Phasevarion-Regulated Virulence in the Emerging Pediatric Pathogen Kingella kingae

ABSTRACT

Kingella kingae is a common etiological agent of pediatric osteoarticular infections. While current research has expanded our understanding of K. kingae pathogenesis, there is a paucity of knowledge about host-pathogen interactions and virulence gene regulation. Many host-adapted bacterial pathogens contain phase variable DNA methyltransferases (mod genes), which can control expression of a regulon of genes (phasevarion) through differential methylation of the genome. Here, we identify a phase variable type III mod gene in K. kingae, suggesting that phasevarions operate in this pathogen. Phylogenetic studies revealed that there are two active modK alleles in K. kingae. Proteomic analysis of secreted and surface-associated proteins, quantitative PCR, and a heat shock assay comparing the wild-type modK1 ON (i.e., in frame for expression) strain to a modK1 OFF (i.e., out of frame) strain revealed three virulence-associated genes under ModK1 control. These include the K. kingae toxin rtxA and the heat shock genes groEL and dnaK. Cytokine expression analysis showed that the interleukin-8 (IL-8), IL-1β, and tumor necrosis factor responses of THP-1 macrophages were lower in the modK1 ON strain than in the modK1::kan mutant. This suggests that the ModK1 phasevarion influences the host inflammatory response and provides the first evidence of this phase variable epigenetic mechanism of gene regulation in K. kingae.

INTRODUCTION

Kingella kingae is a Gram-negative bacterial pathogen belonging to the Neisseriaceae family. While often carried asymptomatically in the respiratory tract, K. kingae is associated with invasive infections, including bacteremia, osteoarticular infections, meningitis, lower respiratory tract infections, and keratitis. Diagnostic advances have resulted in its recognition as a leading cause of osteoarticular infections in young children (1–3). Importantly, it is a member of the HACEK group, associated with infective endocarditis. However, unlike other HACEK organisms, K. kingae-mediated infective endocarditis can progress very rapidly (4–6) and is associated with serious complications, with an overall mortality rate of 16% (1, 6). Adults with invasive infections usually have predisposing medical ailments.

Transmission occurs via close personal contact, and clusters of invasive infections have been reported among day care center attendees (7, 8). Invasive disease begins with colonization of the oropharynx and is often preceded by an upper respiratory tract infection (8, 9). The disrupted respiratory mucosa may then facilitate invasion and hematogenous dissemination to the bones, joints, and heart (10).

Currently two virulence-associated factor have been identified. The major virulence factor is RtxA, a toxin that has been shown to play a key role in virulence in an animal model of infection and allows the bacteria to breach the respiratory tract and joints (10, 11). The second is capsule that is speculated, based on the role of other encapsulated bacteria, to protect the bacterium from the host immune response and to enable mucosal colonization, survival of the organism in the bloodstream, and invasion of deep body sites (12–15). The other suggested K. kingae virulence factors, which have yet to be studied in an in vivo model of infection, include type IV pili that mediate possible initial adherence (16), KnnH, an adhesin allowing possible respiratory tract attachment (12), and outer membrane vesicles produced during infection that may offer a strategic advantage in breaching the respiratory epithelium (17).

Phase variation is a well-known mechanism aiding virulence in pathogenic Neisseria and other bacterial pathogens but has yet to be reported in this pediatric pathogen. Phase variation is the high-frequency reversible on/off switching of gene expression and is commonly mediated by mutations in simple tandem DNA repeats in the open reading frame or promoter region of genes encoding surface-expressed virulence determinants (18). The independent, random switching of these genes results in phenotypically diverse populations that can rapidly adapt to host environments and evasion of immune responses (19).

We have previously reported that phase variation of N⁶-adenosine type III DNA methyltransferases (Mod) in several clinically important bacterial pathogens (20) results in global changes in gene expression due to phase variable methylation of the genome. These phase variable regulons, or phasevarions, control expression of surface antigens and virulence factors, leading to altered phenotypes between Mod ON and OFF variants (where ON indicates in frame for expression and OFF indicates out of frame). Phasevarions have been studied in Haemophilus influenzae (ModA [21]), Neisseria gonorrhoeae and Neisseria meningitidis (ModA, ModB, and ModD [22, 23]), Helicobacter pylori (ModH [24]), and Moraxella catarrhalis (ModM [25]). The ModA, -B, -D, -H, and -M proteins have different alleles based on amino acid differences in their central DNA recognition domain (20, 22, 24–26), and each mod allele methylates different DNA sequences.

In this study, we investigated the modK gene, a phase variable DNA methyltransferase of K. kingae, to determine if it too plays a role in gene regulation.

RESULTS

Two phase variable Mod proteins in K. kingae.

To investigate whether the type III restriction-modification (R-M) system of K. kingae behaved as a phasevarion (21), we first carried out phylogenetic analysis on the mod gene associated with the type III R-M system of K. kingae. A comparison of the 4 available K. kingae genome sequences revealed that each strain contained one distinct phase variable mod gene, which we define as modK. ModK shares 70% identity with ModA of N. meningitidis (22) along the full length of the Mod deduced amino acid sequence. Like modA, the modK gene contains a tract of simple tandem tetranucleotide repeats, 5′-AGCC-3′, that mediates phase variation of modK (Fig. 1A) (22). Bioinformatic analysis of the modK gene of the 4 available genomes and 18 clinical isolates (Tables 1 and 2) indicated that modK was present in 15/18 strains. Sequence analysis of the putative DNA recognition domain (DRD) of the modK gene revealed the presence of two different ModK allele types (ModK1 and ModK2) (Fig. 1B). Of the 15 strains surveyed, 27% contain the modK1 allele and 33% contain modK2. ModK1 and ModK2 share 92 to 93% identity and 93 to 94% similarity along the length of the deduced amino acid sequence, excluding the central variable DNA recognition domain, which shares only 13 to 20% identity and 33 to 38% similarity between the alleles (Fig. 1B). A strain was defined as having the modK1 or modK2 allele when the DNA recognition domain was 95% identical at the nucleotide level to the modK gene of K. kingae strain ATCC 23330 (AFHS01000056) or KK247 (CCJT01000004) (detailed in Table 1 and Fig. 1). Our survey also revealed that 40% of isolates contained another modK allele that shared 92 to 93% identity and 93 to 94% similarity to K. kingella strain PYKK081 (27); we termed this allele modK3. Closer examination of strain PYKK081 showed a single-base-pair deletion within the DRD, resulting in a frameshift and subsequent truncation of the ModK3 protein. Strains containing the modK3 allele also contained this deletion. K. kingae strain 11220434 (28) also contains the modK gene but has a 230-bp deletion and an 11-bp difference at the 3′ end. Examination of the 42 K. kingae incomplete genomes available in NCBI revealed that 40/42 strains, including all healthy isolates, contained this inactive modK4 allele (Table 1). Three clinical isolates that previously did not amplify the modK gene were also found to contain the inactive modK allele.

