Curcumin Stimulates the Overexpression of Virulence Factors in Salmonella enterica Serovar Typhimurium: In Vitro and Animal Model Studies

Salmonella spp. is one of the most common food poisoning pathogens and the main cause of diarrheal diseases in humans in developing countries. The increased Salmonella resistance to antimicrobials has led to the search for new alternatives, including natural compounds such as curcumin, which has already demonstrated a bactericidal effect; however, in Gram-negatives, there is much controversy about this effect, as it is highly variable. In this study, we aimed to verify the antibacterial activity of curcumin against the Salmonella enterica serovar Typhimurium growth rate, virulence, and pathogenicity. The strain was exposed to 110, 220 or 330 µg/mL curcumin, and by complementary methods (spectrophotometric, pour plate and MTT assays), we determined its antibacterial activity. To elucidate whether curcumin regulates the expression of virulence genes, Salmonella invA, fliC and siiE genes were investigated by quantitative real-time reverse transcription (qRT-PCR). Furthermore, to explore the effect of curcumin on the pathogenesis process in vivo, a Caenorhabditis elegans infection model was employed. No antibacterial activity was observed, even at higher concentrations of curcumin. All concentrations of curcumin caused overgrowth (35–69%) and increased the pathogenicity of the bacterial strain through the overexpression of virulence factors. The latter coincided with a significant reduction in both the lifespan and survival time of C. elegans when fed with curcumin-treated bacteria. Our data provide relevant information that may support the selective antibacterial effects of curcumin to reconsider the indiscriminate use of this phytochemical, especially in outbreaks of pathogenic Gram-negative bacteria.


Introduction
Diarrheal disease is an important global health problem and is the third cause of child mortality [1]. Salmonella is one of the most frequent bacteria causing diarrheal diseases [2]. Salmonella enterica serotypes include numerous pathogens of warm-blooded animals, including humans. Salmonella enterica serovar Typhimurium (S. Typhimurium) has been considered the prototypical broad-host-range serotype. It is a frequent cause of acute selflimiting food-borne diarrhea in numerous species, including humans, livestock, domestic fowl, rodents, and birds [3]. Recently, nontyphoidal Salmonella variants were associated

PCR Identification of Genes Encoding Specific Virulence Factors of S. Typhimurium
Virulence factors are essential for the ability of bacteria to cause disease. Salmonella has the ability to survive long-term frozen storage; however, it has been reported that isolated virulent bacterial strains became avirulent during storage or passages. The above is probably due to the loss of virulence plasmids [26]. A polymerase chain reaction (PCR) method confirmed that our storage and growth conditions did not affect the presence of virulence factors from the bacterial strain. Figure 1 shows the presence of three genes of S. Typhimurium involved in epithelial cell adhesion and invasion (invA, fliC and siiE).

Curcumin Did Not Show an Antibacterial Effect.
To determine whether CUR kills S. Typhimurium and to inv ferent environmental or growth conditions on bacteria cell survi posed to dimethyl sulfoxide (DMSO) or CUR for 2 h, and the spectrophotometric (OD600), pour plate and MTT assay. The res 2, indicate that the treatment with 110, 220 and 330 µg/mL of CUR growth. After 4 h of incubation, curcumin provoked a significant stimulation. Untreated and DMSO-treated Salmonella strains growth phase with similar growth rate ( Figure 2A). After 12 h maintained a significant growth increase in the presence of CUR 110 µg/mL 3.8 ± 0.05 SD, 220 µg/mL 4.6 ± 0.06 SD and 330 µg/m 2B). By the pour plate, there was an increment in the number o lowing a dose-response profile (Table 1, Figure 3A). The percen Typhimurium increased significantly from 35% to 57% (Figure 3 sistent with the MTT assay, where formazan cell viability/crystal curcumin concentrations. When formazan was solubilized, the curcumin-treated cultures was higher than the negative control dependent profile (DSMO 0.1925 ± 0.004 vs 110 µg/mL 0.2095 ± 0.04, 330 µg/mL 0.2758 ± 0.018) (Figure 4).

