BMJ Paediatrics Open is committed to open peer review. As ... · Confidential: For Review Only 2 33 Abstract 34 Introduction: Invasive pneumococcal disease (IPD), caused by Streptococcus

BMJ Paediatrics Open is committed to open peer review. As part of this commitment we make the peer review history of every article we publish publicly available. When an article is published we post the peer reviewers’ comments and the authors’ responses online. We also post the versions of the paper that were used during peer review. These are the versions that the peer review comments apply to. The versions of the paper that follow are the versions that were submitted during the peer review process. They are not the versions of record or the final published versions. They should not be cited or distributed as the published version of this manuscript. BMJ Paediatrics Open is an open access journal and the full, final, typeset and author-corrected version of record of the manuscript is available on our site with no access controls, subscription charges or pay-per-view fees (http://bmjpaedsopen.bmj.com). If you have any questions on BMJ Paediatrics Open’s open peer review process please email

[email protected]

on June 3, 2020 by guest. Protected by copyright.

http://bmjpaedsopen.bm

j.com/

bmjpo: first published as 10.1136/bm

jpo-2017-000092 on 31 August 2017. D

ownloaded from

http://bmjpaedsopen.bmj.com/

mailto:[email protected]


Confidential: For Review O

nly

Peripheral blood RNA gene expression in children with

pneumococcal meningitis provides insights into host

response mechanisms

Journal: BMJ Paediatrics Open

Manuscript ID bmjpo-2017-000092

Article Type: Original article

Date Submitted by the Author: 23-May-2017

Complete List of Authors: Kulohoma, Benard ; Centre for Biotechnology and Bioinformatics,

University of Nairobi, Nairobi, Kenya;; University of Oxford, Wellcome Trust Centre for Human Genetics, Nuffield Department of Medicine, Marriage, Fiona; 3. Centre for Integrated Genomic Research, University of Manchester, Manchester, UK. Vasieva, Olga; University of Liverpool, Institute of Integrative Biology Nguyen, Kha; University of Liverpool, Institute of Translational Medicine Mankhambo, Limangeni; 2. Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of Medicine, Blantyre, Malawi. Molyneux, Malcolm; 2. Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of Medicine, Blantyre, Malawi Molyneux, ELizabeth; Paediatrics Day, Philip; 3. Centre for Integrated Genomic Research, University of

Manchester, Manchester, UK. Carrol, Enitan; University of Liverpool, Institute of Infection and Global Health

Keywords: HIV, Immunology, Infectious Diseases, Microbiology, Tropical Paediatrics

https://mc.manuscriptcentral.com/bmjpo

BMJ Paediatrics Open on June 3, 2020 by guest. P

rotected by copyright.http://bm

jpaedsopen.bmj.com

/bm

jpo: first published as 10.1136/bmjpo-2017-000092 on 31 A

ugust 2017. Dow

nloaded from



nly

1

Title: Peripheral blood RNA gene expression in children with 1

pneumococcal meningitis provide insights into host response 2

mechanisms 3

Benard W. Kulohoma 1 † *, Fiona Marriage 3 *, Olga Vasieva 4, Limangeni 4

Mankhambo 2, Kha Nguyen 5, Malcolm E Molyneux 2, Elizabeth M Molyneux 6, 5

Philip JR Day 3, Enitan D Carrol 2 5 6 ‡ 6

7

1. Centre for Biotechnology and Bioinformatics, University of Nairobi. 8

2. Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of 9

Medicine, Blantyre, Malawi. 10

3. Centre for Integrated Genomic Research, University of Manchester, 11

Manchester, UK. 12

4. Institute of Integrative Biology, University of Liverpool, Liverpool, UK. 13

5. Institute of Infection and Global Health, University of Liverpool, Liverpool, 14

UK. 15

6. Department of Paediatrics, University of Malawi College of Medicine, 16

Blantyre, Malawi. 17

18

*Contributed equally. 19

† Current address: Wellcome Trust Centre for Human Genetics, University of 20

Oxford. 21

‡ Corresponding author 22

23

Benard W. Kulohoma ([email protected]) 24

Fiona Marriage ([email protected]) 25

Olga Vasieva ([email protected] ) 26

Limangeni Mankhambo ([email protected]) 27

Kha Nguyen ([email protected]) 28

Malcolm Molyneux ([email protected]) 29

Elizabeth Molyneux ([email protected]) 30

Philip J. R. Day ([email protected]) 31

Enitan D. Carrol ([email protected]) - Corresponding author 32

Page 1 of 39


BMJ Paediatrics Open

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

2

Abstract 33

Introduction: Invasive pneumococcal disease (IPD), caused by Streptococcus 34

pneumoniae, is a leading cause of pneumonia, meningitis and septicaemia 35

worldwide. Morbidity and mortality in pneumococcal meningitis are significantly 36

increased among children with underlying HIV infection. 37

Objectives: We hypothesised that peripheral blood expression profiles from 38

HIV-infected and un-infected children with pneumococcal meningitis would 39

provide insight into the host inflammatory response leading to poorer outcomes. 40

Design and setting: Prospective case control observational study in a tertiary 41

hospital in Malawi 42

Participants: Children up to the age of 16 years with a presumptive diagnosis 43

of meningitis or pneumonia. For the microarray analysis, we studied children 44

with pneumococcal meningitis (n=12) and matched HIV-uninfected controls 45

(n=3). For the real time qPCR, we studied 229 cases and 13 controls with IPD 46

Methods: We used the human genome HGU133A Affymetrix array to explore 47

differences in gene expression between cases and controls and between HIV-48

infected and uninfected cases, and validated gene expression profiles for 34 49

genes using real time qPCR in an independent set of cases and controls. 50

Pathway analysis was used to explore genes differentially expressed. 51

Results: Irrespective of underlying HIV infection, cases showed significant up-52

regulation of: S100 calcium binding protein A12 (S100A12); vanin-1 (VNN1); 53

arginase, liver (ARG1); matrix metallopeptidase 9 (MMP9); annexin A3 54

(ANXA3); interleukin 1 receptor, type II (IL1R2); CD177 molecule (CD177); 55

S100 calcium binding protein A9 (S100A9), cytoskeleton-associated protein 4 56

Page 2 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

3

(CKAP4); and glycogenin 1 (GYG1). The RT-PCR results were in agreement 57

with our microarray analysis findings. There was no differential gene expression 58

in HIV-infected compared to HIV-uninfected cases, but there was significant up-59

regulation of folate receptor 3 (FOLR3), endocytic adaptor protein (NUMB) and 60

S100A12 in survivors compared non-survivors. 61

Conclusion: Children with pneumococcal meningitis demonstrated increased 62

expression in genes regulating immune activation, oxidative stress, leucocyte 63

adhesion and migration, arginine metabolism, and glucocorticoid receptor 64

signalling. 65

66

(298 words) 67

68

Key words: Differential expression, children, invasive pneumococcal disease, 69

host response, pneumococcal meningitis 70

71

Key messages: 72

73

What is known about the subject 74

Invasive pneumococcal disease (IPD) caused by Streptococcus pneumoniae is 75

a leading cause of pneumonia, meningitis and septicaemia worldwide. Globally, 76

IPD is reported to cause about 11% (0.8 million) of all deaths in children less 77

than 5 years of age annually. The overall burden of IPD is 40-fold increased in 78

HIV-infected compared with HIV-uninfected children. However, mechanisms 79

involved in host response during IPD are not yet fully understood. 80

Page 3 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

4

81

What this study adds 82

We describe differences in transcriptional profiles between HIV-infected and 83

HIV-uninfected children with pneumococcal meningitis (as a homogeneous 84

disease entity of invasive pneumococcal disease) and healthy controls. We 85

demonstrate for the first time, increased expression in cases of genes regulating 86

the innate immune response, leucocyte migration, glucose homeostasis and 87

endothelial cell migration. 88

89

90

Introduction 91

92

Streptococcus pneumoniae infection is a leading cause of pneumonia, 93

meningitis and septicaemia worldwide, and results in approximately 1 million 94

deaths in children under the age of 5 years annually 1. The overall burden of 95

invasive pneumococcal disease (IPD) is 40-fold increased in HIV-infected 96

compared with HIV-uninfected children 2. Pneumococcal meningitis is a life-97

threatening disease with poor prognosis associated with neurologic 98

complications and a high case-fatality ratio in African children, which is further 99

increased by HIV co-infection 3. 100

101

Gene expression profiling during sepsis provides new insights into the host 102

response to invasive bacterial disease. Several sepsis studies have 103

demonstrated up-regulation of pathogen recognition receptors and signal 104

Page 4 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

5

transduction pathways, and a down-regulation of zinc homeostasis 4-6. The 105

intricate host inflammatory response is associated with neuronal and vascular 106

injury, even after cerebrospinal fluid (CSF) sterilization with antibiotics. 107

Adjunctive new therapies for bacterial meningitis have to date not shown any 108

conclusive benefit, prompting the need for an improved understanding of key 109

mechanisms that might reveal potential new therapeutic targets 7. Chronic co-110

infection may impact gene expression, even among asymptomatic patients. HIV 111

is a major risk factor for IPD, characterised by waning immunity and 112

dysregulated physiology among infected individuals, greatly escalating disease 113

susceptibility and mortality outcomes8 9, thereby influencing the host’s gene 114

expression. We examine differences in gene-expression using blood samples 115

from children with pneumococcal meningitis and matched healthy controls, and 116

also compare transcriptional profiles between children with pneumococcal 117

meningitis with and without underlying HIV infection. 118

119

Materials and methods 120

121

Patients and controls 122

Children (n=377) were recruited as part of a larger study investigating host 123

determinants of susceptibility to IPD conducted at Queen Elizabeth Central 124

Hospital, Blantyre, Malawi, between April 2004 and October 2006 10. Ethical 125

approval was granted from the College of Medicine Research Committee 126

(COMREC), Malawi and the Liverpool School of Tropical Medicine Local 127

Page 5 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

6

Research Ethics Committee. Written informed consent was given prior to 128

recruitment. 129

130

We excluded children (n=135) infected with other commonly prevalent microbes 131

(Salmonella typhimurium, Escherichia coli, Haemophilus influenzae b, Neisseria 132

