Arginine Decarboxylase Is Essential for Pneumococcal Stress Responses

Polyamines such as putrescine, cadaverine, and spermidine are small cationic molecules that play significant roles in cellular processes, including bacterial stress responses and host–pathogen interactions. Streptococcus pneumoniae is an opportunistic human pathogen, which causes several diseases that account for significant morbidity and mortality worldwide. As it transits through different host niches, S. pneumoniae is exposed to and must adapt to different types of stress in the host microenvironment. We earlier reported that S. pneumoniae TIGR4, which harbors an isogenic deletion of an arginine decarboxylase (ΔspeA), an enzyme that catalyzes the synthesis of agmatine in the polyamine synthesis pathway, has a reduced capsule. Here, we report the impact of arginine decarboxylase deletion on pneumococcal stress responses. Our results show that ΔspeA is more susceptible to oxidative, nitrosative, and acid stress compared to the wild-type strain. Gene expression analysis by qRT-PCR indicates that thiol peroxidase, a scavenger of reactive oxygen species and aguA from the arginine deiminase system, could be important for peroxide stress responses in a polyamine-dependent manner. Our results also show that speA is essential for endogenous hydrogen peroxide and glutathione production in S. pneumoniae. Taken together, our findings demonstrate the critical role of arginine decarboxylase in pneumococcal stress responses that could impact adaptation and survival in the host.


Introduction
Acidification of the phagolysosome and production of reactive oxygen species (ROS)/ reactive nitrogen species (RNS) by immune cells are major defense mechanisms against invading pathogens [1,2]. ROS are produced by specialized phagocytic cells (macrophages and neutrophils) through the nicotinamide adenine dinucleotide phosphate oxidase (NADPH) complex upon detection of pathogen-associated molecular patterns. RNS are generated by inducible nitric oxide synthases (iNOS), which generate NO • , contributing to pathogen killing [3,4]. Phagolysosome maturation generates an acidified environment that reduces pH from 6.5 to 4.0 and is a key defense mechanism against invading pathogens [1]. Thus, to colonize, inhabit, and cause disease in the host, pathogens need to circumvent the antimicrobial effects of the oxidative, nitrosative, and acidified milieu in the host [5]. Bacterial protective measures against ROS and RNS include detoxifying enzymes such as catalases, peroxidases, superoxide dismutase, and DNA repair systems [6]. Bacterial defense against acid stress involves the agmatine deiminase system, ATPase system, and

