quantified as described above. Extractions of the central core tissue resulted in approximately 20 mg of soluble protein per central core. Total protein from pooled exudate samples (50 mg, n = 7 squid and 100 mg, n = 7 squid) and pooled central core samples (40 mg, n = 3 squid) were precipitated with 10% trichloroacetic acid (Fisher Scientific) at 4uC overnight. The protein precipitates of the exudates and central cores were collected by centrifugation (Eppendorf 5810 R, 11,0006 g, 30 minutes, 4uC) and washed twice with ice-cold acetone. The protein pellets were briefly air-dried and then solubilized in 25  For SCX, the tryptic digests of pooled exudate proteins were acidified with 2 ml of 1 M phosphoric acid. A 2.1 mm6200 mm PolySULFOETHYL A TM column (PolyLC Inc.) was used to establish a linear gradient for 118 minutes. The gradient was maintained in 10 mM potassium phosphate, 25% acetonitrile (pH 3.0) and the same buffer with the addition of 1 M potassium chloride. Fractions were collected every 2 minutes at a flow rate of 150 ml/min. All fractions were dried, dissolved in 5 ml of 70% formic acid, and diluted to 15 ml in 0.1% trifluoroacetic acid for subsequent LC MS/MS.
Liquid chromatography tandem mass spectrometry. LC MS/MS of each exudate SCX fraction, central core peptides, and 2D gel spot peptides was completed at the W. M. Keck Biotechnology Resource Laboratory at Yale University. A LTQ Orbitrap mass spectrometer (Thermo Fisher Scientific) equipped with a Waters nanoAcquity UPLC system operated with a Waters SymmetryH C18 180 mm620 mm trap column, and a 1.7 mm, 75 mm6250 mm nanoAcquity TM UPLC TM column (35uC) was used for peptide separation. Trapping was performed at 15 ml/min with Buffer A (100% water, 0.1% formic acid) for 1 minute. Peptide separation was performed at 300 nl/min with Buffer A and Buffer B (100% CH 3 CN, 0.075% formic acid); a 51 minute linear gradient was established starting with 5% Buffer B, increasing to 50% B at 50 minutes, and finally 85% B at 51 minutes. MS was acquired in the Orbitrap using 1 microscan followed by four data dependent MS/MS acquisitions. Neutral loss scans (MS 3 ) were also obtained for 98.0, 49.0, and 32.7 amu.
Data analysis. All MS/MS spectra were analyzed using the Mascot algorithm for uninterpreted MS/MS spectra [18]. The Mascot Distiller program used the MS/MS spectra to generate Mascot compatible files by combining sequential MS/MS scans from profile data that have the same precursor ion. A charge state of +2 and +3 were preferentially located with a signal to noise ratio of 1.2 or greater and a peak list was created for database searching. The peak list was searched by Mascot using V. fischeri amino acid sequence and juvenile E. scolopes light organ expressed sequence tag (EST) databases [19]. Search parameters included partial methionine oxidation, carboxamidomethylated cysteine, a peptide tolerance of 620 ppm, MS/MS fragment tolerance of 60.6 Daltons (Da), and peptide charges of +2 or +3. Normal and decoy database were also searched. Mascot significance scores are based on a MOlecular Weight SEarch (MOWSE) scores and rely on multiple matches of more than one peptide to the same protein [20]. The MOWSE based ions score is equal to (210)*(Log 10 P), where P is the absolute probability that a match is random. For a match to be significant, the probability of it being a random match should be below 5% (E-value,0.05) [21]. The protein threshold score depends on the size of the database being searched, therefore, Mascot determined that scores greater than 68 were significant when searching the juvenile light organ EST database and scores greater than 48 were significant when searching the V. fischeri ES114 amino acid database. Proteins were considered identified when 2 or more peptides matched the same protein and if the Mascot score was above the respective significance threshold. Proteins with putative identifications contained two or more peptide matches, but had a Mascot score below the threshold for the respective database (E-value.0.05).
Mascot also calculates the exponentially modified protein abundance index (empai) which estimates the abundance of protein species by using the number of peptides detected in the analysis compared to the number of possible peptides for a particular protein [22,23].
Host proteins identified by Mascot using the juvenile light organ EST database were further analyzed using the Bioinformatics Utility for Data Analysis of Proteomics using ESTs (BUDAPEST) which removed any peptides matching to non-coding reading frames [24]. BLASTx (E-value cutoff 10 26 ) against the NCBI nr database was used to determine the top protein hit for each EST [25]. In addition, BUDAPEST calculated a peptide score for each protein identified. This score was equal to the number of correct reading frame peptides squared divided by the total number of peptides (all reading frames) identified for that EST. BUDAPEST scores greater than 1 can be considered significant, however, in our study scores of 2 or greater were chosen to represent significant protein identifications.