meningitidis, Staphylococcus aureus, Streptococcus pyogenes) identified by 133

positive laboratory culture. Cases for the microarray analysis were HIV 134

treatment-naive children with confirmed pneumococcal meningitis defined as 135

abnormal CSF white blood cell (WBC) count, >10 WBC/µl plus one or more of 136

the following: CSF culture positive for pneumococci, CSF gram stain consistent 137

with pneumococci, CSF positive for pneumococcal polysaccharide antigen or 138

pneumococcal DNA. Cases for the RT-qPCR were children with confirmed IPD, 139

which was defined as: pneumococcal pneumonia (n=40) or pneumococcal 140

meningitis (n=189), confirmed by either culture, antigen test, or PCR. Controls 141

were healthy afebrile, aparasitaemic children from the same villages as the 142

index cases, and were as closely age matched as possible. Microarray analysis 143

was conducted on 15 samples (12 from cases with pneumococcal meningitis 144

and 3 from controls), and RT-PCR was conducted on 242 samples (229 from 145

cases with IPD and 13 from controls), which included all those used for 146

microarray analysis. 147

148

In the microarray discovery cohort, there were 12 children with pneumococcal 149

meningitis (6 male, 6 female, median age 1.1 years) and 3 controls (2 male, 1 150

female, median age 7 years). There were 3 cases in each of the following 151

Page 6 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

7

categories: HIV-infected survivors, HIV-infected non-survivors, HIV-uninfected 152

survivors and HIV-uninfected non-survivors (Supplementary Table 1). The RT-153

PCT validation cohort had a median age of 3.09 years. There were no known 154

co-infections apart from HIV, and we did not test for other viruses in the CSF, 155

like CMV or EBV, which have been reported to be associated with increased 156

mortality in Malawian adults with bacterial meningitis 11 12. 157

158

Samples 159

Whole blood was drawn at admission from consecutive children with IPD. The 160

methodology for downstream transcriptome analysis from small blood samples 161

has been previously described 13. 162

163

RNA extraction, quantification and hybridization 164

Total RNA was extracted from whole blood using an optimised method for the 165

PAXgeneTM blood RNA kit (Qiagen, West Sussex, UK), as previously described 166

13. The total RNA concentration (ng/µl) and ratios (260/280 and 260/230) were 167

measured using a NanoDrop ND-100 UV-vis spectrophotometer (NanoDrop 168

Technologies, Delaware, USA) and the RNA integrity was assessed using the 169

Agilent 2100 BioAnalyser (Agilent Technologies, Edinburgh, UK) before and 170

after concentration. 171

172

RNA (2µg) was reverse transcribed into cDNA using the Superscript double-173

stranded cDNA synthesis kit (Invitrogen, Paisley, UK) according to the 174

manufacturers’ guidelines. The double stranded cDNA was purified using a 175

Page 7 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

8

GeneChipTM sample clean-up module (Invitrogen, Paisley, UK). The purified 176

cDNA was then biotin labelled with the ENZO BioArray high yield RNA transcript 177

labelling kit (Affymetrix, High Wycombe, UK) and cleaned with a cRNA clean-up 178

module (Invitrogen, Paisley, UK). Aliquots of labelled cRNA (20µg) were 179

fragmented at 94°C for 35 minutes and then hybridized to a Human Genome 180

U133A GeneChipTM array for 16 hours rotating at 60rpm at 45°C in a 181

GeneChipTM Hybridization Oven 640 (Affymetrix, High Wycombe, UK). Each 182

chip was washed and stained on a GeneChipTM Fluidics Station 450 (Affymetrix, 183

High Wycombe, UK) and scanned using a GeneChipTM Scanner 3000 184

(Affymetrix, High Wycombe, UK) employing standard recommended protocols 185

(Affymetrix, High Wycombe, UK). 186

187

Microarray data analysis 188

Microarray experiment data were analysed using R and Biocondutor packages 189

14. Briefly, the human genome HGU133A array (Affymetrix, High Wycombe, UK) 190

scans output was pre-processed using the affy package 15. The limma package 191

was used to evaluate differential expressed genes 16 17. Gene annotations were 192

performed using hgu133a.db, KEGG.db database packages 18-21. Genes were 193

considered differentially expressed if they had a Benjamini and Hochberg (BH) 194

adjusted p-value < 1.5e-3 and > ±2-fold change in gene expression. Bonferroni 195

p-value adjustments were also performed for comparison. Canonical pathways 196

and functional networks that involve the differently expressed genes play were 197

determined using the Ingenuity Pathway Analysis (IPA) catalogue of well-198

characterized metabolic and cell-signaling cascades. Expression data can be 199

accessed using accession number GSE47172 at the NCBI GEO database. 200

Page 8 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

9

201

Reverse transcription for quantitative PCR 202

RNA samples were DNAse (Ambion, Warrington, UK) treated to remove any 203

contaminating genomic DNA. RNA (1µg) was reverse transcribed with 204

SuperScript IITM RNase H- Reverse Transcriptase and oligo (dT)12-18 (0.5µg/µL) 205

following the manufacturer’s guidelines. The cDNA was stored at -40°C until 206

required. 207

208

Real-Time quantitative PCR measurement of target genes 209

The Human Universal Probe Library (UPL, Roche, Switzerland) employing 210

proprietary locked nucleic acid (LNA) analogues was used to design qPCR 211

assays to measure expression levels in genes of interest. Using the Roche 212

Online Assay Design Centre, specific primers and an associated probe were 213

selected for the reference and target transcripts. Gene expression was 214

determined using real time qPCR on a Roche LightCycler 480 (Supplementary 215

Table 2). 216

217

The following 34 genes were identified from literature and prioritised for real-218

time qPCR differential expression analysis between cases and controls: ACSL1, 219

ANXA3, ATP, BAG1, BPGM, C3AR1, CA4, CD177, CD55, CD59, CEACAM, 220

CFLAR, FOLR3, GNAI3, GNLY, GYG1, IL1R2, IL1RN, ITGAM, KLRF1, LCK, 221

LCN2, LTF, MAPK14, MMP9, NUMB, OLAH, PSEN1, RETN, S100A12, 222

SAMSN, SERPINA1, SUB1, and VNN1. We used a previously described real-223

time qPCR data normalization method 22. 224

Page 9 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

10

225

Results 226

227

Transcriptional profiles among the cases and controls 228

We examined whether global transcriptional profiles of peripheral blood from 229

children with pneumococcal meningitis (n=12) were distinct from those of 230

healthy controls (n=3) randomly selected from a larger dataset by microarray 231

expression profile analysis 10. In general, there was a marked distinction in 232

differential expression between cases and controls (Supplementary Figure 1). 233

We identified 10 significantly differentially expressed genes (Benjamini & 234

Hochberg (BH) adjusted p-value < 1.5e-3 and > ±2-fold change) (Figure 1). We 235

observed significant up-regulation of: S100 calcium binding protein A12 236

(S100A12); vanin-1 (VNN1); arginase, liver (ARG1); matrix metallopeptidase 9 237

(gelatinase B, 92kDa gelatinase, 92kDa type IV collagenase) (MMP9); annexin 238

A3 (ANXA3); interleukin 1 receptor, type II (IL1R2); CD177 molecule (CD177); 239

S100 calcium binding protein A9 (S100A9), cytoskeleton-associated protein 4 240

(CKAP4); and glycogenin 1 (GYG1) supporting previous findings that have 241

highlighted the important roles of some of these genes during host pathogen 242

response. These genes have important interconnected functions for cellular 243

interactions and response to infection (Figure 3A and Supplementary Table 3), 244

and play crucial roles during host immune response; cell regulatory processes 245

such as apoptosis and differentiation; and metabolic processes such as amino 246

acid and glucose metabolism. 247

248

249

Page 10 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

11

Differential expression, underlying HIV infection and disease outcomes 250

We established the effect of HIV on host transcriptional profiles during 251

meningitis by comparing HIV-infected cases (n=6) and HIV-uninfected cases 252

(n=6). We also compared transcriptional profiles among survivors (n=6) and 253

non-survivors (n=6) to establish whether differential expression might account 254

for these two different disease outcomes. We did not find any significantly 255

differentially expressed genes for any of these comparisons (Supplementary 256

Figure 2). 257

258

RT qPCR validation 259

RT qPCR validation was performed for a set of 34 prioritised genes selected 260

from literature for cases (n=229) and controls (n=13). The RT-PCR results are in 261

agreement with our microarray analysis findings. All the genes are significantly 262

differentially expressed between the two groups with the exception of guanine 263

nucleotide-binding protein subunit alpha-13 (GNA13), protein numb (NUMB), 264

and presenilin-1 (PSEN1) (Figure 2). There was no significantly increased gene 265

expression in HIV-infected compared to HIV-uninfected cases (Supplementary 266

Figure 3). Interestingly, there was significant up-regulation of folate receptor 3 267

(FOLR3), NUMB and S100A12 in survivors compared non-survivors 268

(Supplementary Figure 4). There was wide variability in relative gene expression 269

in cases, but not controls. 270

271

Pathway analysis 272

Networks were reconstructed using IPA software for the genes differentially 273

expressed between cases and controls in the microarray experiment, and the 274

Page 11 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

12

combined microarray and RT qPCR experiments (Figure 3A and 3B). For the 275

genes defined by the microarray experiments the networks were predominantly 276

related to granulocyte function, but also including antimicrobial and endothelial 277

activation responses (Figure 3A). The merged set demonstrated wider networks 278

involving immune cell activation as well as leucocyte migration and adhesion 279

(Figure 3B). The canonical pathways mapped by genes defined in the 280

microarray experiment demonstrated greatest changes in the arginine and 281

granulocyte pathways. Canonical pathways in the combined microarray and RT 282

qPCR experiments demonstrated greatest change in leucocytes, and especially 283

neutrophils activation and migration, Notch and glucocorticoid receptor 284

signalling pathways (Supplementary Tables 3 and 4). 285

286

Discussion and conclusion 287

We have shown significant differences in RNA transcriptional profiles in children 288

with pneumococcal meningitis compared to controls. Children with 289

pneumococcal meningitis demonstrated increased expression in genes involved 290

in the inflammatory response, and glucose and L-arginine metabolic pathways. 291

Dysregulation of these pathways can lead to an impaired adaptive host 292

response to pneumococcal infection, thereby contributing to the increased 293

morbidity and mortality. Our findings in the initial cohort of pneumococcal 294

meningitis were validated in the larger cohort of all children with IPD, which 295

includes presentations with pneumonia as well as meningitis. These findings 296

add validity to the initial results from the microarray experiments, and suggest 297

that the host response observed is systemic, and not simply localised to the 298

Page 12 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

13

process of blood-brain barrier disruption per se. We observe wide variability in 299

relative gene expression in cases, which perhaps reflects differences in disease 300

onset and robustness of host response within this group. 301

302

Vanin-1 (VNN1) protein is expressed by human phagocytes, and involved in 303

leukocyte adhesion and migration 23. The VNN-1 knockout mice model has 304

provided clear evidence that VNN-1 modulates redox and immune pathways 24. 305

Exposure of human mononuclear cells to oxidative stress results in up-306

regulation of human VNN1 and down-regulation of PPARγ 25. IL1R2 and IL1RN 307

were up-regulated in cases, which is consistent with our previous report of the 308