FIG 1.

K. kingae modK alleles. (A) Schematic representation of the methyltransferase (modK) gene showing the 5′-(AGCC)_n-3′ DNA repeat tract responsible for phase variable expression and the variable central DNA recognition domain (DRD). (B) Predicted amino acid sequences of the DRD of ModK1 and ModK2, as well as two conserved amino acids on each side of the variable region. K. kingae strains that define the active ModK alleles are ATCC 23330 (ModK1) and KK247 (ModK2), with a strain's modK allele classified on the basis of ≥95% nucleotide identity of the DRD to one of these sequences (Table 1).

TABLE 1.

mod alleles and repeat numbers for K. kingae isolate strains^e

Strain	modK allelec	Clinical syndrome/isolation	AGCC repeat tract no.b (ON/OFF)	Sequence source	Country of isolation
ATCC 23330a	modK1	Carrier isolate	11 (OFF)	HMPREF0476_1726-HMPREF0476_1725, reference genome, GL891963.1	Norway
KK789	modK1	Blood culture	8 (OFF)	This study	MDU, Australia
KK791	modK1	Lesion chest wall, chest tissue	15 (OFF)	This study	MDU, Australia
KK002	modK1	Corneal scraping	13 (ON)	This study	Royal Brisbane Hospital, Australia
KK003	modK1	Synovial fluid	13 (ON)	This study	Royal Brisbane Hospital, Australia
KK247a	modK2	Endocarditis	19 (ON)	CCJT01000004	Israel
KK781	modK2	Joint fluid	7 (ON)	This study	MDU, Australia
KK782	modK2	Ankle fluid	14 (OFF)	This study	MDU, Australia
KK792	modK2	Fever, swollen joints, blood culture	11 (OFF)	This study	MDU, Australia
KK793	modK2	Swelling, redness, rash right heel, blood	10 (ON)	This study	MDU, Australia
KK794	modK2	Blood	5 (OFF)	This study	MDU, Australia
KK190	modK2	Endocarditis	ND	PRJEB1631	Israel
KK128	modK2	Endocarditis	ND	PRJEB1630	Israel
PYKK081a	modK3d	Septic arthritis	26 (OFF)	KKB_09104-KKB_09099-KKB_09094 (27), AJGB01000058.1	Israel
KK783	modK3d	Fever, rigors, blood culture	3 (OFF)	This study	MDU, Australia
KK784	modK3d	Febrile, nausea, vomiting, diarrhea, blood culture	2 (OFF)	This study	MDU, Australia
KK790	modK3d	Septic left hip, left hip fluid	3/4 mixed (OFF/ON)	This study	MDU, Australia
KK795	modK3d	Vomiting and diarrhea, blood	3 (OFF)	This study	MDU, Australia
KK001	modK3d	Fluid left knee	25 (ON)	This study	Royal Brisbane Hospital, Australia
KK004	modK3d	Synovial left knee	28 (ON)	This study	Royal Brisbane Hospital, Australia
11220434a	modK4d	Healthy carrier	ND	ALIJ01000087	Israel
KK83	modK4d	Osteoarticular infection	ND	PRJEB1616	Israel
KK12	modK4d	Healthy carrier	ND	PRJEB1638	Israel
KK6	modK4d	Healthy carrier	ND	PRJEB1643	Israel
KK199	modK4d	Endocarditis	ND	PRJEB1628	Israel
KK156	modK4d	Arthritis	ND	PRJEB1618	Israel
KK260	modK4d	Bacteremia	ND	PRJEB4190	Israel
Kk274	modK4d	Arthritis	ND	PRJEB1619	Israel
KK411	modK4d	Endocarditis	ND	PRJEB1629	Israel
KK101	modK4d	Osteomyelitis	ND	PRJEB1613	Israel
AA417	modK4d	Healthy carrier	ND	PRJEB1646	Israel
KK107	modK4d	Healthy carrier	ND	PRJEB1645	Israel
KK145	modK4d	Arthritis	ND	PRJEB1617	Israel
Vir5453	modK4d	Healthy carrier	ND	PRJEB1637	Israel
BB114	modK4d	Healthy carrier	ND	PRJEB1632	Israel
AA068	modK4d	Healthy carrier	ND	PRJEB1647	Israel
KK189	modK4d	Bacteremia	ND	PRJEB4188	Israel
KK75	modK4d	Arthritis	ND	PRJEB1626	Israel
KK141	modK4d	Osteomyelitis	ND	PRJEB1621	Israel
KK245	modK4d	Bacteremia	ND	PRJEB4192	Israel
KK158	modK4d	Bacteremia	ND	PRJEB4189	Israel
BB631	modK4d	Healthy carrier	ND	PRJEB1654	Israel
D7330	modK4d	Healthy carrier	ND	PRJEB1633	Israel
KK3	modK4d	Healthy carrier	ND	PRJEB1636	Israel
PV1748	modK4d	Healthy carrier	ND	PRJEB1652	Israel
KK113	modK4d	Healthy carrier	ND	PRJEB1634	Israel
CC254	modK4d	Healthy carrier	ND	PRJEB1653	Israel
KK171	modK4d	Arthritis	ND	PRJEB1625	Israel
AA105	modK4d	Healthy carrier	ND	PRJEB1639	Israel
BB016	modK4d	Healthy carrier	ND	PRJEB1648	Israel
AA255	modK4d	Healthy carrier	ND	PRJEB1640	Israel
KK97	modK4d	Bacteremia	ND	PRJEB4187	Israel
KK98	modK4d	Bacteremia	ND	PRJEB4191	Israel
BB060	modK4d	Healthy carrier	ND	PRJEB1635	Israel
KK88	modK4d	Osteomyelitis	ND	PRJEB1614	Israel
KK93	modK4d	Bacteremia	ND	PRJEB4186	Israel
D2363	modK4d	Healthy carrier	ND	PRJEB1649	Israel
KK60	modK4d	Endocarditis	ND	PRJEB1627	Israel
KK238	modK4d	Osteomyelitis	ND	PRJEB4185	USA
KK242	modK4d	Arthritis	ND	PRJEB1624	Israel
KK56	modK4d	Arthritis	ND	PRJEB1623	Israel
KK785	modK4d	Blood	16 (ON)	This study	MDU, Australia
KK788	modK4d	Left tibia abscess, abscess aspirate	12 (OFF)	This study	MDU, Australia
KK796	modK4d	Febrile, oncology blood	15 (OFF)	This study	MDU, Australia

^aGenome strains.