Curcumin Did Not Show an Antibacterial Effect
To determine whether CUR kills S. Typhimurium and to investigate the effect of different environmental or growth conditions on bacteria cell survival, 10 7 CFU/mL was exposed to dimethyl sulfoxide (DMSO) or CUR for 2 h, and the effect was evaluated by spectrophotometric (OD600), pour plate and MTT assay. The results, presented in Figure 2, indicate that the treatment with 110, 220 and 330 µg/mL of CUR did not inhibit bacterial growth. After 4 h of incubation, curcumin provoked a significant dose-dependent growth stimulation. Untreated and DMSO-treated Salmonella strains reached the exponential growth phase with similar growth rate ( Figure 2A). After 12 h of incubation, Salmonella maintained a significant growth increase in the presence of CUR (DMSO 3.6 ± 0.09 SD vs 110 µg/mL 3.8 ± 0.05 SD, 220 µg/mL 4.6 ± 0.06 SD and 330 µg/mL 4.9 ± 0.04 SD) ( Figure 2B). By the pour plate, there was an increment in the number of colonies with CUR following a dose-response profile (Table 1, Figure 3A). The percentage of overgrowth in S. Typhimurium increased significantly from 35% to 57% ( Figure 3B). These results are consistent with the MTT assay, where formazan cell viability/crystal formation increases with curcumin concentrations. When formazan was solubilized, the absorbance at 550 nm of curcumin-treated cultures was higher than the negative controls, maintaining the dose-dependent profile (DSMO 0.1925 ± 0.004 vs 110 µg/mL 0.2095 ± 0.01, 220 µg/mL 0.2297 ± 0.04, 330 µg/mL 0.2758 ± 0.018) (Figure 4).       It has been demonstrated that CUR attenuates the virulence pathogens by the down

Virulence Factors Are Upregulated by Curcumin
It has been demonstrated that CUR attenuates the virulence pathogens by the downregulation of transcription of virulence genes [27,28]. We measured fliC, siiE and invA expression levels by relative-quantitative RT-PCR to determine whether CUR directly affected bacterial virulence. All genes showed significant gene expression changes in response to CUR treatment. In S. Typhimurium, treatment with CUR provoked the increase in mRNA expression for siiE, invA and fliC. Among the three genes, siiE had the higher range of mRNA expression (18-and 28-fold), with the concentrations of 220 and 330 µg/mL of CUR, respectively ( Figure 5A), while invA had the smallest range (0.4-and 0.9-fold), with the same doses ( Figure 5B). Finally, fliC had also significantly increased mRNA expression (5-and 10-fold) with 220 and 330 µg/mL of CUR, respectively ( Figure 5C).

Curcumin Enhanced the Pathogenicity of S. Typhimurium in C. elegans.
Earlier reports have shown that S. Typhimurium can kill C. elegans [24,29]. To validate nematode survival in the presence of this pathogenic strain, 8 × 10 8 cells/mL were used as bacterial food for C. elegans, and nematode survival was evaluated. As a negative control, nematodes fed with E. coli OP50 were used. The results obtained show that the mean and maximum life expectancy of nematodes fed with the pathogenic strain decreased significantly, with respect to worms fed with the OP50 strain (negative control) ( Figure 6A). To validate whether CUR increased the virulence of Salmonella, 8 × 10 8 cells/mL were exposed for 2 h to DMSO and 110 and 330 µg/mL of CUR, and subsequently used as nematode food. The results show that feeding C. elegans with CUR-treated bacteria significantly shortens the lifespan of the nematode by 66% at the 330 µg/mL concentration compared to DMSO treatment ( Figure 6B). The above effect correlates with the overexpression of virulence factors in S. Typhimurium due to the use of CUR. Survival curve of worms fed with DMSO-treated bacteria were not different from the curve of worms fed with untreated bacteria ( Figure 6, LogRank p = 0.968 for S. Typhimurium, n = 90). OP50-fed worms

Curcumin Enhanced the Pathogenicity of S. Typhimurium in C. elegans
Earlier reports have shown that S. Typhimurium can kill C. elegans [24,29]. To validate nematode survival in the presence of this pathogenic strain, 8 × 10 8 cells/mL were used as bacterial food for C. elegans, and nematode survival was evaluated. As a negative control, nematodes fed with E. coli OP50 were used. The results obtained show that the mean and maximum life expectancy of nematodes fed with the pathogenic strain decreased significantly, with respect to worms fed with the OP50 strain (negative control) ( Figure 6A). To validate whether CUR increased the virulence of Salmonella, 8 × 10 8 cells/mL were exposed for 2 h to DMSO and 110 and 330 µg/mL of CUR, and subsequently used as nematode food. The results show that feeding C. elegans with CUR-treated bacteria significantly shortens the lifespan of the nematode by 66% at the 330 µg/mL concentration compared to DMSO treatment ( Figure 6B). The above effect correlates with the overexpression of virulence factors in S. Typhimurium due to the use of CUR. Survival curve of worms fed with DMSO-treated bacteria were not different from the curve of worms fed with untreated bacteria (Figure 6, LogRank p = 0.968 for S. Typhimurium, n = 90). OP50-fed worms showed the usual lifespan, ranging from 16 to 22 days.