IL-1Ra single-nucleotide polymorphism rs4251961 playing a key role in the 309

pathophysiology of IPD and in other human infections10. Recent reports also 310

demonstrate IL1R2 expression up-regulation in sepsis, being more pronounced 311

in Gram negative than Gram positive infections 26. 312

313

During infection, host production of the cytokine nitric oxide (NO) after non-314

opsonic phagocytosis, exerts microbicidal effects 27 28. ARG1 expression is 315

inducible in the lungs of mice in response to pneumococcal infection29. 316

Phagocytosis of pneumococci by macrophages also results in increased 317

production of nitric oxide synthase 2 (NOS2) dependent production of NO and 318

reactive nitrogen species 30. Increased plasma arginase activity depletes L-319

arginine concentrations, the substrate for NO synthesis, leading to vascular 320

dysfunction during severe sepsis and supressed NO-mediated microbicidal 321

effects 31. Increased ARG1 activity may also be a bacterial survival strategy to 322

escape the NO-dependent host antimicrobial immune response 30. 323

Page 13 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

14

324

Neutrophil-specific glycoprotein CD177 is expressed on a subset of human 325

neutrophils, and is involved in neutrophil transendothelial migration. A previous 326

microarray study of purified neutrophils from septic shock patients revealed 327

CD177 mRNA has the highest differential expression between cases and 328

controls 32. Consistent with our data, that study also demonstrated increased 329

expression of ARG1, ANXA3, CKAP4, IL1R2, MMP9, and VNN1. ANXA3 330

promotes endothelial cell junction integrity in animal models, and endothelial cell 331

motility in vitro. ANXA3 is up-regulated following neuronal injury, which may 332

explain the finding in pneumococcal meningitis 33. 333

334

S100A12 plays a prominent role in the regulation of proinflammatory processes 335

and immune response by recruiting leukocytes, promoting cytokine and 336

chemokine production, and regulating leukocyte adhesion and migration 34. The 337

S100A8/A9 heterodimer is expressed by myeloid cells, especially neutrophils, 338

and has a protective effect in the host response to pneumococcal infection by 339

increasing circulating neutrophils through increased granulocyte-colony 340

stimulating factor production 35. It is an antimicrobial peptide, but also plays an 341

important role in leukocyte migration 36. During infection, S100 proteins stimulate 342

the proinflammatory immune response through interaction with the 343

immunoglobulin family transmembrane pattern recognition receptors: Receptor 344

for advanced glycation endproducts (RAGE) and Toll-like receptor 4 (TLR4). 345

This leads to NF-κB-mediated pro-inflammatory response with production of 346

proinflammatory cytokines. This inflammatory response in turn leads to 347

increased expression of S100 proteins, and the start of a positive feedback loop. 348

Page 14 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

15

Although as antimicrobial proteins they protect against infection, they can also 349

have a negative detrimental effect on the host by amplifying the destructive pro-350

inflammatory responses. The increased expression in survivors may be 351

explained by this positive feedback loop 37. 352

353

Glucocorticoids also play a significant role in immune response regulation, by 354

supressing immune and inflammatory response, and modulating cytokines that 355

promote host immune responses 38. We speculate that these responses are 356

important in regulating immune responses to avoid host tissue damage. NUMB 357

negatively regulates Notch, which in turn attenuates proinflammatory cytokines 358

and increases anti-inflammatory cytokines, which may explain the increase in 359

NUMB expression in survivors 39 40. 360

361

Our results are consistent with previous studies on transcription profiling in other 362

bacterial diseases, suggesting that some of the mechanisms are not specific to 363

IPD. Although it could be argued that the lack of age matching in cases versus 364

controls could cause differential expression due to maturation of the immune 365

system in older children, the consistency of our results with other studies makes 366

this unlikely 41 42. A limitation of our study is that the microarray analysis was 367

performed using the Affymetrix Human Genome U133A array, and sample 368

number was constrained at the time by cost. The microarray dataset was not 369

large enough to allow all possible multifactorial models of co-morbidity and 370

disease outcomes to be exhaustively examined, but sought to validate our 371

findings using RT-qPCR in a larger cohort of 242 children. Further evaluation is 372

Page 15 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

16

in a larger cohort using whole genome sequencing would provide more insights 373

on the mechanisms of host response. 374

375

In conclusion, this comparative study of gene expression provides mechanistic 376

insight for IPD in children, and demonstrates significant and widespread immune 377

activation, with oxidative stress, recruitment of neutrophils, leucocyte adhesion 378

and migration, activation of antimicrobial peptides, and preservation of 379

endothelial cell junction integrity. 380

381

(2542 words) 382

383

Competing interests 384

None declared 385

386

Author contributions 387

FM, PJRD, EDC conceived and designed the study; LM, EDC collected the 388

clinical samples; FM performed experiments; BWK, FM, OV, KN, SR, PJRD, 389

EDC performed the analysis; BWK, FM, PJRD, EDC wrote the first draft of the 390

manuscript; BWK, FM, SR, LM, KN, OV, MM, EM, PJRD, EDC contributed to 391

writing the manuscript. 392

393

394

395

396

Page 16 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

17

Acknowledgements 397

398

The IPD Study Group (C Antonio, M Chinamale, L Jere, D Mnapo, V Munthali, 399

F Nyalo, J Simwinga, M Kaole, DL Banda, A Manyika, K Phiri) recruited 400

patients. We thank the children included in this study, and their parents and 401

guardians for giving consent for them to participate in the study. We also extend 402

thanks to the nursing and medical staff, at the Malawi-Liverpool-Wellcome Trust 403

Clinical Research Programme (MLW), Research Ward, for their contribution to 404

this study. EDC was supported by a Wellcome Trust Career Development Grant 405

(068026). BWK was supported by a Wellcome Trust Major Overseas 406

programme award (084679/Z/08/Z). The funding bodies had no role in 407

collection, analysis or interpretation of data or in writing the manuscript. 408

409 410 References 411 412 1. O'Brien KL, Wolfson LJ, Watt JP, et al. Burden of disease caused by 413

Streptococcus pneumoniae in children younger than 5 years: global 414 estimates. Lancet 2009;374(9693):893-902. doi: S0140-6736(09)61204-6 415 [pii] 416

10.1016/S0140-6736(09)61204-6 [published Online First: 2009/09/15] 417 2. Madhi SA, Petersen K, Madhi A, et al. Impact of human immunodeficiency 418

virus type 1 on the disease spectrum of Streptococcus pneumoniae in 419 South African children. Pediatr Infect Dis J 2000;19(12):1141-7. 420

3. Nyasulu P, Cohen C, De Gouveia L, et al. Increased risk of death in human 421 immunodeficiency virus-infected children with pneumococcal meningitis 422 in South Africa, 2003-2005. Pediatr Infect Dis J 2011;30(12):1075-80. doi: 423 10.1097/INF.0b013e31822cca05 [published Online First: 2011/07/30] 424

4. Wong HR, Cvijanovich N, Allen GL, et al. Genomic expression profiling 425 across the pediatric systemic inflammatory response syndrome, sepsis, 426 and septic shock spectrum. Crit Care Med 2009;37(5):1558-66. doi: 427 10.1097/CCM.0b013e31819fcc08 [published Online First: 2009/03/28] 428

5. Wong HR, Freishtat RJ, Monaco M, et al. Leukocyte subset-derived 429 genomewide expression profiles in pediatric septic shock. Pediatric 430 critical care medicine : a journal of the Society of Critical Care Medicine 431 and the World Federation of Pediatric Intensive and Critical Care 432

Page 17 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

18

Societies 2010;11(3):349-55. doi: 10.1097/PCC.0b013e3181c519b4 433 [published Online First: 2009/12/17] 434

6. Khoo SK, Petillo D, Parida M, et al. Host response transcriptional profiling 435 reveals extracellular components and ABC (ATP-binding cassette) 436 transporters gene enrichment in typhoid fever-infected Nigerian children. 437 BMC infectious diseases 2011;11:241. doi: 10.1186/1471-2334-11-241 438

1471-2334-11-241 [pii] [published Online First: 2011/09/15] 439 7. Hoffman O, Weber RJ. Pathophysiology and treatment of bacterial meningitis. 440

Ther Adv Neurol Disord 2009;2(6):1-7. doi: 10.1177/1756285609337975 441 [published Online First: 2010/12/25] 442

8. Glennie SJ, Williams NA, Heyderman RS. Mucosal immunity in resource-443 limited setting: is the battle ground different? Trends in microbiology 444 2010;18(11):487-93. doi: 10.1016/j.tim.2010.08.002 445

S0966-842X(10)00139-3 [pii] [published Online First: 2010/09/14] 446 9. Iwajomo OH, Finn A, Moons P, et al. Deteriorating pneumococcal-specific B-447

cell memory in minimally symptomatic African children with HIV infection. 448 The Journal of infectious diseases 2011;204(4):534-43. doi: jir316 [pii] 449

