The Role of Cadaverine Synthesis on Pneumococcal Capsule and Protein Expression

Invasive infections caused by Streptococcus pneumoniae, a commensal in the nasopharynx, pose significant risk to human health. Limited serotype coverage by the available polysaccharide-based conjugate vaccines coupled with increasing incidence of antibiotic resistance complicates therapeutic strategies. Bacterial physiology and metabolism that allows pathogens to adapt to the host are a promising avenue for the discovery of novel therapeutics. Intracellular polyamine concentrations are tightly regulated by biosynthesis, transport and degradation. We previously reported that deletion of cadA, a gene that encodes for lysine decarboxylase, an enzyme that catalyzes cadaverine synthesis results in an attenuated phenotype. Here, we report the impact of cadA deletion on pneumococcal capsule and protein expression. Our data show that genes for polyamine biosynthesis and transport are downregulated in ∆cadA. Immunoblot assays show reduced capsule in ∆cadA. Reduced capsule synthesis could be due to reduced transcription and availability of precursors for synthesis. The capsule is the predominant virulence factor in pneumococci and is critical for evading opsonophagocytosis and its loss in ∆cadA could explain the reported attenuation in vivo. Results from this study show that capsule synthesis in pneumococci is regulated by polyamine metabolism, which can be targeted for developing novel therapies.


Introduction
Streptococcus pneumoniae (pneumococcus) is a Gram-positive encapsulated pathogen that resides asymptomatically in the nasopharynx of healthy humans. In children, elderly, and immunocompromised individuals pneumococci can become pathogenic, causing mild to severe infections such as sinusitis, meningitis, community acquired pneumonia and septicemia [1]. Pneumococcal infections cause approximately 2 million deaths globally with about 900,000 reported cases in the United States alone, resulting in approximately 400,000 hospitalizations annually [2]. There are over 90 pneumococcal serotypes with unique capsular polysaccharide structures [3] and only 24 serotypes are included in the current existing polysaccharide-based vaccines, PCV13 and PPSV23 [4], combined. Increased resistance to antibiotics such as penicillin, cephalosporins, and fluoroquinolones also complicates treatment [5,6]. The diversity of pneumococcal serotypes, coupled with genomic pORI280 [28] by overlap extension (SOEing) PCR [24]. The recombinant product was transformed into TIGR4 as described previously [29]. Transformants were selected on BAP with spectinomycin (100 µg/mL) and cadA gene deletion was confirmed by sequencing. To establish that cadaverine is responsible for loss of the capsule, we complemented the ∆cadA mutant in trans by cloning cadA gene amplified from TIGR4 (Table 1) into pABG5 vector for analysis of transcription [30], and complemented transformants were selected on BAP with kanamycin (50 µg/mL) and confirmed by PCR.

In Vitro Growth of TIGR4 and ∆cadA
TIGR4 and ∆cadA were inoculated into THY (10 5 colony forming units (CFU)/mL) and growth was monitored by measuring optical density at 600 nm (OD 600 nm ) using a Cytation 5 multifunction plate reader (BioTek, Winooski, VT, USA) at 37 • C with 5% CO 2 . We used GrowthRates [31], a software tool that uses the output of plate reader files to automatically monitor growth rate in exponential phase, lag phase and maximal OD to compare TIGR4 and ∆cadA growth curves. We also measured the viability of TIGR4 and ∆cadA in THY by plating cells on BAP every 2 h for CFU enumeration. Morphology was compared by Gram staining of mid-log phase (OD 600 nm 0.4) cultures of bacteria.

Measurement of Capsular Polysaccharides
Capsular polysaccharide (CPS) was quantified by immunoblotting, as described [32]. An isogenic capsular variant of TIGR4 (T4R) in which the cps locus is replaced with the Janus cassette resulting in an unencapsulated phenotype [33] was used as a control. Briefly, bacterial strains were cultured in THY supplemented with 10% fetal bovine serum to an OD 600 nm of 0.2, plated on BAP for CFU enumeration and 1 mL bacteria was pelleted and stored at −20 • C until further use. The CFUs for all strains were~9.0 × 10 7 /mL. CPS was extracted in a lysis buffer (4% deoxycholate, 50 µg/mL DNAseI and 50 µg/mL RNAse) at 37 • C for 10 min and centrifuged at 18,000× g for 10 min. Four threefold serial dilutions of the supernatant in phosphate-buffered saline (PBS) were spotted in duplicate (2 µL) on 0.2-µm-pore-size nitrocellulose membrane (Thermo Fisher Scientific, Waltham, MA, USA) with suction and air dried at 60 • C for 15 min. The membranes were blocked and incubated with rabbit anti-serotype 4 serum (Cedarlane, Burlington, NC, USA) at 1:1000 and a horseradish peroxidase (HRP) conjugated goat anti-rabbit secondary antibody (Thermo Fisher Scientific, Waltham, MA, USA) at 1:10,000. Membranes were developed with enhanced chemiluminiscence (ECL) detection (Thermo Fisher Scientific, Waltham, MA, USA) and scanned using a ChemiDoc XRS+ with Image Lab software (Bio-Rad, Hercules, CA, USA).