Discussion
Diarrheal diseases are one of the leading causes of death in children under 5 years and adults over 65 years. After Rotavirus, enteric bacteria are an important cause of morbidity and mortality. S. Typhimurium is included among the major isolated agents in developing countries [1,30]. Antibiotics are effective in life-threatening cases caused by bacterial pathogens; however, due to increased resistance and the potentially serious side effects of combinatorial therapies, there is a pressing need to have new alternatives [31][32][33][34]. In recent years, CUR, the principal and most active curcuminoid of Curcuma longa L. (C. longa), has gained considerable attention, due to its antimicrobial activity in different strains of bacteria. In 2016, Hayati Gunes et al [19] found that CUR has high antibacterial activity against E. coli, in relation to other bacteria, with a minimum inhibitory concentration (MIC) for CUR of 163 µg/mL. Others found that in combination with antibiotics, the CUR antibacterial activity ranges from 125 to 500 µg/mL [35]. Although most studies suggest that CUR has activity against both Gram-positive and Gram-negative bacteria [11][12][13][16][17][18][19][20][21], its activity against S. Typhimurium is considerably controversial. Meanwhile, in chicken, treatment with C. longa prevents intestinal colonization by S. Typhimurium [36]. In a murine model, CUR increases the pathogenicity of this bacteria [23]. In addition, reports are showing that the efficacy of antibiotics is directly related to the level of inoculum size. Bacteria might appear susceptible when the inoculum is low density (10 5 CFU/mL) but resistant if the inoculum size is increased (high density ~10 9 CFU/mL; depending on the clinical strains) [37][38][39]. In this report, we explore the antibacterial efficacy of CUR, and the killing assay was performed using 10 7 CFU/mL. In the present study, even though we followed the procedure reported by Hayati Gunes et al. [19], CUR at 110, 220 and 330 µg/mL for 16-18 h at 37 °C, 250 rpm, was not active against S. Typhimurium