10.1093/infdis/jir316 [published Online First: 2011/07/28] 450 10. Carrol ED, Payton A, Payne D, et al. The IL1RN promoter rs4251961 451

correlates with IL-1 receptor antagonist concentrations in human infection 452 and is differentially regulated by GATA-1. J Immunol 2011;186(4):2329-453 35. doi: 10.4049/jimmunol.1002402 454

11. Kelly MJ, Benjamin LA, Cartwright K, et al. Epstein-barr virus coinfection in 455 cerebrospinal fluid is associated with increased mortality in Malawian 456 adults with bacterial meningitis. J Infect Dis 2012;205(1):106-10. doi: 457 10.1093/infdis/jir707 458

jir707 [pii] [published Online First: 2011/11/15] 459 12. Benjamin LA, Kelly M, Cohen D, et al. Detection of herpes viruses in the 460

cerebrospinal fluid of adults with suspected viral meningitis in Malawi. 461 Infection 2013;41(1):27-31. doi: 10.1007/s15010-012-0292-z [published 462 Online First: 2012/07/17] 463

13. Carrol ED, Salway F, Pepper SD, et al. Successful downstream application 464 of the Paxgene Blood RNA system from small blood samples in 465 paediatric patients for quantitative PCR analysis. BMC Immunol 466 2007;8:20. doi: 10.1186/1471-2172-8-20 467

14. Gentleman RC, Carey VJ, Bates DM, et al. Bioconductor: open software 468 development for computational biology and bioinformatics. Genome 469 biology 2004;5(10):R80.1-R80.16. 470

15. Irizarry R. From CEL files to annotated lists of interesting genes. In: 471 Gentleman R. CV, Dudoit S., Irizarry R., Huber W., ed. Bioinformatics 472 and Computational Biology Solutions using R and Bioconductor. New 473 York: Springer 2005:431-42. 474

16. Smyth GK. Limma: linear models for microarray data. In: Gentleman R. CV, 475 Dudoit S., Irizarry R., Huber W., ed. Bioinformatics and Computational 476 Biology Solutions using R and Bioconductor. New York: Springer 477 2005:397-420. 478

Page 18 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

19

17. Smyth GK. Linear models and empirical Bayes methods for assessing 479 differential expression in microarray experiments. Statistical Applications 480 in Genetics and Molecular Biology 2004;3(1):3. 481

18. hgu133a.db: Affymetrix Human Genome U133 Set annotation data (chip 482 hgu133a) [program]. R package version 2.7.1 version, 2009. 483

19. annotate: Annotation for microarrays. R package version 1.34.1. [program]. 484 R package version 1.34.1. version, 2008. 485

20. GO.db: A set of annotation maps describing the entire Gene Ontology 486 [program]. R package version 2.12 version, 2013. 487

21. KEGG.db: A set of annotation maps for KEGG [program]. R package version 488 2.7.1 version, 2009. 489

22. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of 490 real-time quantitative RT-PCR data by geometric averaging of multiple 491 internal control genes. Genome Biol 2002;3(7):RESEARCH0034. 492

23. Meghari S, Berruyer C, Lepidi H, et al. Vanin-1 controls granuloma formation 493 and macrophage polarization in Coxiella burnetii infection. Eur J Immunol 494 2007;37(1):24-32. doi: 10.1002/eji.200636054 495

24. Ferreira DW, Naquet P, Manautou JE. Influence of Vanin-1 and Catalytic 496 Products in Liver During Normal and Oxidative Stress Conditions. Curr 497 Med Chem 2015;22(20):2407-16. 498

25. Zhang B, Lo C, Shen L, et al. The role of vanin-1 and oxidative stress-499 related pathways in distinguishing acute and chronic pediatric ITP. Blood 500 2011;117(17):4569-79. doi: 10.1182/blood-2010-09-304931 501

26. Lang Y, Jiang Y, Gao M, et al. Interleukin-1 Receptor 2: A New Biomarker 502 for Sepsis Diagnosis and Gram-Negative/Gram-Positive Bacterial 503 Differentiation. Shock 2017;47(1):119-24. doi: 504 10.1097/SHK.0000000000000714 505

27. Ulett GC, Adderson EE. Nitric oxide is a key determinant of group B 506 streptococcus-induced murine macrophage apoptosis. The Journal of 507 infectious diseases 2005;191(10):1761-70. doi: 10.1086/429693 508 [published Online First: 2005/04/20] 509

28. Gross A, Spiesser S, Terraza A, et al. Expression and bactericidal activity of 510 nitric oxide synthase in Brucella suis-infected murine macrophages. 511 Infection and immunity 1998;66(4):1309-16. [published Online First: 512 1998/04/07] 513

29. Knippenberg S, Brumshagen C, Aschenbrenner F, et al. Arginase 1 activity 514 worsens lung-protective immunity against Streptococcus pneumoniae 515 infection. Eur J Immunol 2015;45(6):1716-26. doi: 10.1002/eji.201445419 516

30. Marriott HM, Ali F, Read RC, et al. Nitric oxide levels regulate macrophage 517 commitment to apoptosis or necrosis during pneumococcal infection. 518 FASEB J 2004;18(10):1126-8. doi: 10.1096/fj.03-1450fje 519

31. Darcy CJ, Woodberry T, Davis JS, et al. Increased plasma arginase activity 520 in human sepsis: association with increased circulating neutrophils. 521 Clinical chemistry and laboratory medicine : CCLM / FESCC 522 2014;52(4):573-81. doi: 10.1515/cclm-2013-0698 [published Online First: 523 2013/10/30] 524

32. Demaret J, Venet F, Plassais J, et al. Identification of CD177 as the most 525 dysregulated parameter in a microarray study of purified neutrophils from 526

Page 19 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

20

septic shock patients. Immunol Lett 2016;178:122-30. doi: 527 10.1016/j.imlet.2016.08.011 528

33. Chong KW, Chen MJ, Koay ES, et al. Annexin A3 is associated with cell 529 death in lactacystin-mediated neuronal injury. Neurosci Lett 530 2010;485(2):129-33. doi: 10.1016/j.neulet.2010.08.089 531

34. Goyette J, Geczy CL. Inflammation-associated S100 proteins: new 532 mechanisms that regulate function. Amino Acids 2011;41(4):821-42. doi: 533 10.1007/s00726-010-0528-0 534

35. De Filippo K, Neill DR, Mathies M, et al. A new protective role for S100A9 in 535 regulation of neutrophil recruitment during invasive pneumococcal 536 pneumonia. FASEB J 2014;28(8):3600-8. doi: 10.1096/fj.13-247460 537

36. Raquil MA, Anceriz N, Rouleau P, et al. Blockade of antimicrobial proteins 538 S100A8 and S100A9 inhibits phagocyte migration to the alveoli in 539 streptococcal pneumonia. J Immunol 2008;180(5):3366-74. 540

37. Zackular JP, Chazin WJ, Skaar EP. Nutritional Immunity: S100 Proteins at 541 the Host-Pathogen Interface. The Journal of biological chemistry 542 2015;290(31):18991-8. doi: 10.1074/jbc.R115.645085 543

38. Sakai A, Han J, Cato AC, et al. Glucocorticoids synergize with IL-1beta to 544 induce TLR2 expression via MAP Kinase Phosphatase-1-dependent dual 545 Inhibition of MAPK JNK and p38 in epithelial cells. BMC molecular 546 biology 2004;5:2. doi: 10.1186/1471-2199-5-2 [published Online First: 547 2004/05/06] 548

39. Kueanjinda P, Roytrakul S, Palaga T. A Novel Role of Numb as A Regulator 549 of Pro-inflammatory Cytokine Production in Macrophages in Response to 550 Toll-like Receptor 4. Scientific reports 2015;5:12784. doi: 551 10.1038/srep12784 552

40. Singla DK, Wang J, Singla R. Primary human monocytes differentiate into 553 M2 macrophages and involve Notch-1 pathway. Can J Physiol Pharmacol 554 2017;95(3):288-94. doi: 10.1139/cjpp-2016-0319 555

41. Langley RJ, Tsalik EL, Velkinburgh JC, et al. An integrated clinico-556 metabolomic model improves prediction of death in sepsis. Sci Transl 557 Med 2013;5(195):195ra95. doi: 10.1126/scitranslmed.3005893 558

5/195/195ra95 [pii] [published Online First: 2013/07/26] 559 42. Liu K, Chen L, Kaur R, et al. Transcriptome signature in young children with 560

acute otitis media due to Streptococcus pneumoniae. Microbes Infect 561 2012;14(7-8):600-9. doi: 10.1016/j.micinf.2012.01.006 562

S1286-4579(12)00035-4 [pii] [published Online First: 2012/02/07] 563

564

Page 20 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

21

Figure and Table legends 565

566

Figure 1: Distribution plot of the differentially expressed genes. The 567

significantly differentially expressed genes are shown in red colour. The 568

significance threshold (P<1.5e-3) is indicated by a dashed red line, and a fold 569

change threshold of more than 2 is shown by the dashed vertical lines. The 570

green line shows P<0.05. (A) Shows results for un-adjusted p-values, (B) 571

results for Benjamini and Hochberg (BH) adjusted p-values, and (C) stringent 572

Bonferroni adjusted p-values, which represents an over-correction. 573

574

Figure 2: Validation of RNA transcription profile differential expression 575

using Real Time quantitative PCR. Relative gene expression in cases 576

compared to controls. The black line shows the boxplot median. The red dot 577

shows the mean, and the Welch two-sample t-test p-value is shown on the top 578

right corner. 579

Figure 3: The gene network for significantly differentially expressed 580

genes in the cases. Networks reconstructed to support the most direct 581

connectivity between genes differentially expressed between cases and 582

controls. IPA (Qiagen) software's Core analysis application has been used to 583

perform an automatic graphical reconstruction of the network via utilisation of 584