Proteomics
Total proteins were isolated from mid-log phase (OD 600 nm 0.4) TIGR4, and ∆cadA (n = 4) cultured in THY and subjected to liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis as described earlier [24,34]. Briefly, proteins were isolated from bacterial pellets sonicated in NP-40 lysis buffer (0.5% NP-40, 150 mM NaCl, 20 mM CaCl 2 ·2H 2 O, 50 mM Tris, pH 7.4) supplemented with 1X protease inhibitor cocktail/ethylenediaminetetraacetic acid (EDTA) using a Covaris S220 focused-ultrasonicator (Covaris, Woburn, MA, USA). Protein concentration from the supernatant was determined using a Pierce bicinchoninic acid assay (BCA) Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) and 30 µg was precipitated with methanol and chloroform (4:1), solubilized in 8 M urea, reduced (0.005 M dithiothreitol (DTT) at 65 • C for 10 min) and alkylated (0.01 M iodoacetamide at 37 • C for 30 min) and digested with porcine trypsin (2 µg at 37 • C, overnight, 50:1 ratio of protein: trypsin, Promega Corporation, Madison, WI, USA). Tryptic peptides were desalted using a C18 spin column (Thermo Fisher Scientific, Waltham, MA, USA) and analyzed by linear trap quadropole (LTQ) Orbitrap Velos mass spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) equipped with an Advion nanomate electrospray ionization (ESI) source (Advion, Ithaca, NY, USA). Peptides (500 ng) were eluted from a C18 column (100-µm id × 2 cm, Thermo Fisher Scientific) onto an analytical column (75-µm ID × 10 cm, C18, Thermo Fisher Scientific) using a 180 min gradient with 99.9% acetonitrile, 0.1% formic acid at a flow rate of 400 nL/min and introduced into an LTQ-Orbit rap. Data dependent scanning was performed by the Xcalibur v 2.1.0 software [35] using a survey mass scan at 60,000 resolution in the Orbitrap analyzer scanning mass/charge (m/z) 400-1600, followed by collision-induced dissociation (CID) tandem mass spectrometry (MS/MS) of the 14 most intense ions in the linear ion trap analyzer. Precursor ions were selected by the monoisotopic precursor selection (MIPS) setting with selection or rejection of ions held to a ±10 ppm window. Dynamic exclusion was set to place any selected m/z on an exclusion list for 45 s after a single MS/MS. Tandem mass spectra were searched against a Streptococcus pneumoniae serotype4 strain ATCC BAA/TIGR4 fasta protein database downloaded from UniProtKB to which common contaminant proteins (e.g., human keratins obtained at ftp://ftp.thegpm.org/fasta/cRAP) were appended. All MS/MS spectra were searched using Thermo Proteome Discoverer 1.3 (Thermo Fisher Scientific) considering fully tryptic peptides with up to two missed cleavage sites. Variable modifications considered during the search included methionine oxidation (15.995 Da), and cysteine carbamidomethylation (57.021 Da). Peptides were identified at 99% confidence with XCorr score cutoffs [36] based on a reversed database search. The protein and peptide identification results were visualized with Scaffold v 3.6.1 (Proteome Software Inc., Portland, OR, USA). Protein identifications with a minimum of two peptides identified at 0.1% peptide false discovery rate (FDR) were deemed correct. Significant changes in protein expression between ∆cadA and TIGR4 were identified by Fisher's exact test at a p-value of ≤0.054 and fold change of ±1.3. Fold changes in protein expression were calculated using weighted normalized spectra with 0.5 imputation value. Various bioinformatics resources such as DAVID [37], KEGG [38] and STRING [39] were utilized to determine the functions of the identified proteins. The PRoteomics IDEntifications (PRIDE) database is a centralized, standards compliant, public data repository for proteomics data. The mass spectrometry proteomics data from this study is deposited to the ProteomeXchange Consortium via the PRIDE partner repository [40] with the dataset identifier PXD008621.

Quantitative Real Time PCR
Gene expression in ∆cadA was measured by quantitative reverse transcription-PCR (qRT-PCR). The primers used for qRT-PCR are listed in Table 1. All primers were validated by performing a melt curve analysis with SYBR Green (Thermo Fisher Scientific Waltham, MA, USA) to ensure the amplification of a single specific product. In brief, total RNA was purified from mid-log phase TIGR4 and ∆cadA grown in THY (n = 3) using the RNeasy Midi kit and QIAcube (Qiagen, Valencia, CA, USA). Purified total RNA (7.5 ng/reaction) was transcribed into cDNA and qRT-PCR was performed using the SuperScript III Platinum SYBR Green One-Step qRT-PCR Kit (Thermo Fisher Scientific, Waltham, MA, USA) as previously described [34]. Relative quantification of gene expression was determined by using the Stratagene Mx3005P qPCR system (Agilent, Santa Clara, CA, USA). Expression of target genes, speE, potD, cps4A, aguA, lys9, and nspC was normalized to the expression of gyrB and fold change determined by the comparative C T method.

Impact of ∆cadA on Pneumococcal Growth
S. pneumoniae TIGR4 and ∆cadA had comparable growth kinetics in THY, as reported previously [24]. The deletion mutant had a shorter lag phase than TIGR4 and had a lower cell density at stationary phase ( Figure 1A(i)). There is evidence that pneumococcal serotypes with high colonization prevalence have a short lag phase in vitro when cultured in complete medium compared to invasive serotypes [41]. The shorter lag phase in ∆cadA could have implications for its invasiveness that needs confirmation in future studies. Since polyamines impact a number of cellular processes that can modulate growth, it is possible that deletion of ∆cadA alters some of these processes that ultimately results in reduced cell density. However, there is no significant difference in the exponential growth rate between TIGR4 (0.011 min −1 ) and ∆cadA (0.015 min −1 ). Both TIGR4 and ∆cadA had comparable CFUs during growth in THY ( Figure 1A(ii)). Mid-log phase TIGR4 ( Figure 1B(i)) and ∆cadA cells ( Figure 1B(ii)) showed no difference morphology as both exhibited the characteristic diplococci morphology. These results are similar to our earlier observation that deletion of lysine decarboxylase has no qualitative difference in pneumococcal growth in vitro. However, lysine decarboxylase is indispensable for survival in murine models of colonization, pneumonia and sepsis [24].