SpeA-Deficient Pneumococci Are Susceptible to Oxidative Stress
Our results show that compared to the WT strain, speA is more susceptible to hydrogen peroxide stress. When cultured in the presence of low concentrations of exogenous hydrogen peroxide (0.5, 0.75, and 1 mM), there was no significant effect on the growth of WT and ∆speA strains at 15 and 30 min post-exposure (data not shown). However, in the presence of 2.5 or 5 mM H 2 O 2 , survival of ∆speA reduced by 33% and 66% relative to the WT strain (0% and 1.1%) at 15 min post-exposure, respectively ( Figure 1A). At 30 min post-exposure, ∆speA viability significantly reduced by 92.4% and 99.5% compared to the WT strain 12% and 29%, at 2.5 and 5 mM H 2 O 2 , concentrations, respectively ( Figure 1B). Susceptibility of ∆speA complemented with the pABG5-speA construct was comparable to that of the WT strain at both time points and H 2 O 2 concentrations ( Figure 1A,B). We measured the impact of the deletion of speA on H2O2 production. Our results show that ΔspeA generates less H2O2 (40%; p = 0.0004) compared to the WT strain (data not shown). This result indicates that it is not higher endogenous H2O2 production that renders ΔspeA more susceptible to exogenous H2O2.
Like hydrogen peroxide, ΔspeA was more susceptible to superoxide stress generated by potassium tellurite. There was no significant difference in the viability of ΔspeA and WT strains at 0.1 mM and 15 min post-exposure, but a significant reduction in ΔspeA viability (~27%) was observed at a concentration of 0.2 mM potassium tellurite (Figure 2A). At 30 min post-exposure, ΔspeA viability significantly reduced by ~23% and ~73% compared to the WT strain (0% and ~27%) at 0.1 mM and 0.2 mM potassium tellurite concentrations, respectively ( Figure 2B). Susceptibility of the complement strain to potassium tellurite was comparable to that of the WT strain at both time points and concentrations (Figure 2A The results represent an average of three independent experiments. The percentage survival relative to untreated controls is shown as a bar with the standard error of the mean, with * indicating p ≤ 0.05 and ** representing p ≤ 0.001, determined by Student's t-test. We measured the impact of the deletion of speA on H 2 O 2 production. Our results show that ∆speA generates less H 2 O 2 (40%; p = 0.0004) compared to the WT strain (data not shown). This result indicates that it is not higher endogenous H 2 O 2 production that renders ∆speA more susceptible to exogenous H 2 O 2 .
Like hydrogen peroxide, ∆speA was more susceptible to superoxide stress generated by potassium tellurite. There was no significant difference in the viability of ∆speA and WT strains at 0.1 mM and 15 min post-exposure, but a significant reduction in ∆speA viability (~27%) was observed at a concentration of 0.2 mM potassium tellurite (Figure 2A). At 30 min post-exposure, ∆speA viability significantly reduced by~23% and~73% compared to the WT strain (0% and~27%) at 0.1 mM and 0.2 mM potassium tellurite concentrations, respectively ( Figure 2B). Susceptibility of the complement strain to potassium tellurite was comparable to that of the WT strain at both time points and concentrations (Figure 2A,B).
Pathogens 2021, 10, x 3 of 13 was comparable to that of the WT strain at both time points and H2O2 concentrations ( Figure 1A,B). We measured the impact of the deletion of speA on H2O2 production. Our results show that ΔspeA generates less H2O2 (40%; p = 0.0004) compared to the WT strain (data not shown). This result indicates that it is not higher endogenous H2O2 production that renders ΔspeA more susceptible to exogenous H2O2.
Like hydrogen peroxide, ΔspeA was more susceptible to superoxide stress generated by potassium tellurite. There was no significant difference in the viability of ΔspeA and WT strains at 0.1 mM and 15 min post-exposure, but a significant reduction in ΔspeA viability (~27%) was observed at a concentration of 0.2 mM potassium tellurite (Figure 2A). At 30 min post-exposure, ΔspeA viability significantly reduced by ~23% and ~73% compared to the WT strain (0% and ~27%) at 0.1 mM and 0.2 mM potassium tellurite concentrations, respectively ( Figure 2B). Susceptibility of the complement strain to potassium tellurite was comparable to that of the WT strain at both time points and concentrations (Figure 2A

Impact of SpeA Deletion on Pneumococcal Gene Expression during Hydrogen Peroxide Stress
To identify gene expression changes that could explain the higher susceptibility of the speA-deficient strain to hydrogen peroxide, we measured the expression of genes known to be part of the pneumococcal oxidative stress response using qRT-PCR. Significant changes in gene expression (p ≤ 0.05 and fold change ≥ 2) in ∆speA relative to untreated and hydrogen-peroxide-treated WT strains were compared to distinguish the impact of gene deletion itself (untreated) from the ∆speA-dependent response to H 2 O 2 ( Table 1 and  Table S1). Deletion of speA resulted in reduced expression, of high-temperature requirement A (htrA) sensor histidine kinase (ciaH), NADH oxidase (noxA), and genes involved in polyamine biosynthesis but led to increased expression of serine protease (prtA) and a gene that encodes the polyamine ATP-binding cassette transporters (ABC) transporter substratebinding protein (potD) ( Table 1). Exposure of ∆speA to H 2 O 2 significantly downregulated thiol peroxidase (tpxD) and agmatine deiminase (aguA) and upregulated the expression of noxA, genes that encode a siderophore transport protein (fhuD), and a manganese ABC transporter substrate-binding lipoprotein (psaA) ( Table 1). Altered expression of these genes in ∆speA shows that arginine decarboxylase is important in pneumococcal hydrogen peroxide stress responses. Surprisingly, most key genes known to be involved in pneumococcal oxidative stress responses were either not significantly altered or had <twofold change in expression (Table S1). We observed a significant change in the expression of genes involved in glutathione (GSH) metabolism that were below our fold-change cutoff (Table S1). Since GSH metabolism is known to be involved in oxidative stress responses, we measured the ratio of reduced/oxidized glutathione (GSH/GSSG) in the mutant and WT strains. We observed a significantly lower GSH/GSSG ratio (0.37 ± 0.09) in the mutant strain compared to the WT strain (1.30 ± 0.06, p < 0.0001). A GSH/GSSG ratio of less than 1 indicates reduced glutathione production, which suggests an impaired redox balance in the mutant strain. These results clearly show that deficiency of polyamines affects the pneumococcal transcriptome, which impairs the redox balance and renders ∆speA more susceptible to stress.