Results
Exudate samples collected from adult E. scolopes light organs were analyzed using a number of proteomic techniques. 1D-and 2D-PAGE revealed that the host soluble fraction of the exudate, derived from host hemocytes and apical surfaces of shed light organ crypt epithelial cells, was comprised of a complex mixture of proteins and peptides, the majority of which are represented between the isoelectric points of 4 to 7 and a size of 7 to 100 kilodaltons (kD) ( Fig. 2A, B). Similar analyses of the symbiont fraction of the exudate also revealed a complex protein profile (Fig. 3A, B). Comparison of the host and symbiont PAGE gels support previous observations that the exudate appears enriched in bacteria. When comparing proteins expressed by V. fischeri in the light organ to proteins expressed by V. fischeri in culture, a protein with an isoelectric point of 5.5 and a molecular weight of 10 kD was present in the light organ, but not in solubilized proteins from culture-grown V. fischeri (Fig. 3C). The protein spot of interest (Fig. 3C, spot 2) and four surrounding protein spots (common to both the light organ and culture) were excised and identified by LC MS/MS ( Table 1). The unique symbiont light organ protein was determined to be a quorum sensing-regulated protein (QsrP), which has been previously identified as being expressed by V. fischeri in the light organ, but remains functionally uncharacterized [26].
In an effort to further characterize the proteins expressed by the host and symbiont we utilized shotgun proteomic techniques (LC MS/MS and MudPIT). These methods allowed us to putatively identify a combined 1581 host and symbiont proteins present in the light organ. For MudPIT, light organ exudate samples of 10 or 20 SCX peptide fractions (see Materials and Methods) were analyzed ( Table 2). In addition, to increase our representation of host proteins we analyzed post-vented central cores by single fraction LC MS/MS (Table 2). A total of 870 unique symbiont proteins were putatively identified by Mascot from all the light organ samples (exudates and central cores; Table S1). 516 of these proteins were above the significance threshold set by Mascot (Evalue,0.05). For the host, we utilized BUDAPEST, a software program developed specifically to identify proteins in the correct open reading frame in cases when only EST databases are available [24]. 676 host proteins with more than 2 peptides matching to the correct reading frame and a BUDAPEST score of greater than or equal to 2 were identified from combining the LC MS/MS and MudPIT data of the exudate and central core samples (Table S2).
All host and symbiont proteins, including putative identifications, were organized functionally according to the Clusters of Orthologus Groups database (COG and KOG; Fig. S1, Table S1,  Table S2) [27,28]. In order to achieve a more thorough understanding of the functions represented by the proteins in our data, we first analyzed the relative abundance of each symbiont protein. The 25 most abundant symbiont proteins determined by empai include the protein subunits of luciferase (LuxAB), QsrP, alkyl hydroperoxide reductase C22 (AhpC), and several cold shock proteins (Table 3). Our analyses also identified a number of symbiont proteins related to functions involved in stress responses, quorum sensing, motility, and signaling pathways, all of which have been previously implicated as being important in the squid/Vibrio association ( Table 4, Table S3; see discussion). Several of these identified proteins including AhpC and the cold shock proteins have symbiotic roles yet to be characterized.
Host proteins detected in the light organ highlight the innate immune system, oxidative stress, and signaling pathways (Table 5). Identified proteins include those involved with the NFkB signaling pathway and the recognition of microbial associated molecular patterns (MAMPs) such as peptidoglycan recognition proteins (PGRPs) and a carbohydrate binding protein, galectin (Table 5). Proteins related to oxidative stress consist of superoxide dismutase, peroxiredoxins and numerous peroxidases, including the E. scolopes halide peroxidase (EsHPO) ( Table 5; see discussion). Additionally, several host proteins involved iron-sequestration were detected in the light organ.