Quantitative Real Time PCR
Gene expression in ΔcadA was measured by quantitative reverse transcription-PCR (qRT-PCR). The primers used for qRT-PCR are listed in Table 1. All primers were validated by performing a melt curve analysis with SYBR Green (Thermo Fisher Scientific Waltham, MA, USA) to ensure the amplification of a single specific product. In brief, total RNA was purified from mid-log phase TIGR4 and ΔcadA grown in THY (n = 3) using the RNeasy Midi kit and QIAcube (Qiagen, Valencia, CA, USA). Purified total RNA (7.5 ng/reaction) was transcribed into cDNA and qRT-PCR was performed using the SuperScript III Platinum SYBR Green One-Step qRT-PCR Kit (Thermo Fisher Scientific, Waltham, MA, USA) as previously described [34]. Relative quantification of gene expression was determined by using the Stratagene Mx3005P qPCR system (Agilent, Santa Clara, CA, USA). Expression of target genes, speE, potD, cps4A, aguA, lys9, and nspC was normalized to the expression of gyrB and fold change determined by the comparative CT method.

Impact of ∆cadA on Pneumococcal Growth
S. pneumoniae TIGR4 and ΔcadA had comparable growth kinetics in THY, as reported previously [24]. The deletion mutant had a shorter lag phase than TIGR4 and had a lower cell density at stationary phase ( Figure 1A(i)). There is evidence that pneumococcal serotypes with high colonization prevalence have a short lag phase in vitro when cultured in complete medium compared to invasive serotypes [41]. The shorter lag phase in ΔcadA could have implications for its invasiveness that needs confirmation in future studies. Since polyamines impact a number of cellular processes that can modulate growth, it is possible that deletion of ΔcadA alters some of these processes that ultimately results in reduced cell density. However, there is no significant difference in the exponential growth rate between TIGR4 (0.011 min −1 ) and ΔcadA (0.015 min −1 ). Both TIGR4 and ΔcadA had comparable CFUs during growth in THY ( Figure 1A(ii)). Mid-log phase TIGR4 ( Figure 1B(i)) and ΔcadA cells ( Figure 1B(ii)) showed no difference morphology as both exhibited the characteristic diplococci morphology. These results are similar to our earlier observation that deletion of lysine decarboxylase has no qualitative difference in pneumococcal growth in vitro. However, lysine decarboxylase is indispensable for survival in murine models of colonization, pneumonia and sepsis [24].

Lysine Decarboxylase is Required for Capsule Production in S. pneumoniae
Isogenic deletion of lysine decarboxylase in S. pneumoniae TIGR4 led to attenuation in murine models of colonization, pneumococcal pneumonia and sepsis [24]. The capsule renders pneumococci resistant to opsonophagocytosis and is essential for pneumococcal virulence [42]. Loss of the capsule associated with impaired cadaverine synthesis could explain the observed attenuation of virulence in S. pneumoniae ∆cadA. We compared total CPS from TIGR4, T4R, ∆cadA and ∆cadA complemented strains (∆cadA (comp)) using serotype 4-specific CPS antibodies. Our results ( Figure 2) clearly show that deletion of cadA results in loss of the capsule in two independently-derived mutants that lack lysine decarboxylase (∆cadA1 [24] and ∆cadA2 (this study)) compared to TIGR4. Both ∆cadA deletion strains exhibit loss of the capsule ( Figure 2) ruling out the possibility that the observed phenotype is due to a random change in the genome elsewhere and not specific to cadA deletion. This was further confirmed by complementation. CPS in ∆cadA is fully restored to the levels comparable to that of TIGR4 by complementation with pABG5-cadA construct. These results clearly demonstrate that deletion of lysine decarboxylase in S. pneumoniae results in the loss of CPS. Loss of the capsule could render ∆cadA susceptible to host defenses resulting in an attenuated phenotype in murine models of colonization and invasive disease. Impact of lysine decarboxylase on capsule synthesis in vivo needs to be evaluated in future studies.

Lysine Decarboxylase is Required for Capsule Production in S. pneumoniae
Isogenic deletion of lysine decarboxylase in S. pneumoniae TIGR4 led to attenuation in murine models of colonization, pneumococcal pneumonia and sepsis [24]. The capsule renders pneumococci resistant to opsonophagocytosis and is essential for pneumococcal virulence [42]. Loss of the capsule associated with impaired cadaverine synthesis could explain the observed attenuation of virulence in S. pneumoniae ∆cadA. We compared total CPS from TIGR4, T4R, ∆cadA and ∆cadA complemented strains (ΔcadA (comp)) using serotype 4-specific CPS antibodies. Our results ( Figure 2) clearly show that deletion of cadA results in loss of the capsule in two independently-derived mutants that lack lysine decarboxylase (∆cadA1 [24] and ΔcadA2 (this study)) compared to TIGR4. Both ∆cadA deletion strains exhibit loss of the capsule ( Figure 2) ruling out the possibility that the observed phenotype is due to a random change in the genome elsewhere and not specific to cadA deletion. This was further confirmed by complementation. CPS in ΔcadA is fully restored to the levels comparable to that of TIGR4 by complementation with pABG5-cadA construct. These results clearly demonstrate that deletion of lysine decarboxylase in S. pneumoniae results in the loss of CPS. Loss of the capsule could render ΔcadA susceptible to host defenses resulting in an attenuated phenotype in murine models of colonization and invasive disease. Impact of lysine decarboxylase on capsule synthesis in vivo needs to be evaluated in future studies. Immunoblot analysis of capsular polysaccharides in TIGR4 and mutant strains. All strains were cultured in THY supplemented with fetal bovine serum (FBS) to mid-log phase. Total capsular polysaccharide (CPS) isolated from equal number of cells for each strain, and 3× dilutions were spotted onto a nitrocellulose membrane. Membranes were probed with rabbit anti-serotype 4 sera and horseradish peroxidase (HRP)-conjugated goat anti-rabbit secondary antibody. Membranes were developed with enhanced chemiluminiscence (ECL) detection and scanned using a ChemiDoc XRS+ with Image Lab software (Bio-Rad, Hercules, CA, USA). Data from representative immunoblot from two independent colonies for each strain are shown.