SpeA Is Required for Pneumococcal Nitrosative Stress Responses
Compared to the WT strain, ∆speA was significantly more susceptible to S-nitrosoglutathione (GSNO), a nitric oxide producer. Exposure to 2.5 mM GSNO resulted in a significant reduction in the percentage survival of ∆speA, which ranged between 37% at 15 min and 74% at 60 min post-exposure compared to the WT strain (0% at 15 min and 36% at 60 min post-exposure). There was no significant difference in the survival of the complement pABG5-speA strain and the WT strain in the presence of GSNO ( Figure 3). sulted in a significant reduction in the percentage survival of ΔspeA, which ranged between 37% at 15 min and 74% at 60 min post-exposure compared to the WT strain (0% at 15 min and 36% at 60 min post-exposure). There was no significant difference in the survival of the complement pABG5-speA strain and the WT strain in the presence of GSNO ( Figure 3). The results represent an average of three independent experiments. The percentage survival relative to untreated controls is shown as a bar with the standard error of the mean, with * indicating p ≤ 0.05 and ** representing p ≤ 0.001, determined by Student's t-test.

Effect of Selection of SpeA on Pneumococcal pHi
We hypothesized that increased susceptibility of the mutant strain to oxidative and nitrosative stress could be due to an intracellular acidified environment due to the deletion of speA and measured intracellular pH (pHi). Our result shows that pHi of ΔspeA (~7.3 ± 0.01) differed from that of the WT strain (~7.5 ± 0.01) by 0.2 units (p < 0.0001; Figure  4). The marginal change in pHi in ∆speA is within the physiological range, suggesting that susceptibility to oxidative and nitrosative stress is independent of pHi. , and re-energized with 10% glucose, and baseline fluorescence readings were established in the first 5 min. Controls were supplemented with 10 μM carbonyl cyanide 3-chlorophenylhydrazone (CCCP) as a protonophore (triangles), and fluorescence of samples, including untreated ones (squares), was measured for an additional 5 min. The pHi of untreated TIGR4 (blue squares) and ΔspeA (red squares) was 7.5 and 7.3, respectively. Then, 20 μM of nigericin was added to both treated and untreated samples to dissipate transmembrane gradients over the last 5 min. Graphs represent the mean of three independent experiments. Statistical significance was determined by Student's t-test at a significance level of p ≤ 0.01.

Effect of Selection of SpeA on Pneumococcal pH i
We hypothesized that increased susceptibility of the mutant strain to oxidative and nitrosative stress could be due to an intracellular acidified environment due to the deletion of speA and measured intracellular pH (pH i ). Our result shows that pH i of ∆speA (~7.3 ± 0.01) differed from that of the WT strain (~7.5 ± 0.01) by 0.2 units (p < 0.0001; Figure 4). The marginal change in pH i in ∆speA is within the physiological range, suggesting that susceptibility to oxidative and nitrosative stress is independent of pH i . sulted in a significant reduction in the percentage survival of ΔspeA, which ranged between 37% at 15 min and 74% at 60 min post-exposure compared to the WT strain (0% at 15 min and 36% at 60 min post-exposure). There was no significant difference in the survival of the complement pABG5-speA strain and the WT strain in the presence of GSNO (Figure 3).

Effect of Selection of SpeA on Pneumococcal pHi
We hypothesized that increased susceptibility of the mutant strain to oxidative and nitrosative stress could be due to an intracellular acidified environment due to the deletion of speA and measured intracellular pH (pHi). Our result shows that pHi of ΔspeA (~7.3 ± 0.01) differed from that of the WT strain (~7.5 ± 0.01) by 0.2 units (p < 0.0001; Figure  4). The marginal change in pHi in ∆speA is within the physiological range, suggesting that susceptibility to oxidative and nitrosative stress is independent of pHi. , and re-energized with 10% glucose, and baseline fluorescence readings were established in the first 5 min. Controls were supplemented with 10 μM carbonyl cyanide 3-chlorophenylhydrazone (CCCP) as a protonophore (triangles), and fluorescence of samples, including untreated ones (squares), was measured for an additional 5 min. The pHi of untreated TIGR4 (blue squares) and ΔspeA (red squares) was 7.5 and 7.3, respectively. Then, 20 μM of nigericin was added to both treated and untreated samples to dissipate transmembrane gradients over the last 5 min. Graphs represent the mean of three independent experiments. Statistical significance was determined by Student's t-test at a significance level of p ≤ 0.01. , and re-energized with 10% glucose, and baseline fluorescence readings were established in the first 5 min. Controls were supplemented with 10 µM carbonyl cyanide 3-chlorophenylhydrazone (CCCP) as a protonophore (triangles), and fluorescence of samples, including untreated ones (squares), was measured for an additional 5 min. The pH i of untreated TIGR4 (blue squares) and ∆speA (red squares) was 7.5 and 7.3, respectively. Then, 20 µM of nigericin was added to both treated and untreated samples to dissipate transmembrane gradients over the last 5 min. Graphs represent the mean of three independent experiments. Statistical significance was determined by Student's t-test at a significance level of p ≤ 0.01.

