Group A Streptococcus Transcriptome Dynamics during Growth in Human Blood Reveals Bacterial Adaptive and Survival Strategies
2005; Elsevier BV; Volume: 166; Issue: 2 Linguagem: Inglês
10.1016/s0002-9440(10)62268-7
ISSN1525-2191
AutoresMorag Graham, Kimmo Virtaneva, Stephen F. Porcella, William T. Barry, Brian B. Gowen, Claire R. Johnson, Fred A. Wright, James M. Musser,
Tópico(s)Bacterial Identification and Susceptibility Testing
ResumoThe molecular basis for bacterial responses to host signals during natural infections is poorly understood. The gram-positive bacterial pathogen group A Streptococcus (GAS) causes human mucosal, skin, and life-threatening systemic infections. During the transition from a throat or skin infection to an invasive infection, GAS must adapt to changing environments and host factors. To better understand how GAS adapts, we used transcript profiling and functional analysis to investigate the transcriptome of a wild-type serotype M1 GAS strain in human blood. Global changes in GAS gene expression occur rapidly in response to human blood exposure. Increased transcription was observed for many genes that likely enhance bacterial survival, including those encoding superantigens and host-evasion proteins regulated by a multiple gene activator called Mga. GAS also coordinately expressed genes involved in proteolysis, transport, and catabolism of oligopeptides to obtain amino acids in this protein-rich host environment. Comparison of the transcriptome of the wild-type strain to that of an isogenic deletion mutant (ΔcovR) mutated in the two-component regulatory system designated CovR-CovS reinforced the hypothesis that CovR-CovS has an important role linking key biosynthetic, catabolic, and virulence functions during transcriptome restructuring. Taken together, the data provide crucial insights into strategies used by pathogenic bacteria for thwarting host defenses and surviving in human blood. The molecular basis for bacterial responses to host signals during natural infections is poorly understood. The gram-positive bacterial pathogen group A Streptococcus (GAS) causes human mucosal, skin, and life-threatening systemic infections. During the transition from a throat or skin infection to an invasive infection, GAS must adapt to changing environments and host factors. To better understand how GAS adapts, we used transcript profiling and functional analysis to investigate the transcriptome of a wild-type serotype M1 GAS strain in human blood. Global changes in GAS gene expression occur rapidly in response to human blood exposure. Increased transcription was observed for many genes that likely enhance bacterial survival, including those encoding superantigens and host-evasion proteins regulated by a multiple gene activator called Mga. GAS also coordinately expressed genes involved in proteolysis, transport, and catabolism of oligopeptides to obtain amino acids in this protein-rich host environment. Comparison of the transcriptome of the wild-type strain to that of an isogenic deletion mutant (ΔcovR) mutated in the two-component regulatory system designated CovR-CovS reinforced the hypothesis that CovR-CovS has an important role linking key biosynthetic, catabolic, and virulence functions during transcriptome restructuring. Taken together, the data provide crucial insights into strategies used by pathogenic bacteria for thwarting host defenses and surviving in human blood. Little is known about how pathogenic bacteria adapt to permit growth in human blood. A model organism to address this issue is group A Streptococcus (GAS), which causes a broad spectrum of human diseases ranging from relatively mild throat and skin infections to fulminant, life-threatening invasive diseases such as puerperal sepsis, myositis, necrotizing fasciitis, and streptococcal toxic shock syndrome.1Musser JM Krause RM The revival of group A streptococcal diseases, with a commentary on staphylococcal toxic shock syndrome.in: Krause RM Emerging Infections. Academic Press, New York1998: 185-218Crossref Scopus (78) Google Scholar, 2Cunningham MW Pathogenesis of group A streptococcal infections.Clin Microbiol Rev. 2000; 13: 470-511Crossref PubMed Scopus (1744) Google Scholar, 3Bisno AL Brito MO Collins CM Molecular basis of group A streptococcal