IPA's Knowledge base database of protein interactions. The meaning of links 585

and shapes is explained in the inserted legend. Functional connections are 586

presented in blue, information connections to the associated categories in 587

pink and grey. (A) Network reconstructed only for the DE genes identified by 588

Page 21 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

22

microarray analysis. (B) Network reconstructed for the merged datasets of DE 589

genes identified via microarray (red) and PCR analysis (green). White blocks 590

correspond to nodes added by IPA's network editor automatically to ensure 591

network connectivity. 592

593

Supplementary Figure 1: Children with pneumococcal meningitis and 594

healthy controls have distinct transcriptional profiles. A heatmap of the 595

top 100 (P<0.05) differentially expressed genes compared in cases and 596

controls. The cases and controls were distinguished into two distinct clusters 597

highlighting their distinct transcriptional profiles. 598

599

600

Supplementary Figure 2: Comparison of microarray transcriptional 601

profiles of HIV infected and HIV-uninfected and survivors and non-602

survivors. There were no significantly differentially expressed genes for any 603

of these comparisons 604

605

Supplementary Figure 3: Comparison of RT-qPCT gene expression for 606

HIV infected and HIV-uninfected. There were no significantly differentially 607

expressed genes for all of these comparisons. 608

609

Supplementary Figure 4: Comparison of RT-qPCT gene expression for 610

survivors and non-survivors. There were three significantly differentially 611

expressed genes: FOLR3, NUMB, and S100A12. 612

Page 22 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

23

613

Supplementary Table 1: Clinical data on children with pneumococcal 614

meningitis and healthy controls. 615

616

Supplementary Table 2: Primer sequences and the UPL probe number 617

for each RT qPCR assay. 618

619

Supplementary Table 3: Canonical pathways of differentially expressed 620

genes between cases and controls in microarray experiment 621

622

Supplementary Table 4: Canonical pathways of differentially expressed 623

genes between cases and controls in combined microarray and RT- 624

qPCR experiments 625

626

627

628

629

Page 23 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

94x24mm (300 x 300 DPI)

Page 24 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

177x177mm (300 x 300 DPI)

Page 25 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

361x481mm (300 x 300 DPI)

Page 26 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

270x203mm (300 x 300 DPI)

Page 27 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

117x42mm (300 x 300 DPI)

Page 28 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

177x177mm (300 x 300 DPI)

Page 29 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

177x177mm (300 x 300 DPI)

Page 30 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyTable S1: Clinical data on children with pneumococcal meningitis and

healthy controls

Sample ID

Age

(years)

Sex Case/control HIV status

Survival status

Previous antibiotics use

Duration of symptoms (days)

94 0.4 F Case Positive Died No 2

95 1.0 F Case Positive

Died Yes 6


Died No 1

96 1.2 M Case Positive

Alive No 3


Alive No 2


Alive Yes 3

82 13.0 M Case Negative Died No 7

115 0.3 M Case Negative

Died Yes 3

121 0.5 F Case Negative

Died No 5


Alive No 3


Alive No 3


Alive Yes 3

C08 7.0 M Control Negative

Alive N/A N/A


Alive N/A N/A

C13 7.0 F Control Negative

Alive N/A N/A

Page 31 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from


Confidential: For Review Only1

Table S2: Primer sequences and the UPL probe number for each RT qPCR assay

Gene Symbol Gene Forward Primer (5'-3') Reverse Primer (5'-3') UPL Probe Accession Number

ACSL1 Acyl-CoA synthetase long-chain family member 1 ccactgtgctgacgttcct gcactctgtctgtccgtatcc 26 NM_001995

ANXA3 Annexin A3 ttgttaaggaatatcaagcagcat gccagagagatcacccttca 17 NM_005139

C4A Complement component 4A (Rodgers blood group) ggaccctgaagacgaaattg gaagccctcgttcacctg 75 NM_007293, NM_001252204

C3AR1 complement component 3a receptor 1

aagaaagcaaggcagtccatt

cgtgtgagctcctcactgaa

16

NM_004054

ILR2A ILR2A interleukin 1 receptor, type II tacgcaccacagtcaaggaa aagaaggccagtgaaagtgg 2 NM_004633, NM_173343

MAPK14

MAPK 14 mitogen-activated protein kinase 14


ccatgagatgggtcaccag

65

NM_139012, NM_139014, NM_001315, NM_139013

MMP9

matrix metallopeptidase 9 (gelatinase B, 92kDa gelatinase, 92kDa type IV collagenase)

gaaccaatctcaccgacagg


6

NM_004994

CD177 CD177 molecule gcctctccctgatctcctaca caaggatcctgggtctgc 79 NM_020406

CD55 CD55 molecule, decay accelerating factor for complement (Cromer blood group) aataatgatgaaggagagtggagtg tggtgggaccttggaagtta 77 NM_000574

CEACAM1 Carcinoembryonic antigen-related cell adhesion molecule 1 (biliary glycoprotein) taggcctaagaggctttctcc tgagggaaagtatgcacagtccgtgtc 80

NM_001712, NM_001184816, NM_001184813, NM_001184815, NM_001205344, NM_001024912

GYG1 Glycogenin 1 ctgaaacagcacaggaccac atcgccactgtccaagacat 75 NM_004130

FOLR3 Folate receptor 3 (gamma) cacaaaggctggaattggac cttgaaggagtggctccaga 19 NM_000804

ITGAM Integrin, alpha M (complement component 3 receptor 3 subunit) tctgaagaccattcagaaccag ggagctgctacttcctgtctg 7 NM_000632

LTF Lactotransferrin ctaatctgaaaaagtgctcaacctc gccatcttcttcggttttacttc 79 NM_002343, NM_001199149

OLAH Oleoyl-ACP hydrolase accgagatcgctccactg ggagaatcaagtgagattttaatagga 19 NM_018324, NM_001039702

RETN Resistin tgcaggatgaaagctctctg catggagcacagggtcttg 45 NM_020415

Page 32 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from




S100A12 S100 calcium binding protein A12 tcatatccctggtagccattg acctactctttgtgggtgtggt 44 NM_005621

SERPINA1 Serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1 aatggggctgacctctcc gtcagcacagccttatgcac 82 NM_001002235, NM_001002236,

NM_000295

SAMSN1 SAM domain, SH3 domain and nuclear localization signals 1 tgaaaatctgtctgacatggtaca tagttgggaatgcgtgttca 21 NM_022136, NM_001256370

VNN1 Vanin 1 tcctgaggtgttgctgagtg agcgtccgtcagttgacac 80 NM_004666

CD59 CD59 molecule, complement regulatory protein caaggagggtctgtcctgtt

agcactgcaggct[a or g]tgacc 66

NM_000611, NM_203331, NM_203330, NM_203329

CFLAR CASP8 and FADD-like apoptosis regulator gcaatccaaaagagtctcaagg

tgagcgccaagctgttcc 7 NM_003879

KLRF1 Killer cell lectin-like receptor subfamily F, member 1

tgatctccttgatcctgttgg

tcttcttgtgccattattcacttt 47 NM_016523

LCN2 Homo sapiens lipocalin 2 tcacctccgtcctgtttagg

aggtaactcgttaatccagggtaa 61 NM_005564

NUMB Numb homolog (Drosophila) gaggaagactgatttccccatt

gcacagggagctgatgctc 16

NM_001005745, NM_003744, NM_001005744, NM_001005743

Page 33 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from


Confidential: For Review Only

Supplementary Table 3: Canonical pathways of differentially expressedgenes between cases and controls in microarray experiment Ingenuity Canonical Pathways -log(p-value) Ratio MoleculesLXR/RXR Activation 2.77 0.0165 IL1R2,MMP9Arginine Degradation I (Arginase Pathway) 2.69 0.25 ARG1Urea Cycle 2.51 0.167 ARG1Arginine Degradation VI (Arginase 2 Pathway) 2.51 0.167 ARG1Glycogen Biosynthesis II (from UDP-D-Glucose) 2.51 0.167 GYG1Granulocyte Adhesion and Diapedesis 2.51 0.0121 IL1R2,MMP9Hepatic Fibrosis / Hepatic Stellate Cell Activation 2.43 0.011 IL1R2,MMP9Airway Pathology in Chronic Obstructive Pulmonary Disease 2.38 0.125 MMP9Citrulline Biosynthesis 2.38 0.125 ARG1Role of IL-17A in Psoriasis 2.17 0.0769 S100A9Superpathway of Citrulline Metabolism 2.14 0.0714 ARG1Inhibition of Angiogenesis by TSP1 1.79 0.0312 MMP9Inhibition of Matrix Metalloproteases 1.71 0.0263 MMP9Neuroprotective Role of THOP1 in Alzheimer's Disease 1.69 0.025 MMP9IL-10 Signaling 1.46 0.0147 IL1R2Glioma Invasiveness Signaling 1.45 0.0143 MMP9Bladder Cancer Signaling 1.36 0.0116 MMP9PPAR Signaling 1.34 0.0111 IL1R2HIF1α Signaling 1.25 0.00885 MMP9p38 MAPK Signaling 1.23 0.00855 IL1R2Pancreatic Adenocarcinoma Signaling 1.23 0.00847 MMP9Atherosclerosis Signaling 1.21 0.00806 MMP9IL-6 Signaling 1.2 0.00787 IL1R2Ovarian Cancer Signaling 1.15 0.00699 MMP9Relaxin Signaling 1.13 0.00671 MMP9Hepatic Cholestasis 1.11 0.00637 IL1R2PPARα/RXRα Activation 1.08 0.00602 IL1R2Agranulocyte Adhesion and Diapedesis 1.06 0.00571 MMP9NF-κB Signaling 1.06 0.00565 IL1R2Regulation of the Epithelial-Mesenchymal Transition Pathway 1.03 0.00535 MMP9ILK Signaling 1.02 0.00521 MMP9IL-8 Signaling 1.01 0.0051 MMP9Leukocyte Extravasation Signaling 0.996 0.00488 MMP9LPS/IL-1 Mediated Inhibition of RXR Function 0.99 0.00481 IL1R2Role of Osteoblasts, Osteoclasts and Chondrocytes in Rheumatoid Arthritis 0.954 0.00441 IL1R2Colorectal Cancer Metastasis Signaling 0.928 0.00413 MMP9Glucocorticoid Receptor Signaling 0.862 0.00352 IL1R2Role of Macrophages, Fibroblasts and Endothelial Cells in Rheumatoid Arthritis 0.84 0.00333 IL1R2Axonal Guidance Signaling 0.686 0.00226 MMP9