Discussion
Diarrheal diseases are one of the leading causes of death in children under 5 years and adults over 65 years. After Rotavirus, enteric bacteria are an important cause of morbidity and mortality. S. Typhimurium is included among the major isolated agents in developing countries [1,30]. Antibiotics are effective in life-threatening cases caused by bacterial pathogens; however, due to increased resistance and the potentially serious side effects of combinatorial therapies, there is a pressing need to have new alternatives [31][32][33][34]. In recent years, CUR, the principal and most active curcuminoid of Curcuma longa L. (C. longa), has gained considerable attention, due to its antimicrobial activity in different strains of bacteria. In 2016, Hayati Gunes et al [19] found that CUR has high antibacterial activity against E. coli, in relation to other bacteria, with a minimum inhibitory concentration (MIC) for CUR of 163 µg/mL. Others found that in combination with antibiotics, the CUR antibacterial activity ranges from 125 to 500 µg/mL [35]. Although most studies suggest that CUR has activity against both Gram-positive and Gram-negative bacteria [11][12][13][16][17][18][19][20][21], its activity against S. Typhimurium is considerably controversial. Meanwhile, in chicken, treatment with C. longa prevents intestinal colonization by S. Typhimurium [36]. In a murine model, CUR increases the pathogenicity of this bacteria [23]. In addition, reports are showing that the efficacy of antibiotics is directly related to the level of inoculum size. Bacteria might appear susceptible when the inoculum is low density (10 5 CFU/mL) but resistant if the inoculum size is increased (high density~10 9 CFU/mL; depending on the clinical strains) [37][38][39]. In this report, we explore the antibacterial efficacy of CUR, and the killing assay was performed using 10 7 CFU/mL. In the present study, even though we followed the procedure reported by Hayati Gunes et al. [19], CUR at 110, 220 and 330 µg/mL for 16-18 h at 37 • C, 250 rpm, was not active against S. Typhimurium (data not shown). It is well documented that the bacterial growth rate determines the bacterial susceptibility to antimicrobials; bacterial overgrowth provokes the nutrients deprivation that induces modifications of the cell envelope [40], and generally, there is no correlation with the antimicrobial concentration. Considering this, CUR treatment was performed throughout each phase of growth. Our results provide evidence that CUR did not inhibit the growth of Salmonella, but promoted a significant overgrowth instead, after 4 h of incubation. Many bacterial species and antibiotic classes exhibit heteroresistance, meaning that a susceptible bacterial isolate harbors a resistant subpopulation that can grow in the presence of an antibiotic. In this work, CUR was in contact with Salmonella for only 2 h, after which it was removed, but, interestingly, in Salmonella, the overgrowth continued after 12 h of incubation. This suggests that CUR enhances the speed at which cells proliferate and that the modification is transmitted to new generations. It has been reported that S. Typhimurium has some genes with diverged expression domains that are involved in different metabolic pathways compared with E. coli, leading to their better survival and propagation [41][42][43]. More studies are necessary to identify how CUR regulates the growth in Salmonella. On the other hand, the standard optical method for quantifying cell density (OD 600 nm) cannot distinguish live from dead bacteria or even particles. Therefore, in order to improve the results, viability and metabolic activity was validated by the pour plate method and MTT assay [44][45][46]. Our results confirm that, after 12 h of incubation, CUR does not affect the growth of S. Typhimurium; we have metabolic active growing cells.
CUR has been found to modulate the activity of several key transcription factors and, in turn, the cellular expression profiles [47]. In bacteria and parasites, CUR is reported to modulate the virulence factor expression [27,28]. The pathogenicity of bacteria is related to many and strain-specific virulence factors. In Salmonella, we analyzed three virulence genes, invA for the Salmonella genus, fliC, and siiE for Typhimurium serovar. Our result showed that all genes were found to be upregulated by CUR. For invA, a gene that mediates invasiveness, the overexpression was only 0.9-fold. The higher overexpression levels were observed with fliC and siiE (10-and 28-fold, respectively). In another Salmonella species, flagella could be dispensable for host cell adhesion, but for S. Typhimurium, the flagellum is a key virulence-associated phenotype. A functional flagellum is necessary for epithelial cell invasion and macrophage uptake; besides, it participates in proinflammatory cytokine expression. In the case of the adhesin SIIE, some studies show that the infection of host organisms by Salmonella involves the cooperative activity of the Salmonella pathogenicity island 1 (SPI1)-encoded type III secretion system (T3SS) and SIIE. Without the function of the SPI4 T1SS or SiiE, Salmonella is highly reduced in adhesion [7,48,49]. Our results suggest that CUR enhances the adhesion ability of Salmonella. Further studies are needed to elucidate the exact mechanism by which CUR upregulates the expression of the major virulence factors of S. Typhimurium [23,50,51]. The higher pathogenic potential, which bacterial strains exposed to CUR possess, was validated using the nematode C. elegans. It is known that S. Typhimurium is pathogenic to C. elegans; even though the nematode expresses numerous antimicrobial proteins, this bacterial strain proliferates and establishes a persistent infection in the intestine of the nematode [52]. In this study, the overexpression of virulence factors in Salmonella by CUR correlates with the short lifespans of C. elegans in a lifespan assay. The rate of mortality of C. elegans fed with untreated S. Typhimurium was similar to that found by other authors [25]; the life expectancy of the nematode was reduced by 66%, in comparison to the E. coli OP50 strain. In nematodes fed with CUR-treated S. Typhimurium, there was a direct correlation between the overexpression of virulence genes and the mortality rate; the complete mortality occurred after 10 and 7 days with 110 and 330 µg/mL, respectively, suggesting increased bacterial infection after exposure to CUR. In contrast, with other Gram-negative bacteria, it has been reported that CUR reduced the production of virulence factors, affecting the adherence and the formation of biofilm [28]. It is important to emphasize that further studies are necessary.

Maintenance and Preservation of Microorganisms
The bacterial strain was grown in nutritive agar (plates) (Becton Dickinson, Maryland, USA) at 37 • C for 18-20 h. The cultures were stored at 4 • C, with streak plating onto fresh agar plates every seven days. A glycerol stock of bacteria was stored at −80 • C.

Extraction of Genomic DNA
Genomic DNA was obtained from S. Typhimurium cultures using the DNeasy ® Blood & Tissue kit (QIAGEN, Hilden, Germany), following the manufacturer's instructions. The DNA was stored at −20 • C. Purity and concentration were determined by 1% agarose (Ultra-Pure-Agarose, Invitrogen, Carlsbad, CA, USA) gel electrophoresis and by spectrophotometry, respectively. Electrophoretic gels were stained with GelRed (Nucleic Acid Gel, Biotium, Landing Pkwy, CA, USA) and visualized on a trans-illuminator (UVP Benchtop 2UV, Fisher Scientific, Waltham, MA, USA).