^bNumber and expression state of repeats within the modK gene. ON, in frame for expression; OFF, out of frame for expression. ND, not able to be determined from draft genomes.

^cA strain was defined as having the modK1, modK2, modK3, or modK4 allele if the DNA recognition region was ≥95% identical at the nucleotide level to modK1 of K. kingae strain ATCC 23330, modK2 of strain KK247, modK3 of strain PYKK081, or modK4 of K. kingae strain 11220434 (Fig. 1).

^dmodK3 and modK4 strains that contain a premature stop codon.

^eRefer to Fig. 1 and to the text.

TABLE 2.

Primers used in this study

Name	Sequence
ModK_F	ATGATAGATAGTAAAATTCAAGACCAATTAAACGAA
ModK_R	TCATTCTTTTTCTCCGTCTTGATAAAACTG
ModK_R2	TCGCCATCTTTTTTCTCCGC
ModK_DRD_F	TATCAAAGGCAATAATCTGATTGCCCTGCAT
ModK_DRD_R	AGTTGTGCCACTGCCTAAATGGTAGTC
ModKF2	TACAGCACACTTTTCAGGGCGCT
ModK_Repeat_F-6FAM	AATGGCGGACAAAGTACCGAAGA
ModK_Repeat_R	CAAAAAGTCGGTCAATTTCATCAAA
Kan_NdeI_F	GGGAATTCCATATGCATTAGGCACCCCAGGCTTTAC
Kan_NdeI_R	GGGAATTCCATATGTGAATCGCCCCATCATCCAG
RtxA_BamHI_F	GGCGCGGATCCGTAGCAGCAGGTAAAGCAACCG
RtxA_XbaI_R	GGCTAGTCTAGAATGTGCTCGCAAGATTACCAGAG
Kan_StyI_F	GGGAACCTAGGCATTAGGCACCCCAGGCTTTAC
Kan_StyI_R	GGGAACCATGGTGAATCGCCCCATCATCCAG
16SF	TCCTGGCTCAGATTGAACGC
16SR	ATTACTCGGTACGTTCCAATATATTACTC
dnakFrt	ACCGCAAAATCGCAGTTTACG
dnakRrt	GAATGTGTCGCCGTTGGTTG
groELFrt	GAATGTGTTGGCAAACGCAGTTC
groELRrt	CAGATACGCCGTCTTTGGTGAT
rtxAFrt	CAGTAGCAGCAGGTAAAGCAACCGTA
rtxARrt	AGTTCATCTGCTGCTTTTATAAGTTCGT
18SF	CGGCTACCACATCCAAGGAA
18SR	GCTGGAATTACCGCGGCT
Il-8Frt	GTTTGATACTCCCAGTCTTGTCATTG
Il-8Rrt	CTGTGGAGTTTTGGCTGTTTTAATCG
IL-βFrt	TCCCCAGCCCTTTTGTTGA
IL-βRrt	TTAGAACCAAATGTGGCCGTG
TNF-αFrt	GGAGAAGGGTGACCGACTCA
TNF-αRrt	CTGCCCAGACTCGGCAA

The length of the 5′-AGCC-3′ repeat tract region of the 18 clinical isolates was determined by GeneScan fragment length analysis using primers ModK_Repeat_F-6Fam/ModK_Repeat_R (22, 26). During this screening, modK AGCC repeat tracts varying in length from 2 to 28 residues were observed (Table 2), resulting in the modK genes being in frame (ON) or out of frame (OFF) for expression, thereby demonstrating phase variation of modK (Fig. 1A).

ModK1 switching alters surface expression of RtxA.

In order to investigate the role of modK1 in K. kingae virulence, we first created an isogenic mutant in strain KK003 by inactivating the modK1 gene by insertion of a kan^r cassette to make the KK003 modK1::kan mutant strain. A comparison of growth rates between wild-type and mutant strains did not reveal any statistically significant variations in growth patterns of these strains (data not shown).

We then carried out proteomic analyses of the secretome fractions of wild-type KK003 modK1 ON and KK003 modK1::kan strains using SDS-PAGE to identify secreted proteins that were differentially expressed. Comparison of the protein profiles revealed a band in the secretome (∼100 kDa) which was strongly expressed in the KK003 modK1 ON strain and very weakly expressed in the KK003 modK1::kan strain (Fig. 2A). The secretome protein was identified as the K. kingae RtxA toxin (10).

FIG 2.

Effects of ModK1 on the expression of secreted proteins by K. kingae modK1 ON and modK1 OFF strains. (A) Strains were grown in BHI broth. Proteins were separated by SDS-PAGE and stained with Coomassie brilliant blue R250. One protein band was observed to be higher in expression in both the KK003 and KK002 modK1 ON strains than the KK003 modK1::kan and KK002 modK1 OFF strains (indicated by asterisks). The band was excised and analyzed by tandem mass spectrometry. The identity of this protein is shown at the right of the gel. (B) Transcriptional analysis of rtxA expression. The rtxA gene expression levels of the KK002 modK1 ON strain compared to modK1 OFF strain and KK003 modK1 ON strain compared to KK003 modK1::kan mutant were quantified by qRT-PCR.

To further confirm if differences in the expression of RtxA were due to modK1 phase variation and to determine if a strain with the same DNA recognition domain as strain KK003 (modK1 allele) would result in rtxA regulation by modK1, we chose the K. kingae clinical isolate, strain KK002 (Table 1), which also contains the modK1 allele.

We isolated natural KK002 modK1 ON and modK1 OFF colonies by screening single colonies of K. kingae strain KK002 by DNA sequencing and GeneScan fragment length analysis to isolate the KK002 modK1 strain with 13 (ON) or 12 (OFF) AGCC repeats (Fig. 1A). The length of the repeat tract and the percentage of the population containing each length in each sample were determined by GeneScan fragment length analysis using primers ModK_Repeat_F-6Fam/ModK_Repeat_R (22, 26).