virulence.Lancet Infect Dis. 2003; 3: 191-200Abstract Full Text Full Text PDF PubMed Scopus (394) Google Scholar GAS has long been known to be capable of replicating in nonopsonizing human blood.2Cunningham MW Pathogenesis of group A streptococcal infections.Clin Microbiol Rev. 2000; 13: 470-511Crossref PubMed Scopus (1744) Google Scholar However, despite years of study, the molecular mechanisms mediating GAS-host interactions remain poorly understood. Several bacterially encoded molecules contribute to GAS immune evasion by interfering with opsonophagocytosis and killing by polymorphonuclear lymphocytes.2Cunningham MW Pathogenesis of group A streptococcal infections.Clin Microbiol Rev. 2000; 13: 470-511Crossref PubMed Scopus (1744) Google Scholar, 3Bisno AL Brito MO Collins CM Molecular basis of group A streptococcal virulence.Lancet Infect Dis. 2003; 3: 191-200Abstract Full Text Full Text PDF PubMed Scopus (394) Google Scholar Others protect GAS by disrupting important innate host defenses such as complement activation and complement-mediated cell lysis.2Cunningham MW Pathogenesis of group A streptococcal infections.Clin Microbiol Rev. 2000; 13: 470-511Crossref PubMed Scopus (1744) Google Scholar, 3Bisno AL Brito MO Collins CM Molecular basis of group A streptococcal virulence.Lancet Infect Dis. 2003; 3: 191-200Abstract Full Text Full Text PDF PubMed Scopus (394) Google Scholar However, additional bacterial proteins likely are involved. Recently, Gryllos and colleagues4Gryllos I Cywes C Shearer MH Cary M Kennedy RC Wessels MR Regulation of capsule gene expression by group A streptococcus during pharyngeal colonization and invasive infection.Mol Microbiol. 2001; 42: 61-74Crossref PubMed Scopus (41) Google Scholar demonstrated that the expression of the hyaluronic acid capsule biosynthesis (has) operon is stimulated in the bloodstream of infected mice. In addition, two transcriptome studies have demonstrated GAS adaptive transcription after in vitro exposure to human polymorphonuclear lymphocytes and iron limitation.5Voyich JM Sturdevant DE Braughton KR Kobayashi SD Lei B Virtaneva K Dorward DW Musser JM DeLeo FR Genome-wide protective response used by group A streptococcus to evade destruction by human polymorphonuclear leukocytes.Proc Natl Acad Sci USA. 2003; 100: 1996-2001Crossref PubMed Scopus (134) Google Scholar, 6Smoot LM Smoot JC Graham MR Somerville GA Sturdevant DE Migliaccio CA Sylva GL Musser JM Global differential gene expression in response to growth temperature alteration in group A streptococcus.Proc Natl Acad Sci USA. 2001; 98: 10416-10421Crossref PubMed Scopus (166) Google Scholar However, no studies have assessed GAS global transcription or the regulatory networks that govern GAS adaptive responses during growth in human blood. A two-component regulatory system (TCS) designated CovR-CovS (Cov, control of virulence; also known as CsrR-CsrS) plays an important role in GAS virulence by negatively regulating the has operon and other genes encoding secreted and membrane-anchored factors that promote survival and virulence in humans.7Levin JC Wessels MR Identification of csrR/csrS, a genetic locus that regulates hyaluronic acid capsule synthesis in group A streptococcus.Mol Microbiol. 1998; 30: 209-219Crossref PubMed Scopus (231) Google Scholar, 8Federle MJ McIver KS Scott JR A response regulator that represses transcription of several virulence operons in the group A streptococcus.J Bacteriol. 1999; 181: 3649-3657Crossref PubMed Google Scholar, 9Graham MR Smoot LM Migliaccio CAL Virtaneva K Sturdevant DE Porcella SF Federle MJ Adams GJ Scott JR Musser JM Virulence control in group A streptococcus by a two-component gene regulatory system: global expression profiling and in vivo infection modeling.Proc Natl Acad Sci USA. 2002; 99: 13855-13860Crossref PubMed Scopus (304) Google Scholar, 10Lei B DeLeo FR Reid SD Voyich JM Magoun L Liu M Braughton KR Ricklefs S Hoe NP Cole RL Leong JM Musser JM Opsonophagocytosis-inhibiting Mac protein of group A streptococcus: identification and characteristics of two genetic complexes.Infect Immun. 