Presence of Virulence Genes
The pathogenicity of Salmonella spp. has been related to numerous virulence genes. The invasion protein InvA is one of the most studied virulence factors. Flic-encoded flagellin protein, and the giant, non-fimbrial adhesin protein SIIE have been also implicated in successful host infection [10,54,55]. The expression of invA (GenBank Accession M90846.1), fliC (GenBank Accession KF589316.1) and siiE (GenBank Accession AJ576316.1) was validated in the bacterial strain. A specific region of each gene was amplified from genomic DNA (DNeasy ® Blood & Tissue, IAGEN) by PCR using the following primers: siiE sense

Spectrophotometric Method
The antibacterial activity of CUR was determined by a growing strain in Luria Bertani broth (LB) (Sigma-Aldrich, Missouri, USA), at 37 • C, 250 rpm. Cultures were allowed to grow until they reached OD600 0.08 (10 7 colony forming units CFU/mL). Cells were pelleted by centrifugation at 1844× g for 5 min (Sigma 1-14K 12092 rotor), resuspended in 3 mL of PBS containing 110, 220 and 330 µg/mL of CUR and incubated for 2 h at 37 • C, 250 rpm. Untreated and 1.2% DMSO-treated cultures were used as negative controls. After the incubation period, cells were harvested by centrifugation, washed with PBS twice to remove the CUR, and grown in LB medium to achieve the exponential phase [18]. Bacterial growth (OD600) in LB medium was measured on a microplate reader (BioTek Synergy HT, Winooski, VT, USA). All experiments were performed in triplicate.

Pour Plate Method
S. Typhimurium strain was exposed to DMSO, 110, 220 or 330 µg/mL of CUR for 2 h following the procedure described above. After CUR treatment, cells were harvested by centrifugation as mentioned above. The pellets were resuspended in 3 mL of PBS, and serial dilutions were performed. For the pour plate method, 100 µL of each dilution was added by pipette to the center of sterile disposable Petri dishes. Then, cooled but still molten agar medium was poured into each Petri dish. The plates were incubated overnight at 37 • C. The dilutions chosen produced between 30 and 300 separate countable colonies. The growth percentage was calculated as ((B2-A2)/A2 100)), where A is the number of colonies untreated, and B is the number of colonies in the presence of CUR. All experiments were performed in triplicate.

Assay MTT
According to previous reports, MTT assay modified by Wang et al. [44] was performed to determine the viability of S. Typhimurium after DMSO, 0, 110, 220 or 330 µg/mL of CUR exposition. Briefly, a bacterial strain was grown at 37 • C in LB broth until the OD600 reached 0.1, and then DMSO or CUR was added to each cell culture. After incubation for 2 h at 37 • C at 250 rpm, cultures were centrifuged at 1844× g for 5 min (Sigma 1-14K 12092 rotor). The resulting bacterial pellets were washed three times in PBS and resuspended in 1 mL of LB. Aliquots of the bacterial cultures (20 µL) were placed on 0.6 mL tubes, which had been preheated to 37 • C for 10 min. Then, 2 µL of MTT (5 mg/mL, Sigma-Aldrich, M5655 St Louis, MO, USA) was added to each tube. After incubation, for 20 min at 37 • C, the tubes were centrifuged at 10,000× g for 1 min (Sigma 1-14K 12092 rotor) in order to precipitate the bacteria and formazan crystals; 20 µL of the medium was removed, and the crystals were dissolved with 250 µL of DMSO. Finally, the coloration was read at 550 nm after 15 min in a microplate reader (BioTek Synergy HT, Winooski, VT, USA). All experiments were performed in triplicate.

Statistical Analysis
All data were presented as mean values with standard deviations and analyzed using two-way ANOVA, followed by Dunnett's multiple comparisons test (GraphPad Prism version 6.01 for Windows, GraphPad Software, La Jolla, CA, USA). p-values of ≤0.05 were considered significantly different.