Page 34 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Ingenuity Canonical Pathways -log(p-value) Ratio MoleculesComplement System 6.35 0.111 CD55,CD59,ITGAM,C3AR1Granulocyte Adhesion and Diapedesis 5.02 0.0303 IL1R2,GNAI3,ITGAM,IL1RN,MMP9LXR/RXR Activation 4.24 0.0331 IL1R2,IL1RN,SERPINA1,MMP9IL-10 Signaling 3.64 0.0441 IL1R2,MAPK14,IL1RNLeukocyte Extravasation Signaling 3.36 0.0195 GNAI3,MAPK14,ITGAM,MMP9p38 MAPK Signaling 2.95 0.0256 IL1R2,MAPK14,IL1RNAtherosclerosis Signaling 2.88 0.0242 IL1RN,SERPINA1,MMP9IL-6 Signaling 2.85 0.0236 IL1R2,MAPK14,IL1RNInhibition of Angiogenesis by TSP1 2.84 0.0625 MAPK14,MMP9Glucocorticoid Receptor Signaling 2.83 0.0141 IL1R2,MAPK14,BAG1,IL1RNIL-17A Signaling in Fibroblasts 2.77 0.0571 MAPK14,LCN2Notch Signaling 2.72 0.0541 NUMB,PSEN1Acute Phase Response Signaling 2.5 0.0179 MAPK14,IL1RN,SERPINA1Amyloid Processing 2.46 0.04 MAPK14,PSEN1Agranulocyte Adhesion and Diapedesis 2.45 0.0171 GNAI3,IL1RN,MMP9NF-κB Signaling 2.44 0.0169 IL1R2,LCK,IL1RNMolecular Mechanisms of Cancer 2.42 0.0109 GNAI3,MAPK14,CFLAR,PSEN1IL-8 Signaling 2.31 0.0153 GNAI3,ITGAM,MMP9LPS/IL-1 Mediated Inhibition of RXR Function 2.24 0.0144 IL1R2,IL1RN,ACSL1CCR5 Signaling in Macrophages 2.21 0.0299 GNAI3,MAPK14Chemokine Signaling 2.2 0.0294 GNAI3,MAPK14Caveolar-mediated Endocytosis Signaling 2.16 0.0282 CD55,ITGAMArginine Degradation I (Arginase Pathway) 2.15 0.25 ARG1Acetate Conversion to Acetyl-CoA 2.15 0.25 ACSL1Toll-like Receptor Signaling 2.14 0.0274 MAPK14,IL1RNRole of Osteoblasts, Osteoclasts and Chondrocytes in Rheumatoid Arthritis 2.14 0.0132 IL1R2,MAPK14,IL1RNIL-15 Signaling 2.1 0.0263 LCK,MAPK14Rapoport-Luebering Glycolytic Shunt 2.06 0.2 BPGMPEDF Signaling 2.02 0.0238 MAPK14,CFLARUrea Cycle 1.98 0.167 ARG1Arginine Degradation VI (Arginase 2 Pathway) 1.98 0.167 ARG1Glycogen Biosynthesis II (from UDP-D-Glucose) 1.98 0.167 GYG1PPAR Signaling 1.96 0.0222 IL1R2,IL1RNIL-1 Signaling 1.95 0.022 GNAI3,MAPK14Cholecystokinin/Gastrin-mediated Signaling 1.87 0.0198 MAPK14,IL1RNAirway Pathology in Chronic Obstructive Pulmonary Disease 1.86 0.125 MMP9Citrulline Biosynthesis 1.86 0.125 ARG1Corticotropin Releasing Hormone Signaling 1.82 0.0187 GNAI3,MAPK14Role of Macrophages, Fibroblasts and Endothelial Cells in Rheumatoid Arthritis 1.81 0.01 IL1R2,MAPK14,IL1RNPaxillin Signaling 1.8 0.0182 MAPK14,ITGAMHIF1α Signaling 1.78 0.0177 MAPK14,MMP9Role of Tissue Factor in Cancer 1.73 0.0168 LCK,MAPK14FXR/RXR Activation 1.69 0.016 IL1RN,SERPINA1CCR3 Signaling in Eosinophils 1.69 0.0159 GNAI3,MAPK14

Supplementary Table 4: Canonical pathways of differentially expressed genes between cases and controls in combined microarrayand RT- qPCR experiments

Page 35 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



GNRH Signaling 1.68 0.0157 GNAI3,MAPK14Role of IL-17A in Psoriasis 1.65 0.0769 S100A9Fatty Acid Activation 1.65 0.0769 ACSL1Superpathway of Citrulline Metabolism 1.61 0.0714 ARG1IL-12 Signaling and Production in Macrophages 1.58 0.0139 MAPK14,SERPINA1Parkinson's Signaling 1.56 0.0625 MAPK14Relaxin Signaling 1.55 0.0134 GNAI3,MMP9γ-linolenate Biosynthesis II (Animals) 1.53 0.0588 ACSL1Mitochondrial L-carnitine Shuttle Pathway 1.53 0.0588 ACSL1Oxidative Ethanol Degradation III 1.53 0.0588 ACSL1Hepatic Cholestasis 1.51 0.0127 IL1R2,IL1RNPPARα/RXRα Activation 1.47 0.012 IL1R2,MAPK14Tec Kinase Signaling 1.46 0.0119 GNAI3,LCKEthanol Degradation IV 1.44 0.0476 ACSL1Differential Regulation of Cytokine Production in Intestinal Epithelial Cells by IL-17A and IL-17F1.4 0.0435 LCN2Role of NFAT in Regulation of the Immune Response 1.4 0.0111 GNAI3,LCKHepatic Fibrosis / Hepatic Stellate Cell Activation 1.4 0.011 IL1R2,MMP9Dendritic Cell Maturation 1.39 0.011 MAPK14,IL1RNEndothelin-1 Signaling 1.39 0.0109 GNAI3,MAPK14IL-22 Signaling 1.38 0.0417 MAPK14Glycolysis I 1.38 0.0417 BPGMRegulation of the Epithelial-Mesenchymal Transition Pathway 1.37 0.0107 MMP9,PSEN1IL-17A Signaling in Gastric Cells 1.37 0.04 MAPK14Role of JAK family kinases in IL-6-type Cytokine Signaling 1.37 0.04 MAPK14Gluconeogenesis I 1.37 0.04 BPGMRole of NFAT in Cardiac Hypertrophy 1.36 0.0106 GNAI3,MAPK14Production of Nitric Oxide and Reactive Oxygen Species in Macrophages 1.35 0.0104 MAPK14,SERPINA1Clathrin-mediated Endocytosis Signaling 1.33 0.0102 NUMB,SERPINA1Thrombin Signaling 1.32 0.01 GNAI3,MAPK14Fatty Acid β-oxidation I 1.29 0.0333 ACSL14-1BB Signaling in T Lymphocytes 1.28 0.0323 MAPK14G Protein Signaling Mediated by Tubby 1.28 0.0323 LCKEthanol Degradation II 1.26 0.0312 ACSL1Systemic Lupus Erythematosus Signaling 1.26 0.00922 LCK,IL1RNCoagulation System 1.22 0.0286 SERPINA1Cardiac Hypertrophy Signaling 1.2 0.00862 GNAI3,MAPK14April Mediated Signaling 1.19 0.0263 MAPK14Inhibition of Matrix Metalloproteases 1.19 0.0263 MMP9Role of PKR in Interferon Induction and Antiviral Response 1.17 0.025 MAPK14B Cell Activating Factor Signaling 1.17 0.025 MAPK14Neuroprotective Role of THOP1 in Alzheimer's Disease 1.17 0.025 MMP9Stearate Biosynthesis I (Animals) 1.17 0.025 ACSL1Role of Hypercytokinemia/hyperchemokinemia in the Pathogenesis of Influenza 1.16 0.0244 IL1RNUVC-Induced MAPK Signaling 1.15 0.0238 MAPK14iNOS Signaling 1.14 0.0233 MAPK14Primary Immunodeficiency Signaling 1.13 0.0227 LCKGraft-versus-Host Disease Signaling 1.13 0.0227 IL1RN

Page 36 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Role of Cytokines in Mediating Communication between Immune Cells 1.06 0.0192 IL1RNRetinoic acid Mediated Apoptosis Signaling 1.01 0.0172 CFLARMSP-RON Signaling Pathway 1.01 0.0169 ITGAMCalcium-induced T Lymphocyte Apoptosis 0.999 0.0167 LCKIL-2 Signaling 0.973 0.0156 LCKUVB-Induced MAPK Signaling 0.96 0.0152 MAPK14Role of IL-17A in Arthritis 0.954 0.0149 MAPK14EGF Signaling 0.948 0.0147 MAPK14Melatonin Signaling 0.942 0.0145 GNAI3Glioma Invasiveness Signaling 0.936 0.0143 MMP9Role of MAPK Signaling in the Pathogenesis of Influenza 0.936 0.0143 MAPK14ErbB4 Signaling 0.936 0.0143 PSEN1GPCR-Mediated Integration of Enteroendocrine Signaling Exemplified by an L Cell 0.93 0.0141 GNAI3Role of PI3K/AKT Signaling in the Pathogenesis of Influenza 0.919 0.0137 GNAI3Ephrin B Signaling 0.919 0.0137 GNAI3STAT3 Pathway 0.919 0.0137 MAPK14BMP signaling pathway 0.913 0.0135 MAPK14CD40 Signaling 0.897 0.013 MAPK14IL-17A Signaling in Airway Cells 0.897 0.013 MAPK14ATM Signaling 0.887 0.0127 MAPK14FLT3 Signaling in Hematopoietic Progenitor Cells 0.872 0.0122 MAPK14Communication between Innate and Adaptive Immune Cells 0.872 0.0122 IL1RNAltered T Cell and B Cell Signaling in Rheumatoid Arthritis 0.867 0.012 IL1RNGPCR-Mediated Nutrient Sensing in Enteroendocrine Cells 0.862 0.0119 GNAI3α-Adrenergic Signaling 0.857 0.0118 GNAI3IL-17 Signaling 0.857 0.0118 MAPK14Neuregulin Signaling 0.852 0.0116 PSEN1LPS-stimulated MAPK Signaling 0.852 0.0116 MAPK14NF-κB Activation by Viruses 0.852 0.0116 LCKBladder Cancer Signaling 0.852 0.0116 MMP9TGF-β Signaling 0.848 0.0115 MAPK14G Beta Gamma Signaling 0.843 0.0114 GNAI3Factors Promoting Cardiogenesis in Vertebrates 0.838 0.0112 MAPK14FGF Signaling 0.834 0.0111 MAPK14Death Receptor Signaling 0.83 0.011 CFLARReelin Signaling in Neurons 0.825 0.0109 LCKErbB Signaling 0.804 0.0103 MAPK14CTLA4 Signaling in Cytotoxic T Lymphocytes 0.8 0.0102 LCKCDK5 Signaling 0.8 0.0102 MAPK14RANK Signaling in Osteoclasts 0.796 0.0101 MAPK14Antioxidant Action of Vitamin C 0.796 0.0101 MAPK14UVA-Induced MAPK Signaling 0.788 0.0099 MAPK14Virus Entry via Endocytic Pathways 0.784 0.0098 CD55SAPK/JNK Signaling 0.78 0.00971 LCKMouse Embryonic Stem Cell Pluripotency 0.772 0.00952 MAPK14Type I Diabetes Mellitus Signaling 0.769 0.00943 MAPK14T Cell Receptor Signaling 0.761 0.00926 LCK