2002; 70: 6880-6890Crossref PubMed Scopus (41) Google Scholar Isogenic Δcov mutant strains are hypervirulent in mouse skin infections and have enhanced resistance in vitro to complement-mediated opsonophagocytic killing by human polymorphonuclear lymphocytes,7Levin JC Wessels MR Identification of csrR/csrS, a genetic locus that regulates hyaluronic acid capsule synthesis in group A streptococcus.Mol Microbiol. 1998; 30: 209-219Crossref PubMed Scopus (231) Google Scholar, 8Federle MJ McIver KS Scott JR A response regulator that represses transcription of several virulence operons in the group A streptococcus.J Bacteriol. 1999; 181: 3649-3657Crossref PubMed Google Scholar, 9Graham MR Smoot LM Migliaccio CAL Virtaneva K Sturdevant DE Porcella SF Federle MJ Adams GJ Scott JR Musser JM Virulence control in group A streptococcus by a two-component gene regulatory system: global expression profiling and in vivo infection modeling.Proc Natl Acad Sci USA. 2002; 99: 13855-13860Crossref PubMed Scopus (304) Google Scholar, 11Heath A DiRita VJ Barg NL Engleberg NC A two-component regulatory system, CsrR-CsrS, represses expression of three streptococcus pyogenes virulence factors, hyaluronic acid capsule, streptolysin S, and pyrogenic exotoxin B.Infect Immun. 1999; 67: 5298-5305Crossref PubMed Google Scholar consistent with increased virulence gene transcription and extracellular capsule production. Frameshift mutations in the covRS locus also arise spontaneously in vivo, and synergistically enhance the virulence of wild-type (WT) bacteria,12Engleberg CN Heath A Miller A Rivera C DiRita VJ Spontaneous mutations in the CsrRS two-component regulatory system of Streptococcus pyogenes result in enhanced virulence in a murine model of skin and soft tissue infection.J Infect Dis. 2001; 183: 1043-1054Crossref PubMed Scopus (162) Google Scholar and hyperencapsulated GAS variants have been isolated after in vitro passage in human blood.13Raeder R Harokopakis E Hollingshead S Boyle MD Absence of SpeB production in virulent large capsular forms of group A streptococcal strain 64.Infect Immun. 2000; 68: 744-751Crossref PubMed Scopus (40) Google Scholar Taken together, these observations suggest that the CovR-CovS TCS responds to molecular signals in human blood. We directly analyzed GAS global transcription during ex vivo culture in human whole blood using a high-density oligonucleotide array. We hypothesized that the CovR-CovS TCS is involved in GAS adaptation allowing growth in blood and that virulence gene expression would be augmented. To test this hypothesis, we compared the transcriptomes of a WT, serotype M1 GAS strain and its isogenic covR-deletion mutant (ΔcovR). The data provide important new insights into the early stages of GAS survival in blood and evidence that the CovR-CovS TCS functions to coordinate bacterial fitness attributes during disseminated host infections. Serotype M1 strain MGAS5005 and the isogenic MGAS5005 ΔcovR derivative (JRS950) have been described.9Graham MR Smoot LM Migliaccio CAL Virtaneva K Sturdevant DE Porcella SF Federle MJ Adams GJ Scott JR Musser JM Virulence control in group A streptococcus by a two-component gene regulatory system: global expression profiling and in vivo infection modeling.Proc Natl Acad Sci USA. 2002; 99: 13855-13860Crossref PubMed Scopus (304) Google Scholar GAS was cultured on Trypticase soy agar containing 5% sheep blood agar (Becton Dickinson, Cockeysville, MD), or in Todd-Hewitt (TH) broth (Becton Dickinson) containing 0.2% (w/v) yeast extract (THY; Difco Laboratories, Detroit, MI), at 37°C in 5% CO2. Bacteria were grown in THY broth to late-exponential phase (OD600 = 0.8), harvested by centrifugation at 6000 × g at 37°C for 8 minutes, suspended in an equal volume of human whole blood maintained at 37°C with 5% CO2, and then incubated. Aliquots were removed at 0, 30, 60, and 90 minutes, and added to 2 vol of RNAProtect bacteria reagent (Qiagen, Valencia, CA). Cells were harvested by centrifugation and stored at −80°C before bacterial RNA isolation. Viable counts were obtained for GAS cultures immediately before time course initiation and after 4-hour co-culture in human blood. Owing to inherent interindividual and gender-related variability of human peripheral blood specimens, 12 human blood donors were used to provide generalizability and sufficient statistical power. Clinical data measurements showed subject and gender-associated variability, so that six donor patients of each gender were used. All