Discussion
The daily expulsion of V. fischeri from the light organ of E. scolopes provides a unique opportunity to characterize the interactions between the host and symbiont in a natural microenvironment. Previous analyses of this exudate have focused on the cellular and biochemical composition of the expelled matrix [5,9]. In this study we characterized the light organ exudate and surrounding epithelial proteome using MudPIT and PAGE. A total of 1581 unique host and symbiont proteins were putatively identified from the light organ, offering the first proteomic analyses of this symbiotic microenvironment.

Innate immune system
MAMPs and host pattern recognition receptors (PRRs) are two components underlying host/microbe interactions and are significantly involved in the development of this association [29]. MAMPs including lipopolysaccharide (LPS), and peptidoglycan and its derivatives, function in determining the specificity of the squid/Vibrio symbiosis as well as initiating morphogenetic changes to the light organ [30][31][32]. We identified several host proteins related to pattern recognition in both the exudate and central core tissues (Table 5). E. scolopes PGRP2 and 3 (EsPGRP2 and EsPGRP3) are involved in detecting peptidoglycan, a major cell wall component of bacteria [33]. EsPGRP2 is secreted into the crypts of the light organ where it is thought to degrade tracheal cytotoxin (TCT), a monomer of peptidoglycan [34]. The role of EsPGRP3 in the symbiosis is currently under investigation, but has been detected in adult and juvenile hemocytes (unpublished data). Certain carbohydrates, such as beta-galactosides, are another type of MAMP that are recognized by carbohydrate binding proteins known as galectins [35]. A putative galectin was identified in both the exudate and central core tissue (Table 5, Table S2) and may have an uncharacterized role in the squid/Vibrio symbiosis.
Aside from PRRs and MAMPs, cellular adhesion is often important for host/microbe cell-to-cell interactions. Outer mem-   brane proteins (OMPs) are localized at the bacterial cell surface and are good candidates for mediating recognition between the partners. OmpU, a symbiont outer membrane protein that we have identified in the light organ (Table S3), was shown to be important in mediating adhesion to adult host hemocytes and during the early stages of colonization [36,37]. Other OMPs identified, such as a hypothetical protein VF_1010, have roles yet to be characterized in binding and adhesion, but may have similar functions (Table S3). Understanding how the symbiont outer membrane proteome varies in the light organ vs. the free-living environment and between symbiosis-competent and incompetent strains may shed light on mechanisms of mediating specificity in this symbiosis. An immune pathway highlighted by our proteomic data includes NFkB signaling ( Table 5). The role of NFkB signaling during the establishment of the squid/Vibrio symbiosis is currently under investigation, however, many important members of the pathway have been identified from juvenile light organ ESTs [33]. We detected NFkB repressing factor (see below) and importin alpha 3, a protein involved in shuttling proteins into the nucleus by recognizing nuclear localization signals (Table 5) [38]. In vitro and in vivo studies using cancer cell lines reveal this protein is a member of the NFkB signaling pathway and aids in the transport of NFkB transcription factors into the nucleus [39].
Recently, E. scolopes has been shown to have a complement pathway that in other systems is involved with mediating inflammation and opsonization [29,40,41]. The function of this pathway has yet to be described in the squid/Vibrio symbiosis, however, we detected putative components of the complement cascade in both the exudate and the central core ( Table 5, Table  S2). Although one of these identifications was annotated as a complement component C3 precursor ( Table 5, Table S2), closest to the cnidarian Nematostella, further analysis of these peptides using E. scolopes transcriptomic data revealed that this protein did not align with the previously described E. scolopes C3 (data not shown). Instead, this protein, along with two others, were identified as thioester-containing proteins (TEPs). Among invertebrates, TEPs play an important role in innate immune response as members of the complement system or as protease inhibitors [42,43].