Comparison of the protein profiles revealed that RtxA was also strongly expressed in the KK002 modK1 ON strain and weakly expressed in the KK002 modK1 OFF strain (Fig. 2A).

Quantitative reverse transcription-PCR (qRT-PCR) comparing expression of the rtxA gene (HMPREF0476_1688) in the wild-type KK003 ON strain to that of the KK003 modK1::kan mutant strain and wild-type KK002 ON strain compared to that of the KK002 OFF strain also further verified our observation that rtxA is downregulated when modK1 is OFF (Fig. 2B).

ModK1 switching alters surface expression of DnaK and GroEL.

To identify surface-associated proteins whose expression is controlled by ModK1, we carried out proteomic analyses of wild-type KK002 modK1 ON and KK002 modK1 OFF strains using SDS-PAGE. Comparison of the outer membrane protein profiles revealed two protein bands in the surface-associated protein fraction (∼70 kDa and ∼60 kDa) highly expressed in the KK002 modK1 OFF strain but lower in expression in the KK002 modK1 ON strain (Fig. 3A). Tandem mass spectrometric analysis identified the surface-associated 70-kDa protein as DnaK and the ∼60-kDa protein as GroEL. A similar profile was observed in the wild-type KK003 modK1 ON strain and KK003 modK1::kan mutant strain (Fig. 3A), where DnaK and GroEL were found to be increased in expression in the mutant.

FIG 3.

Effect of ModK1 on the expression of surface-associated proteins. (A) KK002 modK1 ON and KK002 modK1 OFF strains were grown in BHI broth. Proteins were separated by SDS-PAGE and stained with Coomassie brilliant blue R250. Two protein bands were observed to be higher in expression in the KK002 modK1 OFF strain than in the KK002 modK1 ON strain (indicated by asterisks). The bands were excised and analyzed by tandem mass spectrometry. The identities of these proteins are shown at the right of the gels. (B) Transcriptional analysis of dnaK and groEL expression. The dnaK and groEL gene expression levels of the KK002 modK1 ON compared to the modK1 OFF strain and the KK003 modK1 ON strain compared to the KK003 modK1::kan mutant were quantified by qRT-PCR. (C) Comparison of wild-type KK002 modK1 ON strain versus KK002 modK1 OFF strain and KK003modK1 ON strain versus KK003 modK1::kan mutant in a heat shock killing assay. Cells were incubated at 46°C, and samples were taken at 10-min intervals, diluted, and plated onto HBA plates for determination of viable CFU. Comparison of KK002 modK1 ON and KK002 modK1 OFF strains revealed that the modK1 OFF strain, in which the heat shock proteins GroEL and DnaK are upregulated, is markedly more resistant to heat shock than the KK002 modK1 ON strain. Shown is a representative of three independent experiments. A Mann-Whitney test showed a significant difference at time points of 30 to 90 min (*, P = 0.0260, P = 0.0238, and P = 0.0260). Analysis of KK002 modK1 ON colonies from the 90-min point showed that 16% were phase variants in which the modK1 gene had switched from ON to OFF (i.e., from 13 to 12 AGCC repeats). Similar results were seen when the KK003modK1 ON strain was compared to the KK003 modK1::kan mutant. A Mann-Whitney test showed a significant difference at time points of 30 to 90 min (**, P = 0.0022). (D) Ratio of KK002 modK1 ON to KK002 modK1 OFF strains of the inoculum and at the 90-min time point for KK002 modK1 ON and KK002 modK1 OFF strains, indicating a selection for modK1 OFF organisms over the course of the 90-min assay. The dagger indicates that a statistically significant difference (P = 0.0177) was seen in the ON/OFF ratio between the KK002 modK1 OFF strain at the 0-min time point compared to the KK002 modK1 OFF strain at the 90-min time point.

In addition, quantitative RT-PCR on the dnaK (HMPREF0476_0227) and groEL (HMPREF0476_0560) genes, comparing the KK003 ON strain to the KK003 modK1::kan mutant strain and wild-type KK002 modK1 ON strain to the KK002 modK1 OFF strain, further confirmed that dnaK and groEL are upregulated in expression when modK1 is OFF (Fig. 3B).

ModK1-mediated phase variation of heat shock proteins.

The coordinated phase variation of the heat shock proteins DnaK and GroEL would result in the generation of two distinct bacterial populations, one most likely fitter to respond to environmental stress. To test this hypothesis, a 46°C heat shock killing assay was conducted which compared the survival of the KK002 modK1 ON strain to that of the KK002 modK1 OFF strain. The KK002 modK1 OFF strain, in which the heat shock proteins are upregulated (Fig. 3A), was observed to be more resistant to heat shock than the KK002 modK1 ON strain (Fig. 3C). A similar result was observed for the KK003 modK1::kan mutant strain (Fig. 3C).

Twenty KK002 modK1 ON survivor colonies from the 90-min time point were picked and pooled from four independent assays, and the modK1 5′-AGCC-3′ repeat region was sequenced via fragment analysis to determine whether phase variants inactivating modK1 expression (e.g., ON to OFF) had been selected for in the killing assay. This analysis revealed that 21% of colonies that had survived the heat shock assay had switched from ON to OFF (P = 0.0177) (Fig. 3D), while 97% of modK1 OFF survivors remained OFF.

ModK1 ON suppresses proinflammatory cytokine expression in THP-1 macrophages.

Upon recognition of invading pathogens, macrophages secrete proinflammatory cytokines and chemokines that are required for orchestrating innate immune defenses that limit or clear bacterial infection. To test if the ModK1 phasevarion influenced the expression of proinflammatory cytokine genes, THP-1 macrophages were infected with live K. kingae strains, the KK003 modK1 ON and KK003 modK1::kan strains. Cells from infected and uninfected wells were removed after 6 h, and human cytokine expression was quantified by real-time PCR. The expression levels of genes in treated cells were calculated based on the expression levels in untreated cells. Quantification of the transcript expression levels revealed that interleukin-8 (IL-8), IL-1β, and tumor necrosis factor (TNF) gene expression were lower when cells were infected with the KK003 modK1 ON strain than the KK003 modK1::kan strain (−3.00-, −4.25-, and −3.20-fold decrease, respectively) (Fig. 4).

FIG 4.

modK1 mutant induces significantly increased proinflammatory cytokines in macrophages. THP-1 macrophages were infected with the KK003 modK1 ON wild-type strain or KK003 modK1::kan mutant strain. Quantitative RT-PCR was used to measure transcript expression of the proinflammatory cytokines IL-8, IL-1β, and TNF. Data are expressed as means ± standard deviations (SD) from three independent experiments. P values were determined using a Mann-Whitney test (P value for IL-8, 0.0028; IL-1β, 0.0019; TNF, 0.0314).