Lysine Decarboxylase Effects on Pneumococcal Protein Expression
To identify pneumococcal molecular mechanisms that are responsive to lysine decarboxylase, we carried out mass spectrometry based proteomics with TIGR4 and ∆cadA. A total of 772 proteins were identified (Supplementary Table 1) which represents 34.5% of the annotated protein coding genes in the TIGR4 genome [27]. We identified significant changes in the expression of 132 proteins of which 52 are upregulated and 80 are downregulated in ∆cadA compared to TIGR4 (Supplementary Table 2). Molecular functions and pathways represented by the differentially expressed proteins are discussed in the following sections and shown in Table 2. Immunoblot analysis of capsular polysaccharides in TIGR4 and mutant strains. All strains were cultured in THY supplemented with fetal bovine serum (FBS) to mid-log phase. Total capsular polysaccharide (CPS) isolated from equal number of cells for each strain, and 3× dilutions were spotted onto a nitrocellulose membrane. Membranes were probed with rabbit anti-serotype 4 sera and horseradish peroxidase (HRP)-conjugated goat anti-rabbit secondary antibody. Membranes were developed with enhanced chemiluminiscence (ECL) detection and scanned using a ChemiDoc XRS+ with Image Lab software (Bio-Rad, Hercules, CA, USA). Data from representative immunoblot from two independent colonies for each strain are shown.

Lysine Decarboxylase Effects on Pneumococcal Protein Expression
To identify pneumococcal molecular mechanisms that are responsive to lysine decarboxylase, we carried out mass spectrometry based proteomics with TIGR4 and ∆cadA. A total of 772 proteins were identified (Table 1) which represents 34.5% of the annotated protein coding genes in the TIGR4 genome [27]. We identified significant changes in the expression of 132 proteins of which 52 are upregulated and 80 are downregulated in ∆cadA compared to TIGR4 (Table 2). Molecular functions and pathways represented by the differentially expressed proteins are discussed in the following sections and shown in Table 2.

Capsule Biosynthesis
CPS synthesis in TIGR4 is by the Wzy polymerase dependent mechanism. CPS synthesis is a multistep process that begins with the transfer of sugar-1-phosphate on the cytosolic side onto a C55 lipid undecaprenyl-phosphate (Und-P), followed by the addition of remaining sugars by glycosyl transferases to form a repeat unit. The repeat unit structure of serotype 4 CPS consists of galactose, N-acetylmannosamine, N-acetylfucosamine and N-acetylgalactosamine [43]. Und-P oligosaccharide repeat units are translocated to the outer face of the cytoplasmic membrane by Wzx transporter and polymerized into high molecular weight polysaccharide by Wzy polymerase. All genes involved in capsule biosynthesis are present as a single operon between dexB and aliA in pneumococcal genomes. The first four genes in the operon cpsABCD are important for modulation of synthesis and are common to all serotypes [44]. The rest of the genes in the operon are serotype specific. Changes in the expression of cpsA is a good measure of the transcription cps locus. We expected to see a significant reduction in ∆cadA in the expression of some of the enzymes that catalyze the multi-step CPS biosynthesis. Our proteomics data showed reduced expression of UDP-glucose-4-epimerase (Table 2), that catalyzes the conversion of UDP-glucose to UDP-galactose in CPS biosynthesis [45]. While the reduced expression of this protein is consistent with the observed loss of capsule, it is unlikely that this marginal change can explain the magnitude of the loss of CPS comparable to unencapsulated T4R ( Figure 2) without additional changes in the pneumococcal proteome that directly or indirectly impact capsule synthesis. Expression of bifunctional protein GlmU ( Table 2) was significantly downregulated. This enzyme catalyzes the last two sequential reactions in the de novo biosynthetic pathway for UDP-N-acetylglucosamine (UDP-GlcNAc). GlmU catalyzes the reaction that transfers an acetyl group from acetyl coenzyme A to glucosamine 6-phosphate to synthesize acetylated glucosamine 6-phosphate. NagB is an enzyme that catalyzes the conversion of glucosamine 6-phosphate to fructose 6-phosphate and is known to regulate GlmU [46]. In ∆cadA expression of NagA and NagB, two enzymes involved in UDP-GlcNAc degradation was significantly higher compared to TIGR4. Taken together, the net effect of changes in the expression of GlmU, NagA and NagB would result in lower concentrations of UDP-GlcNAc, a precursor for UDP-ManNAc, which is a constituent of serotype 4 CPS repeat unit, and could contribute to reduced CPS synthesis in ∆cadA.