blood donors were within normal parameters for 24 tested analytes (data not shown). Heparinized human venous blood (125-ml) was collected from the 12 healthy individuals in accordance with a protocol approved by the Institutional Review Board for Human Subjects, National Institute of Allergy and Infectious Diseases. Informed consent was obtained from all study participants. GAS clinical disease history was not assessed. Blood donors (six females, six males) were from many ethnic backgrounds and their ages ranged from 26 to 54 years; (mean age: females, 37.2 years; males, 36.2 years). Heparin was used in preference to ethylenediaminetetraacetic acid as an anti-coagulant because ethylenediaminetetraacetic acid chelates divalent cations, which would influence cellular functions during GAS-blood cell interactions. On collection, venous blood was divided into aliquots for antibody (Ab) testing (1 ml), cytokine analysis (1 ml), and blood analysis (1 ml; Alpha Veterinary Laboratories, Hamilton, MT). The remaining blood was maintained at 37°C with 5% CO2 until the time course was initiated. Bacterial cell pellets were suspended in 5 vol of EL buffer (Qiagen), incubated for 20 minutes on ice, and separated from lysed erythrocytes by centrifugation at 4500 × g at 4°C for 6 minutes. Cells were rinsed with 2 vol of EL buffer. RNA was isolated from the bacterial pellets as described,9Graham MR Smoot LM Migliaccio CAL Virtaneva K Sturdevant DE Porcella SF Federle MJ Adams GJ Scott JR Musser JM Virulence control in group A streptococcus by a two-component gene regulatory system: global expression profiling and in vivo infection modeling.Proc Natl Acad Sci USA. 2002; 99: 13855-13860Crossref PubMed Scopus (304) Google Scholar except that 0.8 μg of bacteriophage MS2 carrier RNA (Roche Bioscience, Indianapolis, IN) and 250 μg of glycogen (Roche) were added. RNA was purified further using the RNeasy 96 kit (Qiagen), with on-column RNase-free DNase I treatment and after treatment with DNAFree (Ambion, Austin, TX). Electrophoretic analysis with an Agilent 2100 Bioanalyzer (Agilent Technologies Inc., Palo Alto, CA) and A260/A280 ratios were used to assess RNA integrity. TaqMan polymerase chain reaction (PCR) assays were performed with RNA templates to ensure contaminating genomic DNA was absent as described.9Graham MR Smoot LM Migliaccio CAL Virtaneva K Sturdevant DE Porcella SF Federle MJ Adams GJ Scott JR Musser JM Virulence control in group A streptococcus by a two-component gene regulatory system: global expression profiling and in vivo infection modeling.Proc Natl Acad Sci USA. 2002; 99: 13855-13860Crossref PubMed Scopus (304) Google Scholar RML GeneChip targets were prepared according to the protocol supplied by the manufacturer (Affymetrix Inc., Santa Clara, CA), with modifications. Control spike transcript mixes (containing 0.025 to 0.000025 pmol each of DAP, LYS, THR, and TRP spike transcript cRNAs) (1 μl) were added to each RNA aliquot, and 4.5 μg of random primers (Invitrogen, Carlsbad, CA) were annealed (10 minutes at 70°C, 10 minutes at 25°C). First-strand cDNA was synthesized with 25 U/μl SuperScript III (Invitrogen) in the presence of 0.5 mmol/L dNTPs, 0.5 U/μl SUPERaseIn RNase inhibitor (Ambion), and 10 mmol/L dithiothreitol (10 minutes at 25°C, 60 minutes at 37°C, 60 minutes at 42°C, 10 minutes at 70°C). RNA was removed by hydrolysis in 1 N NaOH (30 minutes at 65°C), and neutralized with 1 N HCl before cDNA purification using the QIAquick 96 kit (Qiagen) according to the manufacturer's recommendations, except that an extra 10-minute centrifugation was used to remove trace phycoerythrin-ethanol buffer. For cDNA fragmentation, 10.5 μg of cDNA and 1.75 U (0.35U/μg) of DNase I (Roche) were used (10 minutes at 37°C, 10 minutes at 98°C). The fragmented cDNA (averaging 50 to 100 bases) was 3′ end-labeled with biotin-ddUTP using the BioArray terminal labeling kit (Enzo Life Sciences, Inc., Farmingdale, NY) (60 minutes at 37°C) according to the manufacturer's instructions. The fragmented and end-labeled cDNA was added to the hybridization solution without further purification. An anti-sense oligonucleotide array (18-μm feature size) representing ∼249,690 25-mer probe pairs (16 pairs per probe set) was manufactured by Affymetrix Inc.14Lipshutz RJ Fodor SP Gingeras TR Lockhart DJ High density synthetic oligonucleotide arrays.Nat Genet. 