Page 37 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Androgen Signaling 0.754 0.00909 GNAI3p53 Signaling 0.751 0.00901 MAPK14Axonal Guidance Signaling 0.741 0.00452 GNAI3,MMP9iCOS-iCOSL Signaling in T Helper Cells 0.73 0.00855 LCKPancreatic Adenocarcinoma Signaling 0.726 0.00847 MMP9Fc Epsilon RI Signaling 0.723 0.0084 MAPK14Renin-Angiotensin Signaling 0.723 0.0084 MAPK14Natural Killer Cell Signaling 0.72 0.00833 LCKfMLP Signaling in Neutrophils 0.72 0.00833 GNAI3Gαi Signaling 0.72 0.00833 GNAI3Sphingosine-1-phosphate Signaling 0.713 0.0082 GNAI3Type II Diabetes Mellitus Signaling 0.704 0.008 ACSL1CD28 Signaling in T Helper Cells 0.701 0.00794 LCKPKCθ Signaling in T Lymphocytes 0.698 0.00787 LCKCdc42 Signaling 0.692 0.00775 MAPK14p70S6K Signaling 0.689 0.00769 GNAI3Th1 Pathway 0.689 0.00769 PSEN1Role of Pattern Recognition Receptors in Recognition of Bacteria and Viruses 0.686 0.00763 C3AR1HMGB1 Signaling 0.686 0.00763 MAPK14P2Y Purigenic Receptor Signaling Pathway 0.686 0.00763 GNAI3Synaptic Long Term Depression 0.66 0.00714 GNAI3Ovarian Cancer Signaling 0.652 0.00699 MMP9Th2 Pathway 0.644 0.00685 PSEN1Regulation of eIF4 and p70S6K Signaling 0.624 0.00649 MAPK14Dopamine-DARPP32 Feedback in cAMP Signaling 0.614 0.00633 GNAI3CXCR4 Signaling 0.6 0.0061 GNAI3Mitochondrial Dysfunction 0.598 0.00606 PSEN1Gap Junction Signaling 0.595 0.00602 GNAI3Germ Cell-Sertoli Cell Junction Signaling 0.589 0.00592 MAPK14RhoGDI Signaling 0.582 0.00581 GNAI3Ephrin Receptor Signaling 0.582 0.00581 GNAI3Sertoli Cell-Sertoli Cell Junction Signaling 0.58 0.00578 MAPK14Th1 and Th2 Activation Pathway 0.565 0.00556 PSEN1CREB Signaling in Neurons 0.561 0.00549 GNAI3B Cell Receptor Signaling 0.561 0.00549 MAPK14RAR Activation 0.551 0.00535 MAPK14AMPK Signaling 0.549 0.00532 MAPK14NRF2-mediated Oxidative Stress Response 0.545 0.00526 MAPK14ILK Signaling 0.541 0.00521 MMP93-phosphoinositide Biosynthesis 0.539 0.00518 LCKBreast Cancer Regulation by Stathmin1 0.523 0.00495 GNAI3Integrin Signaling 0.505 0.00472 ITGAMcAMP-mediated signaling 0.492 0.00455 GNAI3Superpathway of Inositol Phosphate Compounds 0.481 0.00441 LCKPhospholipase C Signaling 0.47 0.00427 LCKColorectal Cancer Metastasis Signaling 0.458 0.00413 MMP9Signaling by Rho Family GTPases 0.453 0.00407 GNAI3

Page 38 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Protein Ubiquitination Pathway 0.441 0.00394 BAG1Xenobiotic Metabolism Signaling 0.422 0.00372 MAPK14G-Protein Coupled Receptor Signaling 0.42 0.0037 GNAI3Protein Kinase A Signaling 0.313 0.00266 GNAI3

Page 39 of 39



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

Peripheral blood RNA gene expression in children with

pneumococcal meningitis; a prospective case control study


Manuscript ID bmjpo-2017-000092.R1


Date Submitted by the Author: 20-Jul-2017

Complete List of Authors: Kulohoma, Benard ; Centre for Biotechnology and Bioinformatics, University of Nairobi, Nairobi, Kenya;; University of Oxford, Wellcome Trust Centre for Human Genetics, Nuffield Department of Medicine, Marriage, Fiona; 3. Centre for Integrated Genomic Research, University of Manchester, Manchester, UK. Vasieva, Olga; University of Liverpool, Institute of Integrative Biology

Nguyen, Kha; University of Liverpool, Institute of Translational Medicine Mankhambo, Limangeni; 2. Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of Medicine, Blantyre, Malawi. Molyneux, Malcolm; 2. Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of Medicine, Blantyre, Malawi Molyneux, ELizabeth; Paediatrics Day, Philip; 3. Centre for Integrated Genomic Research, University of Manchester, Manchester, UK. Carrol, Enitan; University of Liverpool, Institute of Infection and Global Health



BMJ Paediatrics Open on June 3, 2020 by guest. P

rotected by copyright.http://bm

jpaedsopen.bmj.com

/bm

jpo: first published as 10.1136/bmjpo-2017-000092 on 31 A

ugust 2017. Dow

nloaded from



nlyPeripheral blood RNA gene expression in children with pneumococcal meningitis; a prospective case control study


Manuscript ID bmjpo-2017-000092


Date Submitted by the

Author: 23-May-2017

Complete List of Authors:

Kulohoma, Benard ; Centre for Biotechnology and Bioinformatics,

University of Nairobi, Nairobi, Kenya;; University of Oxford, Wellcome Trust

Centre for Human Genetics, Nuffield Department of Medicine,

Marriage, Fiona; 3. Centre for Integrated Genomic Research,

University of

Manchester, Manchester, UK. Vasieva, Olga; University of Liverpool, Institute of Integrative Biology

Nguyen, Kha; University of Liverpool, Institute of Translational Medicine

Mankhambo, Limangeni; 2. Malawi-Liverpool-Wellcome Trust Clinical

Research Programme, College of Medicine, Blantyre, Malawi.

Molyneux, Malcolm; 2. Malawi-Liverpool-Wellcome Trust Clinical Research

Programme, College of Medicine, Blantyre, Malawi Molyneux, ELizabeth; Paediatrics

Day, Philip; 3. Centre for Integrated Genomic Research, University of

Manchester, Manchester, UK.

Carrol, Enitan; University of Liverpool, Institute of Infection and Global Health


Page 1 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyTitle: Peripheral blood RNA gene expression in children with pneumococcal meningitis: a prospective case control study

Benard W. Kulohoma 1 † *, Fiona Marriage 3 *, Olga Vasieva 4, Limangeni

Mankhambo 2, Kha Nguyen 5, Malcolm E Molyneux 2, Elizabeth M Molyneux 6, Philip

JR Day 3, Enitan D Carrol 2 5 6 ‡

1. Centre for Biotechnology and Bioinformatics, University of Nairobi.

2. Malawi-Liverpool-Wellcome Trust Clinical Research Programme, College of

Medicine, Blantyre, Malawi.

3. Centre for Integrated Genomic Research, University of Manchester, Manchester,

UK.

4. Institute of Integrative Biology, University of Liverpool, Liverpool, UK.

5. Institute of Infection and Global Health, University of Liverpool, Liverpool, UK.

6. Department of Paediatrics, University of Malawi College of Medicine, Blantyre,

Malawi.

*Contributed equally.

† Current address: Wellcome Trust Centre for Human Genetics, University of

Oxford.

‡ Corresponding author

Benard W. Kulohoma ([email protected])

Fiona Marriage ([email protected])

Olga Vasieva ([email protected] )

Limangeni Mankhambo ([email protected])

Kha Nguyen ([email protected])

Malcolm Molyneux ([email protected])

Elizabeth Molyneux ([email protected])

Philip J. R. Day ([email protected])

Enitan D. Carrol ([email protected]) - Corresponding author

Page 2 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyAbstract

Introduction: Invasive pneumococcal disease (IPD), caused by Streptococcus

pneumoniae, is a leading cause of pneumonia, meningitis and septicaemia

worldwide, with increased morbidity and mortality in HIV-infected children.

Objectives: We aimed to compare peripheral blood expression profiles between

HIV-infected and uninfected children with pneumococcal meningitis and controls,

and between survivors and non-survivors, in order to provide insight into the host

inflammatory response leading to poorer outcomes.

Design and setting: Prospective case control observational study in a tertiary

hospital in Malawi

Participants: Children aged 2 months to 16 years with pneumococcal meningitis or

pneumonia.

Methods: We used the human genome HGU133A Affymetrix array to explore

differences in gene expression between cases with pneumococcal meningitis (n=12)

and controls, and between HIV-infected and uninfected cases, and validated gene

expression profiles for 34 genes using real time qPCR (RT qPCR) in an independent

set of cases with IPD (n=229) and controls (n=13). Pathway analysis was used to

explore genes differentially expressed.