We also measured IL-8, IL-1β, and TNF protein levels in the supernatants of THP-1 macrophages infected for either 30 min or 90 min with the wild-type KK003 modK1 ON and KK003 modK1::kan strains using the BD cytometric bead array kit. At 30 min postinfection, IL-8, IL-1β, and TNF protein levels were significantly lower in THP-1 macrophages infected with the wild-type KK003 modK1 ON strain than in THP-1 macrophages infected with the KK003 modK1::kan mutant (P values of 0.0011, 0.0047, and 0.0351, respectively) (Fig. 5). At 90 min postinfection, IL-1β and TNF protein levels were still significantly lower (P values of 0.0004 and 0.0003) in macrophages infected with the wild-type KK003 modK1 ON strain. However, there was no difference in IL-8 levels at this time point between the KK003 modK1 ON and mutant strains (P = 0.358).

FIG 5.

Cytokine release from THP-1 cells infected with K. kingae KK003 modK1 ON, KK003 modK1::kan, and KK003 rtxA::kan strains. Human macrophage THP-1 cells were infected for 30 min or 90 min with the KK003 modK1 ON, KK003 modK1::kan, and KK003 rtxA::kan mutant strains at an MOI of 50 and incubated for 6 h. Released cytokine (IL-8, IL-1β, and TNF) protein levels were quantified from the supernatants of infected THP-1 cells. IL-1β and TNF release were not detectable in the supernatants of uninfected THP-1 cells, but endogenous IL-8 was detectable in the supernatants of uninfected cells. Error bars represent the SD from the means of two independent experiments. P values were determined using Student's t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001). LDH release assays examined the effect of KK003 modK1 ON, KK003 modK1::kan, and KK003 rtx::kan strains on the cytotoxicity for THP-1 macrophages. The RTX toxin has no effect at the 30-min time point but some effect on macrophages at the 90-min time point.

Having established that modK1 switching to the ON phase leads to suppression of IL-8 and IL-1β and to TNF expression, we wanted to test if the RtxA toxin plays a role in this cytokine suppression, since RtxA is increased in expression when ModK1 is ON (Fig. 2). First we created an isogenic rtxA mutant in the K. kingae clinical isolate strain KK003 by insertion of a kan^r cassette to make the KK003 rtxA::kan mutant strain. We measured IL-8, IL-1β, and TNF protein levels in the supernatants of THP-1 macrophages infected for 30 min or 90 min with the KK003 rtxA::kan strain. At 30 min postinfection, IL-8 and TNF levels were not significantly different from those of the wild-type KK003 modK1 ON strain (P values of 0.687 and 0.501, respectively) (Fig. 5). However, IL-1β levels were found to be lower from macrophages infected with the KK003 rtxA::kan mutant than from the wild-type parent strain (P = 0.0038). IL-8, IL-1β, and TNF protein levels were lower in THP-1 macrophages infected with the KK003 rtxA::kan mutant and the KK003 modK1::kan mutant (P values of 0.0015, 0.0003, and 0.0206, respectively) (Fig. 5). At 90 min postinfection, IL-8 and TNF levels were found to be significantly higher in the KK003 rtxA::kan mutant than in the wild-type KK003 modK1 ON and KK003 modK1::kan strains (P = <0.0001). However, as seen at 30 min postinfection, IL-1β levels were significantly lower from macrophages infected with the rtxA::kan mutant than the KK003 modK1 ON strain (P = 0.0315) and the KK003 modK1::kan mutant (P = <0.0001).

To confirm that the RtxA toxin was not significantly affecting the cytokine expression results, we performed lactate dehydrogenase (LDH) release assays. We observed that the RTX toxin had no effect (0% cytotoxicity) on the KK003 modK1 ON strain and KK003 modK1::kan mutant at the 30-min time point (Fig. 5), indicating that the cytokine expression results were not toxin mediated. At 90 min a 25% cytotoxic effect was seen with the KK003 modK1 ON strain and 13% with the KK003 modK1::kan mutant. This suggests there is some impact on cytokine expression levels by the RTX toxin at this time point. Interestingly, a 2-fold difference in the effect of the toxin on macrophages was observed between the KK003 modK1 ON strain and the KK003 modK1::kan mutant.

In summary, the results suggest that phase variation of modK1 expression allows for populations of K. kingae to elicit a lower proinflammatory response. The rtxA toxin does not in itself play a role in suppression 30 min postinfection. However, 90 min postinfection the toxin appears to have a significant role in the suppression of neutrophil recruitment (IL-8) and apoptosis (TNF).

DISCUSSION

K. kingae is a commensal organism that can behave as an opportunistic pathogen under certain conditions. Understanding the processes K. kingae uses to transform from a benign member of the pharyngeal flora to an invasive pathogen of the skeletal system is critically important to the development of effective interventions and treatments. Currently four virulence factors have been identified in K. kingae, including the RtxA toxin, the Knh adhesin, type IV pili, and a polysaccharide capsule. Apart from type IV pili (29), nothing is known about their regulation. Although phase variation is known to mediate bacterial virulence and adaptation to niche environments within the host, it has yet to be reported in this emerging pediatric pathogen. Phase variation of K. kingae type IV pilus expression has been speculated (30), although the mechanism of phase variation has not been determined.

Phasevarions, or phase variable Mod proteins, control the coordinated switching of expression of multiple genes (20) and further enhance bacterial adaptability. Phasevarions are increasingly being recognized as playing an important role in host-adapted pathogens. In H. influenzae (ModA) (21), N. gonorrhoeae and N. meningitidis (ModA, ModB, and ModD) (22, 23), M. catarrhalis (ModM) (25), and H. pylori (ModH) (24), phasevarions control expression of surface antigens and virulence factors, leading to altered phenotypes between Mod ON and OFF variants. For example, the gonococcal ModA13 ON/OFF variants have distinct phenotypes for biofilm formation, resistance to antimicrobials, and survival in primary human cervical epithelial cells (22).