Reactive oxygen and nitrogen stress response
The chemical microenvironment of the light organ crypts likely influences the maintenance of the association and helps to ensure specificity. Although oxygen is critical for the bioluminescence reaction, reactive oxygen species (ROS) and toxic oxygen intermediates have been shown to be abundant in the light organ [44]. Host-derived ROS, such as hypohalous acid, are thought to play key roles in initiation and persistence of the squid/Vibrio  symbiosis [44]. Hypohalous acid, produced by an abundant light organ peroxidase similar to a halide peroxidase, is believed to help to create an oxidative environment that V. fischeri must overcome to colonize the host [45,46]. In addition to the previously described EsHPO, a number of other host peroxidases were present, suggesting that additional ROS may be important to this association (Table 5). Peroxiredoxins are antioxidant proteins, which are abundant in the host proteome and have been shown to detoxify reactive molecular species derived from oxygen and nitrogen [47,48]. Therefore, these ROS mediators may indicate a means by which the host protects its own tissues in the oxidative microenvironment of the light organ.
Another role of host ROS may be maintaining specificity by preventing non-symbiotic bacteria and potential pathogens from infecting the host. The light organ crypts are open to the environment via pores on the surface of the light organ, yet V. fischeri is thought to be the sole symbiont of this highly specific association [1]. Proteins expressed by the symbiont reveal functions involved with protecting cells from host ROS (Table 4). V. fischeri utilizes a periplasmic catalase (katA) to sequester hydrogen peroxide from the host, which can be used by EsHPO to generate hypohalous acid [49]. We identified, in addition to KatA, the antioxidant enzymes AhpC and thioredoxin-dependent thiol peroxidase (Bcp) ( Table 4). A V. fischeri katA mutant showed no additional catalase activity in culture suggesting that KatA is the major scavenger of H 2 O 2 [49]. The additional antioxidant proteins identified in this study may indicate a mechanism by which the symbiont can protect itself from other types of ROS or  RNS. AhpC, a peroxiredoxin capable of reducing hydrogen peroxide, organic peroxides, and peroxynitrite, is the most abundant antioxidant symbiont protein present in the light organ (Table 3). In Vibrio vulnificus, AhpC functions along with another subunit, AhpF, which supplies the reducing equivalents in the form of NADH, to reduce peroxides [50]. However, AhpF is absent from the V. fischeri genome, suggesting that another protein is necessary to reduce peroxides by this pathway. Studies involving Treponema pallidum show that thioredoxin reductase can substitute for organisms lacking an AhpF homolog [51]. For V. fischeri, a thioredoxin reductase FAD/NAD(P)-binding protein (TrxB) was present in our MudPIT data (Table 4) and may have the potential of interacting with AhpC. Along with AhpC, proteins implicating that V. fischeri also detoxifies RNS, include nitric oxide dioxygenase (Hmp), and two peptide-methionine (S)-S-oxide reductases (MsrA and VF_A0005; Table S1) [52,53].
Reactive nitrogen species, such as nitric oxide (NO), contribute to signaling and development in the squid/Vibrio symbiosis [54]. The role of NO as a toxic product to pathogens has been well studied; however, the function of NO in beneficial associations has only been recently analyzed [54,55]. In juvenile squid the epithelial tissue lining the ducts entering the light organ crypts contain high levels of NO, suggesting that the symbionts must overcome NO in order to colonize the light organ [56]. The detection of nitric oxide dioxygenase (Hmp), recently shown to play a role in NO detoxification, suggests that V. fischeri also maintains the ability to manage NO related stress in adult squid (Table 4) [53]. Once V. fischeri colonizes the light organ, nitric oxide synthase (NOS) is down-regulated and lower levels of NO likely allow the symbiont to grow in the crypt spaces under reduced RNS stress [56]. We identified NFkB repressing factor, which in addition to other immune functions, has been shown in vitro to negatively regulate transcription of NFkB pathway effectors, including NOS, by directly interacting with promoter region sequences (Table 5) [57,58]. The results of this study provide a number of new host and symbiont targets involved in mediating ROS and RNS for further analyses.
The availability of iron has also been shown to be an important factor in squid/Vibrio symbioses [53,59,60]. Free iron plays a critical role in host/microbe interactions and under certain circumstances may allow development of pathogenic associations [61,62]. Host proteins involved in sequestering free iron such as ferritin, transferrin, and melanotransferrin were identified (Table 5). These iron-binding proteins provide supporting evidence that iron remains limiting in the light organ and suggest a possible role for these proteins in regulating the growth of V. fischeri [59]. In contrast to the host, putative proteins that the symbiont may utilize for acquiring iron include receptors for the siderophores aerobactin and anguibactin (Table S3). Symbiont proteins involved in utilizing heme, another source of iron, are also present, and include HutZ, HutA, HuvX, and HmuT (Table S3). It is likely that V. fischeri employs several different strategies to meet its necessary iron requirements.