Polyamine Biosynthesis
Deletion of lysine decarboxylase in S. pneumoniae resulted in a significant decrease in the expression of Lys9, Asd, DapA, DapB, Hom, DapH and SP_2096 involved in the biosynthesis of lysine, the substrate for cadA ( Table 2). Expression of N-carbamoylputrescine amidase which catalyzes the synthesis of putrescine from N-carbamoylputrescine in the arginine and proline metabolism was significantly downregulated. Expression of NspC which catalyzes the synthesis of spermidine from carbamoyl spermidine was also significantly downregulated ( Table 2). The net effect of these protein expression changes in polyamine biosynthesis pathways in ∆cadA, is expected to result in reduced intracellular concentrations of cadaverine, putrescine and spermidine (Figure 3). We reported reduced intracellular concentrations of cadaverine, putrescine and spermidine in ∆cadA [24] previously. Results from this study explain this observed impact on intracellular polyamine concentrations in ∆cadA. genomes. The first four genes in the operon cpsABCD are important for modulation of synthesis and are common to all serotypes [44]. The rest of the genes in the operon are serotype specific. Changes in the expression of cpsA is a good measure of the transcription cps locus. We expected to see a significant reduction in ΔcadA in the expression of some of the enzymes that catalyze the multi-step CPS biosynthesis. Our proteomics data showed reduced expression of UDP-glucose-4-epimerase (Table 2), that catalyzes the conversion of UDP-glucose to UDP-galactose in CPS biosynthesis [45].
While the reduced expression of this protein is consistent with the observed loss of capsule, it is unlikely that this marginal change can explain the magnitude of the loss of CPS comparable to unencapsulated T4R ( Figure 2) without additional changes in the pneumococcal proteome that directly or indirectly impact capsule synthesis. Expression of bifunctional protein GlmU (Table 2) was significantly downregulated. This enzyme catalyzes the last two sequential reactions in the de novo biosynthetic pathway for UDP-N-acetylglucosamine (UDP-GlcNAc). GlmU catalyzes the reaction that transfers an acetyl group from acetyl coenzyme A to glucosamine 6-phosphate to synthesize acetylated glucosamine 6-phosphate. NagB is an enzyme that catalyzes the conversion of glucosamine 6-phosphate to fructose 6-phosphate and is known to regulate GlmU [46]. In ∆cadA expression of NagA and NagB, two enzymes involved in UDP-GlcNAc degradation was significantly higher compared to TIGR4. Taken together, the net effect of changes in the expression of GlmU, NagA and NagB would result in lower concentrations of UDP-GlcNAc, a precursor for UDP-ManNAc, which is a constituent of serotype 4 CPS repeat unit, and could contribute to reduced CPS synthesis in ∆cadA.

Polyamine Biosynthesis
Deletion of lysine decarboxylase in S. pneumoniae resulted in a significant decrease in the expression of Lys9, Asd, DapA, DapB, Hom, DapH and SP_2096 involved in the biosynthesis of lysine, the substrate for cadA ( Table 2). Expression of N-carbamoylputrescine amidase which catalyzes the synthesis of putrescine from N-carbamoylputrescine in the arginine and proline metabolism was significantly downregulated. Expression of NspC which catalyzes the synthesis of spermidine from carbamoyl spermidine was also significantly downregulated ( Table 2). The net effect of these protein expression changes in polyamine biosynthesis pathways in ΔcadA, is expected to result in reduced intracellular concentrations of cadaverine, putrescine and spermidine ( Figure 3). We reported reduced intracellular concentrations of cadaverine, putrescine and spermidine in ΔcadA [24] previously. Results from this study explain this observed impact on intracellular polyamine concentrations in ΔcadA.

Peptidoglycan
The rigid, stable shape of bacteria is provided by the peptidoglycan layer which is made of N-acetylmuramic acid-(β-1, 4)-N-acetylglucosamine (MurNAc-GlcNAc) disaccharides cross-linked by peptides. The peptidoglycan layer, teichoic acid (TA) and lipoteichoic acid (LTA) constitute the cell wall of Gram-positive bacteria such as S. pneumoniae [47]. Both LTA and TA contain phosphorylcholine (PC), which plays a major role in S. pneumoniae adhesion and also forms attachment of choline binding surface proteins (CBPs) [48]. The peptidoglycan layer provides attachment for many structural components including the polysaccharide capsule [49] and it is important in the adhesion of S. pneumoniae to the host tissues. Our results show reduced expression of penicillin-binding protein 2X (Table 2), which is involved in peptidoglycan biosynthesis that contributes to bacterial cell division and growth [50,51]. Choline kinase, an enzyme which catalyzes the synthesis of PC from choline was downregulated in ∆cadA relative to TIGR4. In some Gram-positive pathogens, choline kinase is known to be important for the production of cell wall elements and LTA [52]. PC is necessary for the adherence of S. pneumoniae during the transition from colonization to invasive disease [48]. Reduced choline kinase expression in ∆cadA could also modulate virulence.