In this study, we have characterized the phase variable type III DNA methyltransferase ModK of K. kingae. Phase variation via repeat tracts is reported for the first time in K. kingae via the modK tetranucleotide repeat tract. Variations in the 5′-AGCC-3′ tract correlated with variable ModK1 expression. We identified four ModK alleles in K. kingae, and each allele differed extensively within the DNA recognition domain. Strains with the modK1 and modK2 alleles result in a full-length Mod (Fig. 1), suggesting that two phasevarions exist within K. kingae, with each allele regulating a different set of genes. Strains carrying either the modK3 or modK4 allele most likely lack an active phasevarion, as they contain frameshifts within their DNA recognition domain, resulting in truncated ModK proteins. The majority of strains we surveyed appeared to contain the inactive modK4 allele, suggesting that phasevarion-mediated regulation in the context of pathogenesis of invasive K. kingae disease is limited to only a small subset of K. kingae strains. However, most of these strains were from Israel, and a larger, more comprehensive global survey is required to assess the prevalence of modK1 and modK2 alleles. It is possible that the presence of either allele contributes to increased fitness, transmissibility, and/or invasiveness.

Proteomic analysis revealed that three major proteins were differentially expressed in K. kingae grown in brain heart infusion (BHI) broth when ModK1 is ON compared to OFF (Fig. 2 and 3), indicating that modK1 phase variation is involved in the epigenetic regulation of proteins and mediates the coordinated switching of a phasevarion.

The ModK1 phasevarion included the RtxA toxin, as RtxA was increased in expression when ModK1 is ON. A recent study found that RtxA plays a key role in K. kingae virulence in an infant rat model (11) and is suggested to improve the organism's chances of surviving in the host and invading skeletal tissues (10). ModK-mediated switching of the RtxA toxin may enable evasion of host immune stimulation or promote invasion and inflammation at certain sites.

The heat shock chaperones DnaK and GroEL were also found to be under ModK1 control. Transient induction of heat shock proteins is a vital protective mechanism to cope with various sources of physiological and environmental stress at the cellular level. DnaK and GroEL are surface expressed and immunoreactive in several pathogenic bacteria (31, 32). For example, DnaK, may play a direct role in the pathogenesis of H. influenzae via cell adhesion (33), and H. pylori GroEL is associated with adhesion of H. pylori to human gastric epithelial cells (34). GroEL also facilitates the adhesion of Escherichia coli to macrophages and may play a role in enhancing invasiveness (31). It is possible that modK1 phase variation randomizes expression of heat shock proteins to create a subpopulation of individuals of increased fitness that are better suited to changes within microenvironments.

In osteoarticular infections by pathogens, bone and joint damage results mainly from the inflammatory reaction elicited by the infection (35). In a mouse model of S. aureus arthritis, polymorphonuclear leukocytes and macrophages are seen in the synovial tissue early in the infection (36). High levels of TNF and IL-1β are detected in the synovial fluid of patients with bacterial arthritis (37). The proinflammatory cytokine IL-8 plays a role in acute inflammation by activating and attracting leukocytes to sites of inflammation. High levels of IL-8 are associated with severe infectious conditions, such as endocarditis (36).

Our quantitative RT-PCR cytokine expression data as well as our cytokine secretion data showed lower IL-8, IL-1β, and TNF expression when modK1 was switched ON compared to when modK1 was switched OFF, suggesting that modK1 phase varying to ON results in innate immune subversion (Fig. 4 and 5). The RTX toxin appears to play no role in cytokine expression levels at 30 min, as macrophages infected with the KK003 modK1 ON and KK003 modK1::kan mutant strains displayed no signs of cytotoxicity (Fig. 5). At 90 min postinfection, macrophages infected with these strains displayed some signs of cytotoxicity, possibly due to the RtxA toxin (Fig. 5). Microbial heat shock proteins can stimulate macrophages to produce a variety of cytokines, including TNF and IL-1β (38–40). In addition, DnaK and GroEL were found to be upregulated in expression when modK1 was OFF, and it is possible that increased expression of DnaK and GroEL stimulated macrophages to produce higher levels of proinflammatory cytokines (41).

The IL-8 and TNF levels from the KK003 rtxA::kan mutant were significantly increased compared to those of both KK003 modK1 ON and KK003 modK1::kan strains, which may relate to greater macrophage survival. Cytotoxicity was not observed with the KK003 rtxA::kan mutant at either time point. It has been suggested that RtxA plays a role in the low inflammatory response elicited by K. kingae infections in children who have mild clinical presentation (11). Hence, RtxA production being decreased if KK003 modK1 is OFF may contribute to higher levels of TNF for the modK1 mutant. Similarly, cytokine expression is greatly reduced in the presence of the E. coli RTX toxin HlyA (42).

In conclusion, we have confirmed regulation of gene expression via the random ON/OFF switching of the modK1 gene. We found that the ModK1 phasevarion is comprised of at least three genes with verified expression changes or effects in a cell culture model system. This modulation of expression could aid K. kingae adaptation to changing host environments, potentially contributing to increased fitness for transmission and/or ability to cause disease.

MATERIALS AND METHODS

Bacterial strains and growth conditions.

K. kingae strains were grown at 37°C with 5% CO₂ on horse blood agar (HBA) plates or in BHI broth supplemented with IsoVitaleX (Becton Dickinson) and Levinthal. Kanamycin (50 μg ml⁻¹) was added if required. E. coli strains DH5α and XL1-Blue (Promega) were used to propagate plasmids and were grown at 37°C in Luria-Bertani (LB) broth supplemented with either ampicillin (100 μg ml⁻¹) or kanamycin (100 μg ml⁻¹).

DNA manipulation and analysis.

All enzymes were sourced from New England BioLabs. Sequencing was performed on PCR products using a QIAquick PCR purification kit (Qiagen) and Big-Dye (PerkinElmer) sequencing kits. Data were analyzed using MacVector v11.1.2 (Accelrys).

ModK allele-specific PCR.

PCR products specific for the DNA recognition domain and repeat regions of modK were generated using the primers listed in Table 2. K. kingae clinical isolates were used as templates. The reaction was performed in 50 μl using 1× GOTaq green master mix (Promega) with the following cycling conditions to amplify the ∼2.0-kb modK gene: 30 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 2 min and 1 cycle of 72°C for 7 min with 5 μM the primer pair ModK_F and ModK_R. The DNA recognition domain was determined by sequencing the PCR product using the primers ModK_DRD_F and ModK_DRD_R and comparison to the available genome strains to determine the modK allele group. For the 3 clinical isolates that did not produce a product using the ModK_R primer, the modK gene was amplified using primer pair ModKF/ModKR₁ under the above-described conditions.

Fragment analysis.