Quorum Sensing
First described in V. fischeri, quorum sensing regulates bioluminescence, the light from which provides the host with an anti-predatory mechanism known as counterillumination [6,63,64]. Lux proteins involved in light production were identified and among the most abundant symbiont proteins (Table 3). Previous PAGE and transcriptomic analyses first revealed additional quorum sensing-regulated proteins, which were also detected by our characterization of the adult light organ proteome [26,65]. QsrP is one of the most abundant proteins present in the symbiont proteome (Table 3), yet this novel protein remains functionally uncharacterized. Another quorum sensingregulated protein identified in this study is a putative surface protein (VF_A0894) with immunoglobulin-like domains (Table 4). This putative surface protein is similar to the Leptospira immunoglobulin-like proteins (LigA, LigB and LigC) of pathogenic Leptospira spp., which are thought to mediate adhesion to host cells [66]. These quorum sensing-regulated proteins may be important to a symbiotic lifestyle. We also detected LuxS, AI-2 synthase, which is involved in a second quorum sensing system in V. fischeri and has been implicated in regulating motility in Vibrio alginolyticus [67][68][69]. A link between LuxS and motility, may implicate a role for quorum sensing and the onset of motility prior to symbiont expulsion from the light organ (see below).

Symbiont Signaling
Two-component signaling pathways are important mechanisms by which bacteria can sense the environment and have been identified in V. fischeri [70][71][72][73]. The roles in colonization for some of these regulators, which were present in our proteomic data (Table 4), such as GacA and ArcA, have been studied in detail, and mutagenesis of these genes has demonstrated that they are important in the association [71][72][73]. Although many regulators have already been characterized with respect to the symbiosis, many proteins involved in two-component signaling have unknown functions in the light organ. For example, CpxP, an abundant symbiont protein (Table 4), is a periplasmic component of Escherichia coli and Vibrio cholerae and involved in modulating the cell envelope stress response through CpxAR signaling, thus providing an appealing target for future studies [74,75].

Other Related Stresses
Although several were identified in this study, cold shock proteins have yet to be described with respect to the light organ symbiosis. Of the top 25 most abundant symbiont proteins present in the light organ, three were cold shock proteins (CspD, CspG, and VF_A0595; Table 3). Cold shock proteins often bind nucleic acids and function in general stress responses. Furthermore, they have been shown to play a role in regulating bacterial growth at stationary phase and may even serve as MAMPs recognized by hosts [76,77]. One cold shock protein identified in the light organ, CspD, prevents replication from occurring in stationary phase E. coli cells by binding to single stranded DNA and blocking replication [78]. Prior to expulsion at dawn, the symbiont population is at its most dense during the day/night cycle. Therefore, cold shock proteins may play a role in either maintaining high cell densities in the light organ and/or assisting during the transition between the symbiotic and free-living state.