ABC Transporters
Our results show that expression of ABC transporters that bind metal ions is significantly altered in ∆cadA. Manganese, a transition metal ion is a prosthetic group in superoxide dismutase, has direct antioxidant properties and is known to be important for pneumococcal physiology. An ABC-type permease PsaBCA [53] transports manganese and psa mutants are avirulent [54] due to their hypersensitivity to oxidative stress [51]. Expression of Manganese ABC transporter-substrate-binding lipoprotein (PsaA) and Manganese ABC transporter, ATP -binding protein (PsaB) is significantly higher in ∆cadA (Table 2). An ABC transporter potentially responsible for iron-siderophore transport (FhuD), specifically ferric hydroxamate is significantly lower in ∆cadA compared to TIGR4. Iron is a critical cofactor for many enzymes, and uptake and efflux of this critical micronutrient from different host niches that differ in the quantity and form of iron is an important aspect of pneumococcal pathogenesis. Reduced expression of FhuD could impair iron homeostasis and have an impact on virulence. An alternate explanation involves ∆cadA response to oxidative stress by reducing uptake of iron due to reduced expression of FhuD. High intracellular iron concentrations can result in an increase in oxidative stress [55] and ∆cadA with reduced intracellular polyamine concentrations would be more susceptible to oxidative stress [21]. Increased expression of PsaA and PsaB could also be in response to increased susceptibility to oxidative stress in ∆cadA.
Expression of five ABC transporters, four of which are involved in oligopeptide trafficking (SP_2073, OppA, OppD, OppB and OppF) were downregulated in ∆cadA (Table 2). Reduced expression of oligopeptide transporter proteins could significantly alter transport of substrates that would in turn affect biosynthesis of amino acids, polyamines and other cellular components. There was reduced expression of lysine-transfer RNA (tRNA) ligase, a protein involved in amino acid metabolism. Expression of RplU, which binds to 23S ribosomal RNA (rRNA) in the initial stages of protein translation and ribosome maturation factor (RimP), which modulates final stages of translation was significantly reduced in ∆cadA. Our results show that lysine decarboxylase deficiency results in reduced expression of a number of proteins involved in amino acid transport and metabolism which could impair pneumococcal growth and fitness. For survival in the host, pathogenic bacteria have to cope with phosphate (Pi) limiting or enriched host microenvironments. Phosphate is the component of nucleic acids, phospholipids and energy storage (ATP). Bacteria acquire phosphate by Pi-specific transport systems. Under Pi limitation Pho regulon is activated and results in Pi import through an ABC transporter complex PstSACB [56]. Our data showed increased expression of PstS2, PstB3, PstC and PhoU (Table 2) which can impact Pi homeostasis. Low Pi is known to alter virulence factor expression in a number of pathogenic bacteria including E. coli [57]. It is known that activation of Pho regulon in pathogenic bacteria under Pi starvation conditions, activates oxidative stress response through mechanisms that are yet to be described [56]. Increased expression of Pst system could be part of oxidative stress response and possibly contribute to altered virulence in ∆cadA through mechanism that are not known at present.

Pentose Phosphate Pathway
The ability of bacterial pathogens to survive in the host largely depends on acquiring nutrients and adapting their metabolism to different host microenvironments. Pneumococci have the ability to utilize a variety of carbohydrates as carbon sources via Embden-Meyerhof-Parnas (EMP) pathway (glycolysis) and pentose phosphate pathway (PPP) [27]. Pyruvate and ATP are the end products of glycolysis. The oxidative branch of PPP generates ribulose 5-phosphate and reduced nicotinamide adenine dinucleotide phosphate (NADPH) while the non-oxidative branch generates a number of sugar phosphates that provide precursors for nucleotide, amino acid and vitamin B6 synthesis. Our data supports increased carbon flux through the non-oxidative branch of PPP due to the observed increase in the expression of transketolase (Tkt), an enzyme that catalyzes the interconversion of sugar-phosphates in the pathway. Expression of all proteins SgaR2, SgaB2, SgaT2, including TktN and TktC, encoded in a single regulon belonging to BgIG family transcriptional regulator is higher in ∆cadA ( Table 2). We also identified a significantly higher expression for one of the two general proteins of PTS (phosphotransferase system), phosphocarrier protein HPr (PtsH) which could support the proper functioning of PTS transport systems in the BgIG regulon in ∆cadA. A shift in metabolism towards PPP is often in response to oxidative stress to maintain NADH/NADPH redox homeostasis and to synthesize ribose-5-phosphate for DNA repair [58].

Carbohydrate Metabolism
Our results show increased expression of proteins that are involved in galactose and tagatose catabolism. Expression of LacB is higher in ∆cadA (Table 2) which generates the substrate tagatose 6-phosphate for the enzyme LacD. Enzymatic action of LacB and LacD would result in higher levels of glyceraldehyde 3-phosphate, which can be channeled into PPP by transaldolase and transketolase. We identified reduced expression of catabolite control protein A (CcpA), a protein that regulates carbon catabolite repression in our lysine decarboxylase mutant. CcpA is known to control the expression of a number of virulence factors in Gram-positive bacteria [59]. For instance, in S. pneumoniae, CcpA contributes to sugar metabolism and virulence [60].

Capsule Biosynthesis
We did not identify major differences in the expression of proteins that catalyze different steps in CPS synthesis. To determine whether CPS synthesis is regulated at the transcriptional level, we compared cps4A mRNA expression between TIGR4 and ∆cadA by qRT-PCR. Our results show a significant reduction in the expression of cps4A (Table 3), which could explain reduced CPS synthesis in ∆cadA.

Polyamine Synthesis and Transport
Our proteomics data identified significantly reduced expression of proteins involved in the biosynthesis of putrescine, spermidine and cadaverine (Figure 2). The genes encoding proteins NspC, Lys9, AguA and AguB are predicted to constitute a single operon in the genome [61]. Based on our proteomics results, we expected to see reduced expression of polyamine biosynthesis (lys9, aguA, and nspC) genes and our qRT-PCR data shows reduced expression of these genes in ∆cadA (Table 3). Impaired lysine decarboxylase also resulted in a significant reduction in the expression of spermidine synthase (Table 3). Pneumococci can compensate for reduced polyamine synthesis by increasing the import of extracellular polyamines. Extracellular polyamine uptake in pneumococci is predicted to be via a single ABC transporter, organized as a four gene operon, potABCD [62]. The proposed structure of the putrescine/spermidine transporter has PotD, an extracellular substrate binding domain that binds polyamines, PotB, and PotC which form transmembrane channels that transport polyamines and PotA that is a membrane associated cytosolic ATPase [23]. We did not detect polyamine transport proteins in our proteomics data. We measured the expression of potD mRNA in ∆cadA and observed a significant decrease (Table 3), which would further contribute to reduced intracellular concentrations of putrescine and spermidine, putative substrates for the PotABCD transporter.