Effect of Deletion of SpeA on Pneumococcal Acid and Thermal Stress Responses
We observed that ∆speA is more susceptible to acid and not thermal stress. Comparison of WT, ∆speA, and pABG5-speA strains in Todd-Hewitt broth with 0.5% yeast extract (THY) at different pH values showed inhibition of growth of all strains at pH ≤ 5.5 (data shown). Comparison of the growth rate and maximum optical density of the WT, ∆speA, and pABG5-speA strains at pH 6.0 and 7.4 identified no significant difference at p ≤ 0.05. However, there Pathogens 2021, 10, 286 6 of 13 was a significant difference between the lag phase of the mutant strain compared to the WT/pABG5-speA strain at pH 5.7 (p ≤ 0.01) ( Figure 5A). ∆speA had no noticeable growth at pH 5.7, and normal growth for ∆speA was observed only at pH 6.0 and above ( Figure 5B). Pneumococci must adapt to varying temperatures at different sites in the human body. Growth of WT and ∆speA strains was comparable at 30 • C and 40 • C ( Figure S1), indicating that speA function may not be necessary for pneumococcal thermal stress responses.
We observed that ΔspeA is more susceptible to acid and not thermal stress. Comparison of WT, ΔspeA, and pABG5-speA strains in Todd-Hewitt broth with 0.5% yeast extract (THY) at different pH values showed inhibition of growth of all strains at pH ≤ 5.5 (data shown). Comparison of the growth rate and maximum optical density of the WT, ΔspeA, and pABG5-speA strains at pH 6.0 and 7.4 identified no significant difference at p ≤ 0.05. However, there was a significant difference between the lag phase of the mutant strain compared to the WT/pABG5-speA strain at pH 5.7 (p ≤ 0.01) ( Figure 5A). ΔspeA had no noticeable growth at pH 5.7, and normal growth for ΔspeA was observed only at pH 6.0 and above ( Figure 5B). Pneumococci must adapt to varying temperatures at different sites in the human body. Growth of WT and ΔspeA strains was comparable at 30 °C and 40 °C (Figure S1), indicating that speA function may not be necessary for pneumococcal thermal stress responses.

Discussion
The major finding in this study is that speA is important for pneumococcal oxidative, nitrosative, and acid but not thermal stress responses. Our results also show that glutathione and H2O2 production is dependent on speA in pneumococci, while the reverse is true for pHi. Our qRT-PCR results show that enhanced susceptibility of ΔspeA to exogenous hydrogen peroxide could be due to the deletion of speA itself and its impact on the expression of other genes involved in manganese and iron transport, polyamine biosynthesis, protein repair, and detoxification (Table 1). Polyamines are known to be involved in ROS scavenging, pH homeostasis, and regulation of antioxidant systems. Therefore, reduced polyamine synthesis that results in altered levels of intracellular polyamines is expected to render ΔspeA more susceptible to stress [4,8,9,[19][20][21][22][23][24][25][26][27]. Indeed, when we characterized the proteome and transcriptome of ΔspeA, we observed a shift in the carbon flow toward the pentose phosphate pathway [13,14], a signature of oxidative stress [16]. In this study, measurement of parameters indicative of cellular stress and rigorous analysis of the susceptibility of WT and ΔspeA strains to various stressors validated the previously reported signature of oxidative stress at the molecular level.
Although catalase negative, S. pneumoniae generates high levels of H2O2 (2 mM) via pyruvate oxidase [28,29]. Intrinsic H2O2 generation is critical for pneumococcal pathogenesis. H2O2 can inhibit or kill other common inhabitants of the respiratory tract. It is Figure 5. Growth of S. pneumoniae TIGR4, ∆speA, and pABG5-speA cultured in Todd-Hewitt broth with 0.5% yeast extract (THY) at varying pH was monitored by measuring absorbance at 600 nm and is shown in blue, red, and green, respectively, while the blank is shown in black circles in graph B. Graph (A) shows growth of the strains at pH 7.4 (squares) and 5.7 (triangles), and (B) shows growth of the strains at pH 7.4 (squares) and 6.0 (triangles). The results represent an average of three independent experiments. Statistical significance was determined by Student's t-test at a significance level of p ≤ 0.05 and p ≤ 0.01.