The length of the 5′-AGCC-3′ repeat tract in modK and the percentage of each fragment length was determined by GeneScan fragment analysis (Applied Biosystems International) (22, 26). The primer pair Modk_Repeat_F-6FAM and ModK_Repeat_R (Table 2) was used to amplify the repeat region of modK. Strain KK002 modK1 ON and OFF colonies were isolated and quantified using GeneScan fragment length analysis, as described previously (22, 26). The following cycling conditions were used for the modK repeat region: 30 cycles of 94°C for 30 s, 55°C for 30 s, and 72°C for 30 s and 1 cycle of 72°C for 7 min with 5 μM the primer pair ModK_Repeat_F-6FAM and ModK_Repeat_R. PCR products were cleaned using the Wizard SV gel and PCR clean-up kit (Promega).

Generation of modK mutant strain and insertion into K. kingae strain KK003.

The modK open reading frame was amplified using PCR with primers ModK_F and ModK_R (Table 2). K. kingae strain KK003 was used as the template. The PCR product was first cloned into pCR2.1-TOPO and named pCR2.1-TOPOmodK. This plasmid then was cut with EcoRI and SacI and cloned into the corresponding sites of vector pGEM-Teasy (Promega) and named pGEMmodK. The pGEMmodK construct then was digested with EcoRI and cloned into pBlueScript KS to generate pbluescriptmodK. The Tn903 Kan resistance gene from the pUC4K vector (Pharmacia) was amplified by PCR using the primer pair Kan_NdeI_F/Kan_NdeI_R and inserted into the NdeI site of pGEMmodK. Previous work has demonstrated that the pUC4Kan kanamycin cassette has no promoter or terminator that is active in pathogenic Neisseria and will neither affect transcription nor have a polar effect on expression of adjacent genes (43). The resulting plasmid, pGEMmodK::kan, was linearized by digestion with SphI and used to transform competent K. kingae strain KK003. The resulting KK003 modK::kan transformants were selected on HBA plates containing 50 μg ml⁻¹ kanamycin. Transformants were confirmed by PCR and sequence analysis using primer ModK_F2 and kanamycin-specific primers (Table 2).

Generation of rtxA mutant strain and insertion into K. kingae strain KK003.

The modK open reading frame was amplified using PCR with primers RtxA_BamHI_F and RtxA_XbaI_R (Table 2). K. kingae strain KK003 was used as the template. The PCR product was cut with BamHI/XbaI and inserted into the BamHI/XbaI-digested vector pUC19. The construct was named pUC19rtxA. The Tn903 Kan resistance gene from the pUC4K vector (Pharmacia) was amplified by PCR using the primer pair KanStyIF/KanStyIR and inserted into the StyI site of pUC19rtxA. The resulting plasmid, pUC19rtxA::kan, was linearized by digestion with SphI and used to transform K. kingae strain KK003. KK003 rtxA::kan transformants were selected on HBA plates containing 50 μg ml⁻¹ kanamycin. Transformants were confirmed by PCR and sequence analysis using primer RtxA_BamHI_F and a kanamycin-specific primer (Table 2).

Preparation of surface-associated proteins.

K. kingae KK002 modK1 ON and OFF strains were cultivated in BHI broth overnight at 37°C, diluted to an optical density at 600 nm (OD₆₀₀) of 0.4 in 10 ml of BHI broth, and then grown at 37°C to late-log phase (OD₆₀₀ of 1.0). Primers ModK_Repeat_F-6FAM and ModK_Repeat_R were used to check the ON/OFF state of the modK1 repeat region.

Cells were harvested by centrifugation at 3,000 × g for 10 min, resuspended in 160 μl of phosphate-buffered saline (PBS; pH 7.4), vortexed at high speed for 1 min, and subsequently incubated at 60°C for 30 min with intermittent vortexing. The samples were then pelleted by centrifugation at 3,000 × g for 10 min and the supernatant was transferred to a fresh tube, where it was mixed with NuPAGE lithium dodecyl sulfate sample reducing buffer and boiled at 95°C for 5 min. The samples then were separated by SDS-PAGE using 4 to 12% Bis-Tris NuPAGE gels (Invitrogen), and the separated proteins were stained with Coomassie brilliant blue R250.

Preparation of secreted proteins.

K. kingae KK002 modK1 ON and OFF strains were cultivated in BHI broth overnight at 37°C, diluted 1:10 in 10 ml of BHI broth, and then grown at 37°C to late-log phase (OD₆₀₀ of 1.0). Primers ModK_Repeat_F-6FAM and ModK_Repeat_R were used to check the ON/OFF state of the modK1 repeat region. Cells were harvested by centrifugation (5,000 × g, 10 min), and the supernatant, containing the extracellular proteins, was passed through a 0.20-μm-pore-size filter (Sartorius). Proteins in the supernatants were precipitated with 20% (vol/vol) trichloroacetic acid on ice for 1 h, washed in 25% (vol/vol) acetone, separated by SDS-PAGE using 4 to 12% Bis-Tris NuPAGE gels, and stained with Coomassie brilliant blue R250.

Protein identification.

Tandem mass spectrometry was performed at the Walter and Eliza Hall Institute for Medical Research, Proteomics Laboratory, Melbourne, Australia.

Quantitative real-time PCR.

To verify global gene expression data, bacterial cultures of K. kingae KK002 modK1 ON, KK002 OFF, KK003 modK1 ON, and KK003 modK1::kan strains were cultivated in BHI overnight at 37°C. Triplicate cultures of a 1:100 dilution were grown to an OD₆₀₀ of 0.6 to 0.7, and 1 ml of each culture was combined and treated with RNAprotect solution (Qiagen) according to the manufacturer's instructions. About 1 μg of RNA was isolated as detailed previously (22). The RNA was further treated with DNase I (2.7 Kunitz units) for 1 h at 37°C, followed by inactivation of the enzyme with 2 μl of 25 mM EDTA and heating for 5 min at 65°C. The reaction mix was subjected to reverse transcription using the QuantiTect reverse transcriptase kit (Qiagen) by following the manufacturer's instructions. The lack of residual genomic DNA in each sample was verified prior to quantification of gene transcription. The primer pairs rtxAFrt/rtxARrt, groELFrt/groEL-Rrt, dnaKFrt/dnaKRrt, and 16SFrt/16SRrt were used to amplify rtxA, groEL, dnaK, and the16S rRNA gene, respectively. The Brilliant II SYBR green qPCR master mix was used to perform PCR in real time, as recommended by the manufacturer (Agilent Technologies). Reactions were performed in triplicate and contained 500 nM concentrations of each primer in a total volume of 20 μl. Amplification and detection of specific products were performed with a CFX96 real-time PCR detection system and a C1000 thermal cycler (Bio-Rad Laboratories), using the following protocol: 95°C for 30 s, followed by 40 cycles of 95°C for 5 s and 55°C for 20 s. The data were analyzed by using CFX Manager, version 2.0, software (Bio-Rad Laboratories). Obtained threshold cycle (C_T) values of the target genes were normalized to the 16S rRNA gene and expressed as the fold difference by calculating the 2^−ΔΔCT value.