Discussion
Polyamines are important for host-pathogen interactions during bacterial infections. Polyamines in pathogenic bacteria play an important role in physiological stress responses and adaptation to growth in vivo. Putrescine is a constituent of the cell wall in a number of Gram-negative bacteria, such as Salmonella enterica, E. coli and Proteus mirabilis [63,64], while cadaverine is covalently linked to the peptidoglycan layer of Veillonella alcalescens [65]. Spermidine modulates autolysis and ion trafficking across the cell membrane in S. pneumoniae, protecting pneumococci from cationic antimicrobial compounds [7]. Cadaverine regulates porins that control the permeability of membranes [30], and enables E. coli to survive acidic stress [66]. Current literature clearly demonstrates that deletion of polyamine synthesis and/or transport in pathogenic bacteria (including pneumococci) leads to reduced virulence in animal models [24,67,68]. To date, studies that describe specific host innate immune mechanisms induced by pathogenic bacteria with altered polyamine metabolism, or specific effects of impaired polyamine metabolism on pathogen molecular mechanisms are largely unknown. A few examples of specific roles of polyamines in bacterial pathogens include the following: in intracellular pathogen Shigella, cadaverine is shown to be important for reducing enterotoxic activity [69], inhibiting trans-epithelial migration of polymorphonuclear neutrophils, increasing survival in macrophages, and enhancing antioxidant defenses in vitro [20]. Salmonella typhimurium mutant deficient in polyamine biosynthesis has reduced invasive potential and survival in epithelial cells in vitro and is attenuated in a mouse model of typhoid fever [70]. In Yersinia pestis, loss of intracellular spermidine and putrescine affects biofilm formation and biosynthesis defective Y. pestis is less virulent in a murine model of bubonic plague [71]. When Francisella tularensis is cultured in the presence of spermine or spermidine prior to macrophage infection assays, there is reduced pro-inflammatory response in vitro [55]. Thus altered polyamine metabolism has varying impacts on bacterial virulence.
Invasive infections caused by S. pneumoniae, a commensal in the nasopharynx, pose significant risk to human health. The available polysaccharide-based conjugate vaccines are effective in reducing vaccine serotypes in population, but ultimately lead to serotype replacement [72] from a reservoir of more than 90 capsular serotypes [3]. The critical role of the upper respiratory microbiota and the consequence of its perturbation in various pathophysiological conditions mandates a cautious approach to drug and vaccine discovery. Anti-virulence strategies that target genes/proteins that are necessary for fitness during invasive infection can offer serotype-independent coverage without impacting nasopharyngeal colonization, that is, disarm, but not eradicate, pneumococci [68]. Specific aspects of bacterial physiology and metabolism that allow pathogens to adapt to the host are a promising avenue for the discovery of novel therapeutics. Intracellular polyamine concentrations are tightly regulated by biosynthesis, transport and degradation. Polyamine transport and synthesis genes are conserved in pneumococci. Deletion of the polyamine transport (potABCD) operon or spermidine synthase and lysine decarboxylase biosynthesis genes had no significant impact on pneumococcal growth in vitro [24]. However, enhanced bacterial clearance in murine models of colonization, pneumococcal pneumonia, and sepsis were seen with mutant strains [24]. In a murine model of pneumococcal pneumonia, ∆potABCD failed to elicit host defenses that are intact in TIGR4, and was cleared more efficiently by opsonophagocytosis by neutrophils [34]. TIGR4 cultured in vitro (in THY) had spermidine as the most abundant intracellular polyamine, followed by cadaverine and putrescine [24]. Polyamine transport and metabolism impaired ∆potABCD, ∆cadA and ∆speE showed reduced levels of spermidine, cadaverine and putrescine, relative to TIGR4. [24].
Here we report in vitro characterization of lysine decarboxylase deficient pneumococci. Our results clearly demonstrate loss of CPS in ∆cadA. This loss of the capsule is specific to the deletion of the cadA gene, as complementation of the wild type gene restored capsule, comparable to that of TIGR4. Invasive pneumococcal serotypes can resist complement mediated opsonophagocytosis by neutrophils due to the presence of the capsule [71,73]. All three complement pathways are activated for opsonophagocytosis [73,74]; classical, lectin and alternative. Activation of the classical pathway is by antibody (including non-specific immunoglobulin M (IgM) produced during infection), or by C-reactive protein (CRP), an acute phase protein. Steric inhibition of the interaction between complement components and the Fc portion of immunoglobulins by the capsule enables pneumococci to evade this host defense. The alternative pathway is constitutively activated at low levels by complement protein C3b, and capsule-mediated resistance to opsonophagocytosis also includes decreased cleavage of C3b to iC3b [75]. Pneumococci regulate capsule expression as they invade host tissues. In the initial stages of colonization of the nasopharynx, they express the capsule to evade mucus, and in the subsequent stage of adhesion, the capsule is downregulated to expose surface adhesins for adherence and colonization. The invasive phase requires upregulation of the capsule to resist opsonophagocytosis [76]. Some of the known regulatory mechanisms for CPS synthesis involve phase variation [77], deletion of pgdA and adr, acetylases of peptidoglycan and increased transcription of cpsA [78] and deletion of pyruvate oxidase (spxB) that all lead to increased CPS [79], mutations in arcD that result in reduced CPS [80], and tyrosine phosphorylation of CpsBCD proteins that modulate CPS levels [81]. Here, for the first time, we show that polyamines, specifically cadaverine modulate CPS synthesis. Inactivation of lysine decarboxylase in pneumococci results in reduced capsule synthesis. The mechanisms that result in reduced CPS synthesis are due to the combinatorial effects on the regulation of UDP-GlcNAc synthesis and degradation ( Figure 4) that could reduce the availability of UDP-ManNAc