Discussion
The major finding in this study is that speA is important for pneumococcal oxidative, nitrosative, and acid but not thermal stress responses. Our results also show that glutathione and H 2 O 2 production is dependent on speA in pneumococci, while the reverse is true for pH i . Our qRT-PCR results show that enhanced susceptibility of ∆speA to exogenous hydrogen peroxide could be due to the deletion of speA itself and its impact on the expression of other genes involved in manganese and iron transport, polyamine biosynthesis, protein repair, and detoxification (Table 1). Polyamines are known to be involved in ROS scavenging, pH homeostasis, and regulation of antioxidant systems. Therefore, reduced polyamine synthesis that results in altered levels of intracellular polyamines is expected to render ∆speA more susceptible to stress [4,8,9,[19][20][21][22][23][24][25][26][27]. Indeed, when we characterized the proteome and transcriptome of ∆speA, we observed a shift in the carbon flow toward the pentose phosphate pathway [13,14], a signature of oxidative stress [16]. In this study, measurement of parameters indicative of cellular stress and rigorous analysis of the susceptibility of WT and ∆speA strains to various stressors validated the previously reported signature of oxidative stress at the molecular level.
Although catalase negative, S. pneumoniae generates high levels of H 2 O 2 (2 mM) via pyruvate oxidase [28,29]. Intrinsic H 2 O 2 generation is critical for pneumococcal pathogenesis. H 2 O 2 can inhibit or kill other common inhabitants of the respiratory tract. It is cytotoxic to host cells, causes apoptosis in respiratory epithelial cells, and promotes colonization of the upper respiratory tract [30][31][32]. Therefore, reduced ability to generate H 2 O 2 in ∆speA observed in this study could reduce the colonization and invasive potential of ∆speA. Increased sensitivity of ∆speA to exogenous H 2 O 2 and superoxide radicals reported here suggests a role of polyamines in pneumococcal oxidative stress responses, which is consistent with earlier reports. Putrescine and spermidine protect cells from ROS by increasing the expression of regulators or genes that encode free-radical scavengers [8,24,25,33]. Spermine and spermidine are known to scavenge ROS and work synergistically with superoxide dismutase to reduce single-stranded DNA breakages due to ROS [25,34,35].
In addition, cadaverine protects Escherichia coli and Vibrio vulnificus in high-oxygen environments and superoxide stress, respectively [20,36]. We recently reported that speA deletion resulted in significant reduction in the intracellular levels of agmatine, the product of arginine decarboxylase activity of SpeA. Therefore, this study implicates that agmatine is a critical polyamine that could regulate pneumococcal stress responses, which warrants further investigation in the future. However, superoxide susceptibility results of this study contradict our earlier report [12] where we observed similar survival rates for WT and ∆speA strains when treated with superoxide produced by paraquat. This discrepancy could be due to the difference in the compounds used to generate superoxide anions and/or the higher concentration of paraquat used. The high concentration of paraquat that we used earlier could have been equally toxic to both WT and mutant strains.
We carried out qRT-PCR with RNA-isolated pneumococci exposed to 2.5 mM H 2 O 2 for 15 min. The 15 min exposure time, which is shorter than the pneumococcal-doubling time, represents transcriptional changes due to the deficiency of speA during H 2 O 2 stress, without the confounding effects of growth and adaptation. These results show that speA affects the expression of genes involved in pneumococcal stress responses, which concurs with reports that polyamines regulate the expression of genes involved in oxidative stress responses [33]. Pneumococci deficient in tpxD and htrA are susceptible to exogenous H 2 O 2 , and ∆htrA was attenuated in murine models of nasopharyngeal colonization, pneumonia, and bacteremia [37][38][39][40][41]. Therefore, reduced expression of tpxD and htrA in ∆speA could result in reduced survival in the host, which could explain its reported in vivo attenuation [37]. Furthermore, a reduced GSH/GSSG ratio indicates impaired redox homeostasis in ∆speA, which could further render the mutant susceptible to oxidative stress. Reduced glutathione and glutathione metabolism are involved in oxidative stress responses in other bacterial pathogens [42][43][44][45][46]. Controlled influx and efflux of cellular Mn is important for normal growth and oxidative stress responses in many pathogenic bacteria [47,48]. The Fenton reaction with H 2 O 2 generates hydroxyl radicals, which is the primary cause of damage to biomolecules such as DNA. Increased expression of the iron transporter fhuD and the manganese transporter psaA could exacerbate the effects of stress in the mutant, which could impair pneumococcal survival in vivo, especially during transition between different niches. Our results agree with the reported role of polyamines in the regulation of bacterial stress responses by modifying the expression of stress response genes [33].