1999; 21: 20-24Crossref PubMed Scopus (1858) Google Scholar The custom GeneChip (RMLChip herein) contains 2636 probe sets (42,351 probe pairs) for 2636 predicted GAS open reading frames (ORFs). These features represent a composite superset of six GAS genomic sequences representative of serotypes M1, M3, M5, M12, M18, and M49 (sequenced strains are designated SF370, MGAS315, Manfredo, MGAS9429, MGAS8232, and CS101, respectively). To facilitate the analysis of GAS samples in the presence of host cells, all probe set sequences were pruned during the design process to exclude cross-hybridizing sequences (those exhibiting sequence similarity) with human, rat, and mouse genome ORFs represented on Affymetrix Inc. arrays, and 12 additional bacterial genome sequences. Although the RMLChip was not designed based on the genome sequence of strain MGAS5005, the genome sequence has since been obtained and annotated for this strain under GenBank accession no. CP000017 (Sumby PA, Madrigal A, Kent KD, Porcella SF, Ricklefs SM, Virtaneva K, Sturdevant D, Graham MR, Vuopio-Varkila J, Hoe NP, Musser JM, submitted), and the composite RMLChip contains 1692 redundant probe sets (high BLAST score match to MGAS5005) that represent more than 90% coverage of the total number of predicted coding regions (1869 ORFs) encoded by this M1 GAS genome. Target hybridizations, washing, staining, and scanning were performed by the National Institute of Allergy and Infectious Diseases Affymetrix core facility (Science Applications International Corporation (SAIC) Frederick, MD), following the manufacturer's recommendations (Affymetrix). For each of the 12 human blood donors, arrays were hybridized in a complete two-factor experimental design with two treatment levels (WT or mutant GAS strain) and four time points (0, 30, 60, and 90 minutes). To minimize experimental variability, all 12 blood samples were collected within a 2-hour time period and GAS culturing was conducted in parallel. Cultured samples were randomized before all preparation procedures were performed. Expression estimates for each gene were obtained using the PM-MM difference model of dCHIP version 1.3 software available at .15Li C Wong WH Model-based analysis of oligonucleotide arrays: expression index computation and outlier detection.Proc Natl Acad Sci USA. 2001; 98: 31-36Crossref PubMed Scopus (2702) Google Scholar The gene expression estimates were further normalized across samples by simple quadratic scaling on all genes with the median expression for each gene as a baseline.16Yoon H Liyanarachchi S Wright FA Davuluri R Lockman JC de la Chapelle A Pellegata NS Gene expression profiling of isogenic cells with different TP53 gene dosage reveals numerous genes that are affected by TP53 dosage and identifies CSPG2 as a direct target of p53.Proc Natl Acad Sci USA. 2002; 99: 15632-15637Crossref PubMed Scopus (72) Google Scholar Two-dimensional scatterplots were generated for all pairs of samples within a factor level to examine the uniformity of the normalized expression values across donors; five samples with low correlation to the other within-factor samples were removed as outliers (data not shown). Principal component analyses were performed using all MGAS5005 genome-specific probe sets (n = 1925). Hierarchical clustering also was performed to explore single gene effects. A mixed-effects analysis of variance model was applied to an absolute square root transform of the dCHIP expression estimates, with time, treatment, and gender as fixed effects, and subject as a random effect using Partek Pro 5.1 (Partek Inc., St. Louis, MO). In reporting the significance of effects, both the nominal P values and the false discovery rate (FDR) Q-value17Storey JD Tibshirani R Statistical significance for genomewide studies.Proc Natl Acad Sci USA. 2003; 100: 9440-9445Crossref PubMed Scopus (7033) Google Scholar were reported (Supplementary Table 1 ; supplementary data available at ) because it is important to account for multiple testing. The experimental design permitted differentially expressed genes to be identified with very high confidence, corresponding to FDRs of 0.06% for time, strain, and subject effects. FDR levels of 0.06% are