Heat shock killing assay.

K. kingae strains were grown for 16 h at 37°C with 5% CO₂, followed by inoculation of 10 ml of BHI broth (Oxoid, Basingstoke, United Kingdom) supplemented with IsoVitaleX and Levinthal. The next day, 10 ml of BHI broth was inoculated with either the wild-type KK002 modK ON or KK002 modK OFF strain supplemented with IsoVitaleX and Levinthal and grown, with aeration, to logarithmic phase at 37°C. Heat shock treatment was performed at time intervals of 0 to 90 min by incubating 10 × 10¹⁰ cells per ml under aerobic conditions at 46°C. A sample was taken at each point, serial dilutions were carried out in BHI broth, and 10 μl of each dilution was spotted onto HBA plates in triplicate. P values for the heat shock assay were determined using a Mann-Whitney test. Twenty colonies from the KK002 modK ON and KK002 modK OFF strains at 0- and 90-min time points were isolated and pooled, and the mod repeat region was amplified by PCR using primers ModK_Repeat_F-6FAM and ModK_Repeat_R and sequenced by fragment analysis to determine changes in repeat numbers. P values were determined using a Mann-Whitney test. A Student's t test was used to determine the statistical significance between the percentage of KK002 modK1 ON and KK002 modK1 OFF strains from the KK002 modK1 ON inoculum and the percentage of KK002 modK1 ON and KK002 modK1 OFF strains from colonies from the 90-min time point in four independent assays.

THP-1 macrophage cell cultures.

Human THP-1 macrophages were grown in RPMI 1640 (GIBCO) containing l-glutamate and supplemented with 10% (vol/vol) fetal bovine serum (FBS), 50 IU/ml of penicillin, and 50 μg ml⁻¹ of streptomycin. Culture flasks were incubated at 37°C with humidity and 5% CO₂. Human THP-1 monocytes are macrophage-like monocytic cells that grow in suspension and can be differentiated into adherent macrophages using phorbol myristate acetate (PMA). Briefly, THP-1 monocytes were treated with PMA at 20 ng/million cells and incubated at 37°C with 5% CO₂ for 3 days with fresh RPMI 1640 medium containing 10% FBS. To prepare for infection, adherent THP-1 macrophages then were harvested, counted, and adjusted to 1 million cells/ml and placed in 6-well tissue culture plates.

THP-1 macrophage infection assay.

Freshly grown human THP-1 macrophages were harvested and adjusted to 1 million cells/ml without antibiotics, transferred into 6-well tissue culture plates (2 ml/well), and infected with live or heat-killed K. kingae at a multiplicity of infection (MOI) of 50 for 30 min or 90 min at 37°C with 5% CO₂. Cells then were incubated at 37°C with 5% CO₂ for 6 h in media supplemented with 50 μg ml⁻¹ gentamicin with or without 20 ng ml⁻¹ TNF (Calbiochem, EMD Biosciences, USA). Uninfected cells in duplicate wells were also incubated simultaneously and used as a no-infection control. Supernatants were harvested and saved at −20°C for determination of cytokine release, and cells were used for RNA extraction.

Cytokine measurement.

The cytometric bead array human inflammatory cytokine kit was used for the measurement of IL-8, IL-1Β, and TNF protein levels according to the manufacturer's protocol (BD Biosciences). Data were acquired on a FACSArray instrument and analyzed using cytometric bead array software, version 3.0 (BD Biosciences). The cytokine levels were expressed as picograms per milliliter. P values were determined using a Mann-Whitney test.

Cytotoxicity assays.

Freshly grown human THP-1 macrophages were harvested and seeded at 10⁶ cells/ml in 24-well tissue culture trays and treated with PMA (50 ng/ml; AdipoGen Life Sciences) to induce differentiation 48 h prior to infection. THP-1 macrophages were infected with K. kingae at an MOI of 50 for 30 min or 90 min at 37°C with 5% CO₂. Cytotoxicity was measured by LDH release using a CytoTox 96 nonradioactive cytotoxicity assay (Promega) per the manufacturer's instructions. The maximal LDH release was defined as 100% and was determined by adding the lysis solution (Cytotox One kit) to uninfected monolayers, determining the absorbance, and then subtracting the background value. The assays were performed in triplicate with biological duplicates.

RNA extraction.

Total RNA was extracted with TRIsure (Bio-line) according to the manufacturer's instructions as follows. One ml of TRIsure was added to each well and scraped by a scraper. The cells were collected in an Eppendorf tube, 0.2 ml chloroform was added, and the tube was shaken vigorously. The tube was incubated at room temperature for 5 min and then centrifuged at 12,000 × g for 15 min at 4°C. The upper layer was collected and 0.5 ml isopropyl alcohol added, and the tube was incubated at room temperature for 15 min and then centrifuged at 12,000 × g for 15 min at 4°C. The supernatant was removed and the pellet was washed once with 1 ml 70% ethanol. RNA was stored at −80°C.

Quantitative real-time PCR to measure cytokine transcript levels.

Following RNA extraction, genomic DNA was removed by DNase treatment and gDNA Wipeout buffer (Qiagen) by following the manufacturer's instructions. The extracted RNA was then reverse transcribed to cDNA using the QuantiTect reverse transcriptase kit (Qiagen) by following the manufacturer's instructions. Quantitative RT-PCR was carried out using SsoFast EvaGreen supermix (Bio-Rad) and a standard SYBR green PCR master mix (Life Technologies) protocol on the QuantStudio real-time PCR system (Applied Biosystems) according to the instructions from the respective manufacturer. The primer pairs used were IL-8Frt/IL-8Rrt, IL-βFrt/IL-βRrt, and TNF-αFrt/TNF-αRrt (Table 2). 18S was used as an internal control. Respective ΔC_T values were obtained by normalization to unstimulated cells. Relative expression was calculated with respect to the control. The results were expressed as 2^−ΔΔCT.