∆speA was more sensitive to nitrosative stress compared to the WT strain. Findings of this study agree with reports on the protective role of polyamines in nitrosative stress responses: putrescine and spermidine in Salmonella typhimurium against nitrosative stress [22] and cadaverine against nitrosative stress in E. coli [49]. The increased susceptibility of the mutant to nitrosative stress will impact pneumococcal in vivo adaptation in host immune cells such as macrophages, epithelial cells, and dendritic cells known to produce nitric oxide via inducible nitric oxide synthases for bacterial clearance [50]. We hypothesized that the increased sensitivity of ∆speA to oxidative and nitrosative stress could be due to impaired buffering capacity. However, the pH i of WT and ∆speA strains remained within the physiological pH range (Figure 4). These results conform to reports that bacteria maintain their pH i within a narrow pH range despite exposure and growth under varied extracellular pH conditions [51,52]. Although pneumococci are reported to be tolerant of pH 4.4 [51] in phagosome vesicles, our results show increased sensitivity of ∆speA to acid stress, suggesting that impaired polyamine synthesis renders pneumococci more susceptible to acid stress. During the lag phase, bacteria synthesize proteins and other molecules necessary for replication. A prolonged lag phase in the mutant at all pH values suggests that speA's indirect effects could be essential for replication. These results concur with our earlier study where we reported reduced expression of genes involved in protein and capsule synthesis in ∆speA [13,14]. These results are consistent with studies on other pathogenic bacteria that report the neutralizing effects of polyamines (cadaverine) at low pH [26,36,53,54]. Since the synthesis of polyamines via the decarboxylation of amino acids consumes a proton and generates ammonia, which protects bacteria against acid stress [54][55][56], speA deficiency could render the mutant more susceptible to low pH. Nevertheless, findings of this study contradict our earlier conclusion [12]. Compared to our earlier report, in this study, we characterized the growth of the strains over a wide range of pH (4.0-7.8), while the earlier study used only pH 5.5, which showed no significant difference in the survival of the WT strain. Results at pH 5.5 are similar with the findings of this study. Growth kinetics of WT and ∆speA strains at different temperatures indicated that deletion of speA has no noticeable impact on pneumococcal thermal stress responses (Supp. Figure S1). In S. pneumoniae D39 [7], the substrate-binding protein potD was reported to be essential for thermal stress responses. We recently reported that speA deletion resulted in significant reduction in the intracellular levels of only agmatine, the product of arginine decarboxylase. Therefore, this study indicates that agmatine is a critical polyamine that could regulate pneumococcal stress responses, which warrants further investigation in the future. Organisms alter their gene expression to limit energy-consuming processes. Inhibition of capsular polysaccharide (CPS) synthesis in ∆speA by metabolic reprogramming could be an adaptive response to counter increased intracellular oxidative and nitrosative stress coupled with impaired stress responses. This is supported by the downregulation of genes involved in carbohydrate metabolism and nucleotide synthesis in favor of the pentose phosphate pathway that generates NADPH, a cofactor for antioxidant systems [13,14].
In summary, deletion of arginine decarboxylase, an enzyme from the putrescine/spermidine biosynthesis pathway, adversely effects pneumococcal oxidative, nitrosative, and acid stress responses, which impacts its ability to survive in the human host. Expression of pneumococcal oxidative stress response genes such as tpxD, htrA, and aguA is dependent on speA, which warrants further investigation. For the first time, our data suggest that speA regulates the production of hydrogen peroxide and glutathione in pneumococci. These results once again illustrate the need for polyamine homeostasis in bacterial pathogens, the disruption of which has implications for physiology and virulence, providing insight into the potential for identifying novel therapeutic targets. However, comprehensive omics studies are needed to elucidate a crosstalk between polyamine metabolism, pneumococcal stress responses, capsule synthesis, and pathogenesis.