of the CPS repeat unit in serotype 4 pneumococci with concomitant transcriptional downregulation of cps4A gene. A moderate reduction in cpsA expression in TIGR4 capsule promoter mutant resulted in twofold reduction in CPS [78]. In ∆cadA, there is a twofold reduction in the expression of cps4A mRNA (Table 3), which would be expected to have a greater impact on CPS synthesis. The intersection between central metabolism and virulence is becoming evident for many bacterial pathogens including pneumococci. As pneumococci sense and adapt to different host niches, they modulate capsule synthesis which requires a shift in the metabolism towards increased synthesis of precursors for CPS synthesis. A recent study showed that reduced acetyl-coA levels result in loss of capsule [56], supporting the link between central metabolism and capsule formation. synthesis. A recent study showed that reduced acetyl-coA levels result in loss of capsule [56], supporting the link between central metabolism and capsule formation. Given the well-established role of polyamines in bacterial adaptation to oxidative stress [24], the reduced intracellular polyamines due to reduced expression of enzymes catalyzing synthesis (proteomics data, Table 2) and transport (qRT-PCR data, Table 3) could explain the observed shift in metabolism that increases carbon flux through the pentose phosphate pathway. Reduced uptake of iron and upregulation of manganese transporter proteins, known pneumococcal virulence factors linked to oxidative stress response support this idea. It is possible that increased oxidative stress in ∆cadA results in metabolic adaptation that diverts the energy from capsule production towards stress responses. With data presented here, it is difficult to distinguish between correlation and causation due to impaired cadaverine synthesis. Future studies that utilize additional omics approaches and data analysis in an integrated manner are required to deconvolute the complexity of pneumococcal response to altered polyamine metabolism. Although we have begun to identify some of the molecular mechanisms that are responsive to polyamines in pneumococci that govern virulence, annotation of polyamine metabolism genes is limited in pneumococci. Our knowledge of polyamine metabolism in pathogenic bacteria and specific effects of polyamines on bacterial translation relies heavily on studies from E. coli. Basic biochemistry pertaining to substrate specificity of the transporters and biosynthesis genes is yet to be determined. Our data shows that reduced expression of putrescine and spermidine biosynthesis in ΔcadA could ultimately result in the altered intracellular polyamines that we reported earlier [24]. Future studies focused on determining individual contribution of these polyamine biosynthesis genes to pneumococcal virulence are necessary for a comprehensive description of the role of polyamine synthesis in pneumococcal pathogenesis. Specific effects of polyamines on protein synthesis in bacteria are Given the well-established role of polyamines in bacterial adaptation to oxidative stress [24], the reduced intracellular polyamines due to reduced expression of enzymes catalyzing synthesis (proteomics data, Table 2) and transport (qRT-PCR data, Table 3) could explain the observed shift in metabolism that increases carbon flux through the pentose phosphate pathway. Reduced uptake of iron and upregulation of manganese transporter proteins, known pneumococcal virulence factors linked to oxidative stress response support this idea. It is possible that increased oxidative stress in ∆cadA results in metabolic adaptation that diverts the energy from capsule production towards stress responses. With data presented here, it is difficult to distinguish between correlation and causation due to impaired cadaverine synthesis. Future studies that utilize additional omics approaches and data analysis in an integrated manner are required to deconvolute the complexity of pneumococcal response to altered polyamine metabolism. Although we have begun to identify some of the molecular mechanisms that are responsive to polyamines in pneumococci that govern virulence, annotation of polyamine metabolism genes is limited in pneumococci. Our knowledge of polyamine metabolism in pathogenic bacteria and specific effects of polyamines on bacterial translation relies heavily on studies from E. coli. Basic biochemistry pertaining to substrate specificity of the transporters and biosynthesis genes is yet to be determined. Our data shows that reduced expression of putrescine and spermidine biosynthesis in ∆cadA could ultimately result in the altered intracellular polyamines that we reported earlier [24]. Future studies focused on determining individual contribution of these polyamine biosynthesis genes to pneumococcal virulence are necessary for a comprehensive description of the role of polyamine synthesis in pneumococcal pathogenesis. Specific effects of polyamines on protein synthesis in bacteria are known. Studies in E. coli have identified a "polyamine modulon", a set of 17 genes whose translation is enhanced by polyamines [82]. Mechanisms by which polyamines increase protein synthesis include formation of the initiation complex due to structural changes in the Shine-Dalgarno sequence and the initiation codon AUG, initiation from inefficient initiation codon, and +1 frame shifting and suppression of nonsense codons. Here we report altered expression of a number of pneumococcal proteins in response to impaired lysine decarboxylase. It is possible that some of these proteins are regulated by polyamines utilizing the mechanisms described in E. coli. Future studies focused on understanding the specific mechanisms utilized by polyamines to impact protein expression can result in mechanistic insights into the impact of altered metabolism on pneumococcal virulence. Despite these knowledge gaps, our previous work has clearly established the importance of polyamine transport and synthesis genes in pneumococcal pathogenesis. In this study we show that impaired polyamine synthesis results in reduced capsule synthesis. This foundational knowledge at the intersection of polyamine metabolism and pneumococcal virulence reinforces the need for future mechanistic studies focused on developing novel therapeutics targeting polyamine metabolism in pneumococci.