equivalent to ∼1 false positive in a genome encoding ∼1900 ORFs (approximately the size of the GAS genome).Table 1Flow Cytometric Analysis of Group A Streptococcus Surface ProteinsDesignationProteinDescriptionMutantWTP valueSPy0319Surface lipoprotein22.43 ± 0.6015.00 ± 0.680.0001SPyM18_0281OppASurface lipoprotein23.31 ± 1.0515.03 ± 0.980.0006SPy2191SceDSecreted protein22.90 ± 0.7018.20 ± 0.500.0007SPy0453MtsASurface lipoprotein12.74 ± 0.809.60 ± 0.450.0040SPy1592Surface lipoprotein69.87 ± 5.0850.02 ± 4.200.0064SPy1245PstSSurface lipoprotein25.70 ± 0.9519.73 ± 1.970.0090SPyM3_1204SLASecreted protein13.75 ± 0.7810.58 ± 0.950.0111Auto-fluorescenceFACs control4.14 ± 0.054.29 ± 0.040.0127SPy0252Surface lipoprotein10.78 ± 0.828.93 ± 0.430.0258Secondary AbFACs control4.07 ± 0.084.15 ± 0.120.3414MGAS5005 (WT) and JRS950 (ΔcovR) cells were harvested at OD600 = 0.8 after in vitro growth in THY broth at 37°C with 5% CO2. Immunostaining was performed with affinity-purified rabbit polyclonal antibodies, or control rabbit α-SLA antibodies. GAS surface antigens were detected with a phycoerythrin-conjugated donkey anti-rabbit IgG secondary antibody, and analyzed by flow cytometry.Listed are mean fluorescence within the analysis gate ± SD from two independent experiments comprised of triplicate measurements. Minimums of 17,900 gated events (representing GAS cells) were analyzed for each replicate. Statistical significance was assessed at the P < 0.05 level after Bonferroni correction for 10 comparisons (adjusted P < 0.005). Open table in a new tab MGAS5005 (WT) and JRS950 (ΔcovR) cells were harvested at OD600 = 0.8 after in vitro growth in THY broth at 37°C with 5% CO2. Immunostaining was performed with affinity-purified rabbit polyclonal antibodies, or control rabbit α-SLA antibodies. GAS surface antigens were detected with a phycoerythrin-conjugated donkey anti-rabbit IgG secondary antibody, and analyzed by flow cytometry. Listed are mean fluorescence within the analysis gate ± SD from two independent experiments comprised of triplicate measurements. Minimums of 17,900 gated events (representing GAS cells) were analyzed for each replicate. Statistical significance was assessed at the P < 0.05 level after Bonferroni correction for 10 comparisons (adjusted P < 0.005). To elucidate the biology underlying the GAS transcriptional response, we looked for evidence that sets of genes belonging to a functional category showed a coordinated response to experimental factors. Functional annotation for the MGAS5005 genome was generated through in-house compilation. All probe sets were assigned to 1 of 17 functional categories (including unknown), and further classified into 1 of 52 subcategories. The differential expression of a functional category was assessed across both time and treatment by an approach first used in Virtaneva and colleagues,18Virtaneva K Wright FA Tanner SM Yuan B Lemon WJ Caligiuri MA Bloomfield CD de la Chapelle A Krahe R Expression profiling reveals fundamental biological differences in acute myeloid leukemia with isolated trisomy 8 and normal cytogenetics.Proc Natl Acad Sci USA. 2001; 98: 1124-1129Crossref PubMed Scopus (245) Google Scholar that is similar to recent efforts, such as Mootha and colleagues19Mootha VK Lindgren CM Eriksson KF Subramanian A Sihag S Lehar J Puigserver P Carlsson E Ridderstrale M Laurila E Houstis N Daly MJ Patterson N Mesirov JP Golub TR Tamayo P Spiegelman B Lander ES Hirschhorn JN Altshuler D Groop LC PGC-1alpha-responsive genes involved in oxidative phosphorylation are coordinately downregulated in human diabetes.Nat Genet. 2003; 34: 267-273Crossref PubMed Scopus (5955) Google Scholar For each gene, F-statistics were obtained for the time and treatment effects from a fixed-effects analysis of variance model. A two-sample Wilcoxon ranked sum statistic was then computed for the rank statistics of genes belonging to the functional category relative to the ranks of all remaining genes. Empirical P values for each functional category were obtained by recomputing Wilcoxon statistics across 10,000 permutations of the array assignments. In each permutation, the subject assignment of each array was held constant, whereas the treatment and time assignments were randomized. This approach allowed computation of permutation-based estimates for the FDR20Yekutieli D Benjamini Y Resampling based FDR controlling multiple hypotheses testing.J Stat Plan Infer. 