Bacterial Strains and Growth Conditions
S. pneumoniae serotype 4 clinical isolate TIGR4 [57], ∆speA, and pABG5-speA [14] strains were used in this study [14]. Pneumococci were grown in Todd-Hewitt broth with 0.5% yeast extract (THY) or on blood agar plates (BAP) at 37 • C in 5% CO 2 unless otherwise specified. For all stress susceptibility assays, the percentage survival of cells treated with a chemical stressor was calculated relative to untreated cultures. All assays were performed in triplicate in three independent experiments.

Hydrogen Peroxide Survival Assay
Mid-log phase cultures of TIGR4, ∆speA, and pABG5-speA (optical density at 600 nm (OD 600nm ) = 0.4-0.5) were supplemented with final concentrations of hydrogen peroxide ranging from 0.5 to 5 mM and incubated at 37 • C with 5% CO 2 for 30 min. The control reactions contained untreated bacteria. At 15 min intervals, aliquots were serially diluted in sterile phosphate-buffered saline (PBS) and plated on BAP for colony-forming unit (CFU) enumeration.

Potassium Tellurite Susceptibility Assay
Mid-log phase cultures of TIGR4, ∆speA, and pABG5-speA were centrifuged at 10,000× g for 2 min and cells suspended in PBS. Bacteria (10 8 CFU/mL) in 1 mL of PBS were exposed to 0.1 and 0.2 mM potassium tellurite (Millipore-Sigma, St. Louis, MO, USA) and incubated for 30 min at 37 • C in 5% CO 2 . The control reactions contained untreated bacteria. At 15 min intervals, aliquots were serially diluted in sterile PBS and plated on BAP for CFU enumeration.

Hydrogen Peroxide Production Assay
H 2 O 2 generated from mid-log phase cultures of (10 mL) TIGR4 and ∆speA was compared using a quantitative peroxide assay (Pierce, Thermo Fisher Scientific Waltham, MA, USA) following the manufacturer's instructions. Briefly, 1 mL of bacterial culture (10 8 CFU/mL) was centrifuged at 10,000× g at 4 • C for 2 min and the supernatant filtered with a 0.22 µM filter. The concentration of H 2 O 2 was measured in the filtrate . The assay was performed in triplicate in three independent experiments.

S-Nitrosoglutathione Susceptibility Assay
Mid-log phase cultures of TIGR4, ∆speA, and pABG5-speA were centrifuged at 10,000× g for 2 min and cells suspended in PBS. Bacteria (10 7 CFU/mL) in 100 µL of PBS were supplemented to a final concentration of 2.5 mM S-nitrosoglutathione (GSNO; Millipore-Sigma, St. Louis, MO, USA) and incubated at 37 • C in 5% CO 2 for 60 min. Control reactions contained only bacteria in PBS. At 15 min intervals, aliquots were serially diluted in sterile PBS and plated on BAP for CFU enumeration. CFUs were determined by serial dilution and plating on BAP, and results were expressed as the percentage survival of treated bacteria relative to untreated controls.

Measurement of Intracellular pH (pH i )
The intracellular pH (pH i ) was determined using the method described in [58] with slight modifications. Briefly, mid-log phase (OD600 nm 0.4) cultures of TIGR4 and ∆speA grown in THY (n = 3) were harvested, washed, and suspended in PBS. Cells (10 8 CFU) in 1 mL of the culture were loaded with 5 mM 2'-7'-bis(carboxyethyl)-5(6)-carboxyfluoresceinacetoxymethyl (BCECF-AM) dye (Millipore-Sigma, St. Louis, MO, USA) and incubated for 30 min at 30 • C in the dark. Cells were pelleted, washed, and re-energized with 10 mM glucose in PBS. To obtain an in vivo calibration curve, measured specifically for each strain, 400 µL of energized cells were pelleted and suspended in potassium buffers ranging from pH 6.5 to 8.0. Nigericin (1 mM) (Thermo Fisher Scientific, Waltham, MA, USA) was added to the samples (to equilibrate the intracellular pH of the cells to the pH of the surrounding buffer) and incubated at 37 • C for 5 min. Fluorescence was measured by a Synergy plate reader (BioTEk, Winooski, VT, USA), and a calibration curve was obtained by plotting fluorescence against the pH of the buffers. To measure the pH of individual samples, 200 µL of loaded and energized cells were added to the wells of a 96-well plate in duplicate and fluorescence was detected using a plate reader for 5 min. Next, 10 µM of carbonyl cyanide 3-chlorophenylhydrazone (CCCP) was added to one well (to serve as a control), and fluorescence was measured for another 5 min. CCCP is a protonophore that can uncouple the proton motive force and cause a sudden decrease in intracellular pH (Millipore-Sigma, St. Louis, MO, USA). Nigericin (1 mM) was added to both CCCP-treated controls and the untreated sample to equilibrate the pH i of the bacteria to the pH of the buffer, and fluorescence was measured for an additional 5 min. Fluorescence levels were calculated for each of the controls and sample, and pH i was interpolated from the calibration curve. The assays were carried out in triplicate in three independent experiments.