Relative-Quantitative RT-PCR
The effect of CUR on the expression of invA, fliC and siiE, genes associated with the virulence of S. Typhimurium, was evaluated by semi-quantitative qRT-PCR using the primers described above. First, the bacteria strain was exposed to DMSO, 110, 220 or 330 µg/mL of CUR for 2 h following the procedure described above. After CUR remotion, they were grown overnight in LB medium. Total RNA was obtained from DMSO, or CUR treated bacterial cultures using a Total RNA Purification kit (NORGEN), following the manufacturer's instructions. cDNAs were synthesized by a reverse transcriptase reaction (Verso cDNA Synthesis Kit, Thermo Scientific) using 1 µg of RNA and Oligo dt20 primer (Integrated DNA). Relative-quantitative RT-PCR was performed in a StepOneTM Real-Time PCR System (Applied Biosystems TM , Foster City, CA, USA) using Maxima SYBR Green qPCR Master Mix (Thermo Scientific) to evaluate the amplification reaction. The gene expression was normalized to the expression level of glyceraldehyde 3-phosphate dehydrogenase genes (GenBank accession no. DQ644683.1) using the following primers: gapdh sense 5 -GGT TTT GGC CGT ATC GGT CGC A-3 and gapdh antisense 5 -ACC GGT AGC TTC AGC CAC TAC G-3 . Melting curves confirmed the absence of primer dimerization. The amplification conditions were as follows: hot start at 95 • C 10 min, 40 cycles of 95 • C 15 s, 60 • C 30 s and 72 • C 30 s. The comparative ∆∆Ct method calculated changes in expression [57]. Significant differences (defined as p < 0.05, indicated by asterisks in figures) were calculated by ANOVA tests using the GraphPad Prism version 6.01 for Windows (GraphPad Software, La Jolla, CA, USA). Error bars indicate standard deviations for experiments with more than one trial.

Maintenance and Preservation of C. elegans
The wild-type C. elegans variety Bristol N2 strain used in this study was provided by the Caenorhabditis Genetics Center (CGC, Minneapolis, MN, USA). Adult worms were used for all experiments, age-synchronized according to standard methods. Nematodes were grown and maintained monoxenically at 20 • C, on nematode growth medium (NGM) [58]. Animals were grown on Petri plates seeded with Escherichia coli strain OP50 as a food source [59,60].

Pathogenicity Assays of Salmonella Strains on the C. elegans Model
A pathogenicity assay is a lifespan assay, where C. elegans strains on NGM agar plates are fed with pathogenic bacteria (known here as killing plates) instead of the regular E. coli OP50 [29]. In this study, S. Typhimurium was grown overnight in LB at 37 • C, 250 rpm and then resuspended at an OD600 = 1. Then, cells were harvested by centrifugation, and the pellets were resuspended in 3 mL of PBS containing DMSO, 110 or 330 µg/mL of CUR and incubated for 2 h at 37 • C, 250 rpm. The killing plates were prepared by dropping 10 µL of bacterial suspension (8 × 10 8 CFU/mL) onto NGM agar plates and incubated for 16 h at room temperature [61]. After that, 30 age-synchronized L4 worms were transferred to the killing plates. Since C. elegans starts laying eggs on day 1 of adulthood, the worms were transferred to fresh killing plates from the 2nd to 14th days to prevent a mistaken offspring score. In the remaining days of the trial, the transfers were carried out within a longer time (when food ran out) because the worms were in the non-reproductive phase. Worms that were alive, dead, or missing were determined and counted every other day along the time course of dying for the population, using a touch movement assay for death [62]. This assay consists of visually inspecting the worm for movement; if there is movement, then it scored as alive, and if there is no movement, even when its body is gently touched with a worm picker, it is scored as dead. Missing worms, those lost or burrowing into the medium or climbing the plate walls and drying up, were censored from the analysis. NGM agar plates seeded with untreated S. Typhimurium and E. coli OP50 were used as positive and negative controls. Every experiment was repeated three times. Data were analyzed using the Kaplan-Meier survival test and weighted log-rank tests [63]. Actual P-values are included in the figures; asterisks indicate significant differences. Differences were considered significant at p < 0.05.

Conclusions
This research provides new evidence on the antibacterial activity of CUR against one of the major enteric bacterial pathogens, Salmonella. We demonstrated that CUR increases the cell proliferation of S. Typhimurium, and we also observed a deregulation of three genes involved in the pathogenicity of S. Typhimurium leading to an increase in virulence in agreement with the results of in vivo assays. These results urge us to reconsider the indiscriminate use of CUR, especially in outbreaks of pathogenic Gram-negative bacteria.