1999; 82: 171-196Crossref Google Scholar to account for the multiple testing of functional categories. Nominal P values and FDR estimates are reported for the set of 17 categories and 52 subcategories (Supplementary Table 2). Real-time PCR assays were conducted to confirm a subset of the microarray data as described,6Smoot LM Smoot JC Graham MR Somerville GA Sturdevant DE Migliaccio CA Sylva GL Musser JM Global differential gene expression in response to growth temperature alteration in group A streptococcus.Proc Natl Acad Sci USA. 2001; 98: 10416-10421Crossref PubMed Scopus (166) Google Scholar except that Platinum Quantitative PCR SuperMix (Invitrogen) was used and each PCR reaction was performed in quadruplicate. Strains MGAS5005 and JRS950 were grown in vitro in THY broth to late-exponential phase. Bacteria were collected by centrifugation, rinsed once with Dulbecco's phosphate-buffered saline (PBS) (Sigma-Aldrich, St. Louis, MO), suspended in Dulbecco's PBS in 96-well plates, and maintained at 4°C throughout the staining procedure. Cells were blocked with 2% human serum in Dulbecco's PBS (staining buffer, SB) for 10 minutes before immunostaining for 30 minutes with GAS-specific, affinity-purified primary Abs (Bethyl Laboratories, Montgomery, TX) in SB.21Lei B Smoot LM Menning HM Voyich JM Kala SV Deleo FR Reid SD Musser JM Identification and characterization of a novel heme-associated cell surface protein made by Streptococcus pyogenes.Infect Immun. 2002; 70: 4494-4500Crossref PubMed Scopus (88) Google Scholar GAS antigens that were more highly expressed at the RNA level in the ΔcovR mutant strain were selected for the analysis. Control Abs were raised in rabbits against rSLA (spyM3_1204), an ORF that is not encoded in the GAS serotype M1 MGAS5005 genome. Detection was achieved using phycoerythrin-conjugated polyclonal donkey anti-rabbit IgG (1:500; Jackson ImmunoResearch Laboratories, Inc., West Grove, PA) in SB for 30 minutes, and subsequent flow cytometric analysis performed with a FACscalibur flow cytometer (Becton Dickinson, Mountain View, CA).21Lei B Smoot LM Menning HM Voyich JM Kala SV Deleo FR Reid SD Musser JM Identification and characterization of a novel heme-associated cell surface protein made by Streptococcus pyogenes.Infect Immun. 2002; 70: 4494-4500Crossref PubMed Scopus (88) Google Scholar To model bacteria proliferating rapidly during host sepsis, GAS cells in the late-exponential phase of growth were recultured at approximately the same cell density in freshly heparinized human whole blood for 0 to 90 minutes, and their transcribed cDNAs used to prepare microarray hybridization targets. Evaluation of scatterplots, spike-in control transcripts (data not shown), and histograms of the ex vivo expression data (Supplementary Figure 1 ) indicated high quality for the resultant data set (comprised of 91 RMLChips). The resultant principal component analyses plots and clustering dendrogram discriminated by treatment within time (Supplementary Figure 2 ).Figure 2Statistical analysis of GAS functional categories for treatment effects during ex vivo blood culture. A and B: Plots show the cumulative distributions for ranked test statistics of genes within the selected functional categories for treatment effects in blood (hatched lines). Genes differentially expressed with treatment that exhibit a leftward shift indicate up-regulation in the ΔcovR strain and those showing a rightward shift indicate down-regulation. All genes exhibiting significant differential expression at the nominal 0.05 level (t > 1.99) are shaded in the plot. The effect of treatment in blood on GAS amino acid metabolism (A) showing up-regulation in the ΔcovR strain (nominal P value <0.05; FDR = 0.14), and virulence interacting (B), showing up-regulation in the ΔcovR strain, but not in a statistically significant manner (nominal P value = 0.11; FDR = 0.69).View Large Image Figure ViewerDownload Hi-res image Download (PPT) The expression data revealed that extensive remodeling of the transcript profile occurred in both strains during ex vivo blood culture. Within 30 minutes, 716 transcripts were more abunda
Referência(s)