In Vitro Growth under Acid and Thermal Stress
To compare the growth of TIGR4, ∆speA, and pABG5-speA at varying pH, the pH of THY was adjusted using either HCl or NaOH. Growth (10 5 CFU/mL) of all strains in THY with pH adjusted between 4 and 7.5 with an interval of 0.2 units was monitored for 24 h by measuring the optical density at 600 nm (OD 600 ) using a Cytation 5 multifunction plate reader (BioTek, Winooski, VT, USA). For thermal stress, growth of all strains in THY at 30 • C, 37 • C, and 40 • C was monitored for 24 h by measuring OD 600 using a Cytation 5 multifunction plate reader.

RNA Extraction and Quantitative Real-Time PCR
Quantitative reverse transcription-PCR (qRT-PCR) was used to compare gene expression changes in ∆speA in response to H 2 O 2 stress compared to the WT strain (see Table S2 for a list of primers used in this study). To ensure amplification of a single specific product, primers were validated using melt-curve analysis with SYBR Green (Thermo Fisher Scientific Waltham, MA, USA). Total RNA was purified from cells at the mid-log phase in THY (pH 7.4) and cells exposed to 2.5 mM hydrogen peroxide for 15 min. RNA was extracted and purified using a RNeasy Midi kit and QIAcube (Qiagen, Valencia, CA, USA). RNA from H 2 O 2 -treated and control bacteria was isolated from three independent cultures. Purified RNA (7.5 ng/reaction) was reverse-transcribed into complementary DNA (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 [14]. Relative quantification of gene expression was determined using the Stratagene M × 3005P qPCR system (Agilent, Santa Clara, CA, USA). Expression of selected genes known to be involved in oxidative stress responses [40], polyamine biosynthesis, and transport was measured and normalized to the expression of gyrB and fold change determined by the comparative C T method.

Measurement of Intracellular Glutathione
The ratio of reduced and oxidized intracellular glutathione concentrations in TIGR4 and ∆speA were determined using the GSH/GSSG-Glo™ Assay Kit (Promega, Madison, WI, USA). Mid-log 10 mL cultures, n = 3 were harvested at 5000× g for 10 min at 4 • C, suspended in PBS, and transferred to beadbeater tubes (MP Biomedicals, Irvine, CA, USA). Cell suspensions were lysed with a FastPrep-24™ Classic benchtop homogenizer (45 s, 6.5 m/s × 3) (MP Biomedicals, Irvine, CA, USA) and clarified by centrifugation at 6000× g for 5 min at 4 • C. The extracts were then processed according to the manufacturer's instructions. Luminescence was measured with a Cytation™ 5 cell imaging multi-mode reader (BioTek, Winooski, VT, USA) and used to calculate glutathione concentrations. Protein concentrations of the extracts were determined with the Pierce BCA Protein Assay Kit (Thermo Fisher Scientific, Waltham, MA, USA) and used to normalize glutathione concentrations. GSH/GSSG (reduced/oxidized glutathione) ratios were calculated from the normalized glutathione concentrations according to the kit manufacturer's instructions.

Statistical Analysis
Significant differences in susceptibility between WT and deletion strains to different stressors, GSH/GSSG ratio, changes in endogenous H 2 O 2 , and changes in gene expression measured by qRT-PCR were determined by Student's t-test at a p-value of ≤ 0.05. Gene expression changes identified at p ≤ 0.05 and a fold change of 2 were considered for biological interpretation. GrowthRates [59], a software tool that uses the output of plate reader files to determine the growth rate in the exponential phase, lag phase, and maximal optical density (OD), was used to analyze growth curves of the bacterial strains used in this study.
Supplementary Materials: The following are available online at https://www.mdpi.com/2076-081 7/10/3/286/s1: Figure S1: Effect of temperature on the growth of S. pneumoniae TIGR4 and ∆speA; Table S1: Changes in gene expression between ∆speA and TIGR4 cultured in the presence or absence of 2.5 mM hydrogen peroxide that are < 2.0 folds; Table S2: Sequences of primers used in this study.