Motility
Research involving the role of motility in the squid/Vibrio symbiosis has focused on the initiation of colonization. Within the light organ V. fischeri cells become differentiated with the loss of their flagella [79]. Upon release from the light organ at dawn, V. fischeri cells are believed to fully regenerate their flagella within several hours [79]. Our proteomic data show the presence and putative identification of several proteins related to flagellar structure including filamental proteins (FlaA, FlaC), basal body proteins (FlgB, FlgH, FlgI), and a motor protein (MotB; Table 4). Proteins related to flagellar regulation (FlrA and FlgM) and chemotaxis (CheW and CheZ) were also detected. A recent study indicated an increase in flagellar gene expression by light organ symbionts in the hours preceding dawn and V. fischeri mutants of FlaA and FlrA have been shown to be important for symbiotic competence [10,80,81]. FlrA was also found to be expressed by V. fischeri in the light organs of E. scolopes and a different squid species, E. tasmanica, but not in strains grown in seawater [82]. Together, the data from this present study and others suggests that V. fischeri cells are generating flagella prior to expulsion from the light organ and may be preparing for the transition from the symbiotic to the free-living state. Future studies should focus on signals in the changing microenvironment that may initiate this transition.

Symbiont Metabolism
Within the light organ, V. fischeri employs a number of metabolic strategies [9,10,83,84]. The daily rhythm of the light organ symbiont population coincides with fluctuations in symbiont metabolism [10]. Transcriptomics revealed a unique pattern in which during the night the symbiont ferments chitin as a means of obtaining energy. After the majority of the symbiont population is expelled from the light organ, the remaining symbionts anaerobically respire glycerol during the hours in which the light organ becomes replenished with a full symbiont population. The results of this study show abundant symbiont chitin binding proteins and chitinases, thus supporting these previous findings (Table S3). The diel shift in metabolism is one piece of evidence that supports the light organ as being a dynamic microenvironment that is under the regulation of both the host and symbiont [10].

Summary
Proteomic studies of symbioses utilizing high-throughput techniques are becoming more common and have been used for analyses of the pea aphid-Buchnera symbiosis, nitrogen fixing symbioses of leguminous plants, human gut microbiota, and in characterizing the function of uncultivable symbionts in hydrothermal vent symbioses [85][86][87][88][89]. Characterization of the light organ proteome with high-throughput techniques allowed for the identification of a large number of host and symbiont proteins using little starting material and demonstrates the value of proteomic analyses in an effort to understand the relationship of a symbiotic association. The results of this study complement prior transcriptomic data, but have also identified a number of proteins of previously unknown function in the squid/Vibrio symbiosis [10]. The high-throughput techniques used here offer new methods for identification of host and symbiont proteins likely important for the maintenance of this and other host/microbe associations.

Supporting Information
Figure S1 Functional analysis of host and symbiont light organ proteomes. A. COG category counts for all symbiont proteins present in the light organ (including putative identifications). B. KOG category counts for all host proteins present in the light organ (including putative identifications) using represen-tative light organ ESTs. (COG/ KOG key: J-translation, ribosomal structure, and biogenesis, A-RNA processing and modification, Ktranscription, L-replication, recombination and repair, B-chromatin structure and dynamics, D-cell cycle control, cell division and chromosome partitioning, Y-nuclear structure, V-defense mechanisms, T-signal transduction mechanisms, M-cell wall, membrane and envelope biogenesis, N-cell motility, Z-cytoskeleton, Wextracellular structures, U-intracellular trafficking, secretion and vesicular transport, O-posttranslational modification, protein turnover and chaperones, C-energy production and conversion, G-carbohydrate transport and metabolism, E-amino acid transport and metabolism, F-nucleotide transport and metabolism, Hcoenzyme transport and metabolism, I-lipid transport and metabolism, P-inorganic ion transport and metabolism, Qsecondary metabolites biosynthesis, transport and catabolism, Rgeneral function prediction only, S-function unknown).