Results: Irrespective of underlying HIV infection, cases showed significant up-

regulation compared to controls of: S100 calcium binding protein A12 (S100A12);

vanin-1 (VNN1); arginase, liver (ARG1); matrix metallopeptidase 9 (MMP9); annexin

A3 (ANXA3); interleukin 1 receptor, type II (IL1R2); CD177 molecule (CD177);

S100 calcium binding protein A9 (S100A9), cytoskeleton-associated protein 4

(CKAP4); and glycogenin 1 (GYG1). RT qPCR confirmed differential expression in

keeping with microarray results. There was no differential gene expression in HIV-

infected compared to HIV-uninfected cases, but there was significant up-

regulation of folate receptor 3 (FOLR3), endocytic adaptor protein (NUMB) and

S100A12 in survivors compared non-survivors.

Conclusion: Children with IPD demonstrated increased expression in genes

regulating immune activation, oxidative stress, leucocyte adhesion and migration,

arginine metabolism, and glucocorticoid receptor signalling.

(280 words)

Page 3 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyKey words: Differential expression, children, invasive pneumococcal disease, host

response, pneumococcal meningitis

Key messages:

What is known about the subject

Invasive pneumococcal disease (IPD) caused by Streptococcus pneumoniae is

a leading cause of pneumonia, meningitis and septicaemia worldwide. Globally,

IPD is reported to cause about 11% (0.8 million) of all deaths in children less

than 5 years of age annually. The overall burden of IPD is increased 40-fold in

HIV-infected compared with HIV-uninfected children. However, mechanisms

involved in host response during IPD are not yet fully understood.

What this study adds

We demonstrate for the first time, differences in transcriptional profiles between HIV-

infected and HIV-uninfected children with pneumococcal meningitis (as a

homogeneous disease entity of invasive pneumococcal disease) and healthy

controls. We demonstrate increased expression in cases of genes regulating the

innate immune response, leucocyte migration, glucose homeostasis and endothelial

cell migration.

Introduction

Streptococcus pneumoniae infection is a leading cause of pneumonia, meningitis

and septicaemia worldwide, and results in approximately 1 million deaths in children

under the age of 5 years annually 1. The overall burden of invasive pneumococcal

disease (IPD) is increased 40-fold in HIV-infected compared with HIV-uninfected

children 2. Pneumococcal meningitis is a life-threatening disease with poor prognosis

associated with neurologic complications and a high case-fatality ratio in African

children, which is further increased by HIV co-infection 3.

Gene expression profiling during sepsis provides new insights into the host response

to invasive bacterial disease. Several sepsis studies have demonstrated up-

regulation of pathogen recognition receptors and signal transduction pathways, and

Page 4 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlya down-regulation of zinc homeostasis 4-6. The intricate host inflammatory response

is associated with neuronal and vascular injury, even after cerebrospinal fluid (CSF)

sterilization with antibiotics. Adjunctive new therapies for bacterial meningitis have to

date not shown any conclusive benefit, prompting the need for an improved

understanding of key mechanisms that might reveal potential new therapeutic targets

7. Chronic co-infection may impact gene expression, even among asymptomatic

patients. HIV is a major risk factor for IPD, characterised by waning immunity and

dysregulated physiology among infected individuals, greatly escalating disease

susceptibility and mortality outcomes 8 9, thereby influencing the host’s gene

expression. We examine differences in gene-expression using blood samples from

children with pneumococcal meningitis and matched healthy controls, and also

compare transcriptional profiles between children with pneumococcal meningitis with

and without underlying HIV infection.

Materials and methods

Patients and controls

Children (n=377) were recruited as part of a larger study investigating host

determinants of susceptibility to IPD conducted at Queen Elizabeth Central Hospital,

Blantyre, Malawi, between April 2004 and October 2006 10. Ethical approval was

granted from the College of Medicine Research Committee (COMREC), Malawi and

the Liverpool School of Tropical Medicine Local Research Ethics Committee. Written

informed consent was given prior to recruitment.

We excluded children (n=135) infected with other commonly prevalent microbes

(Salmonella typhimurium, Escherichia coli, Haemophilus influenzae b, Neisseria

meningitidis, Staphylococcus aureus, Streptococcus pyogenes) identified by positive

laboratory culture of blood, CSF or lung aspirate. Cases for the microarray analysis

were HIV treatment-naive children with confirmed pneumococcal meningitis defined

as abnormal CSF white blood cell (WBC) count, >10 WBC/∝l plus one or more of the

following: CSF culture positive for pneumococci, CSF gram stain consistent with

pneumococci, CSF positive for pneumococcal polysaccharide antigen or

pneumococcal DNA. Cases for the RT-qPCR were children with confirmed IPD,

Page 5 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlywhich was defined as: pneumococcal pneumonia (n=40) or pneumococcal meningitis

(n=189), confirmed by either culture, antigen test, or PCR. Controls were healthy

afebrile, malaria aparasitaemic children from the same villages as the index cases,

and were as closely age matched as possible. Microarray analysis was conducted

on 15 samples (12 from cases with pneumococcal meningitis and 3 from controls),

and RT-PCR was conducted on 242 samples (229 from cases with IPD and 13 from

controls), which included all those used for microarray analysis.

There were no known co-infections apart from HIV, and we did not test for other

viruses in the CSF, like CMV or EBV, which have been reported to be associated

with increased mortality in Malawian adults with bacterial meningitis 11 12.

Samples

Whole blood was drawn at admission from consecutive children with IPD. The

methodology for downstream transcriptome analysis from small blood samples has

been previously described 13.

RNA extraction, quantification and hybridization

Total RNA was extracted from whole blood using an optimised method for the

PAXgeneTM blood RNA kit (Qiagen, West Sussex, UK), as previously described 13.

The total RNA concentration (ng/µl) and ratios (260/280 and 260/230) were

measured using a NanoDrop ND-100 UV-vis spectrophotometer (NanoDrop

Technologies, Delaware, USA) and the RNA integrity was assessed using the

Agilent 2100 BioAnalyser (Agilent Technologies, Edinburgh, UK) before and after

concentration.

RNA (2µg) was reverse transcribed into cDNA using the Superscript double-stranded

cDNA synthesis kit (Invitrogen, Paisley, UK) according to the manufacturers’

instructions. The double stranded cDNA was purified using a GeneChipTM sample

clean-up module (Invitrogen, Paisley, UK). The purified cDNA was then biotin

labelled with the ENZO BioArray high yield RNA transcript labelling kit (Affymetrix,

High Wycombe, UK) and cleaned with a cRNA clean-up module (Invitrogen, Paisley,

UK). Aliquots of labelled cRNA (20µg) were fragmented at 94°C for 35 minutes and

then hybridized to a Human Genome U133A GeneChipTM array for 16 hours rotating

Page 6 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyat 60rpm at 45°C in a GeneChipTM Hybridization Oven 640 (Affymetrix, High

Wycombe, UK). Each chip was washed and stained on a GeneChipTM Fluidics

Station 450 (Affymetrix, High Wycombe, UK) and scanned using a GeneChipTM

Scanner 3000 (Affymetrix, High Wycombe, UK) employing standard recommended

protocols (Affymetrix, High Wycombe, UK).

Microarray data analysis

Microarray experiment data were analysed using R and Biocondutor packages 14.

Briefly, the human genome HGU133A array (Affymetrix, High Wycombe, UK) scans

output was pre-processed using the affy package 15. The limma package was used

to evaluate differential expressed genes 16 17. Gene annotations were performed

using hgu133a.db, KEGG.db database packages 18-21. Genes were considered

differentially expressed if they had a Benjamini and Hochberg (BH) adjusted p-value

< 1.5e-3 and > ±2-fold change in gene expression. Bonferroni p-value adjustments

were also performed for comparison. Canonical pathways and functional networks

that involve the differently expressed genes play were determined using the

Ingenuity Pathway Analysis (IPA) catalogue of well characterized metabolic and cell

signalling cascades. Expression data can be accessed using accession number

GSE47172 at the NCBI GEO database.

Reverse transcription for quantitative PCR

RNA samples were DNAse (Ambion, Warrington, UK) treated to remove any

contaminating genomic DNA. RNA (1µg) was reverse transcribed with SuperScript

IITM RNase H- Reverse Transcriptase and oligo (dT)12-18 (0.5µg/µL) following the

manufacturer’s guidelines. The cDNA was stored at -40°C until required.

Real-Time quantitative PCR measurement of target genes

The Human Universal Probe Library (UPL, Roche, Switzerland) employing

proprietary locked nucleic acid (LNA) analogues was used to design qPCR assays to

measure expression levels in genes of interest. Using the Roche Online Assay

Design Centre, specific primers and an associated probe were selected for the

reference and target transcripts. Gene expression was determined using real time

qPCR on a Roche LightCycler 480 (Supplementary Table 2).

Page 7 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyThe following 34 genes were identified from literature and prioritised for real-time

qPCR differential expression analysis between cases and controls: ACSL1, ANXA3,

ATP, BAG1, BPGM, C3AR1, CA4, CD177, CD55, CD59, CEACAM, CFLAR, FOLR3,

GNAI3, GNLY, GYG1, IL1R2, IL1RN, ITGAM, KLRF1, LCK, LCN2, LTF, MAPK14,

MMP9, NUMB, OLAH, PSEN1, RETN, S100A12, SAMSN, SERPINA1, SUB1, and

VNN1. We used a previously described real-time qPCR data normalization method

22.

Statistical analysis of genes prioritised for real-time qPCR differential

expression analysis

First, we derived the relative gene expression in cases compared to controls for the

34 genes under assessment. We used a previously described real-time qPCR data

normalization method 22. Briefly, the amounts of target genes expressed in a sample

were normalized to the average of the three endogenous controls. This is given by

∆Cq, where ∆Cq is determined by subtracting the average endogenous gene Cq

value from the average target gene Cq value:

Cq target gene – Cq average endogenous gene = ∆Cq.

The calculation of relative expression, ∆∆Cq, involves subtraction of ∆Cq value for the

controls from the ∆Cq value for the cases:

∆Cq target gene(case) – ∆Cq target gene(control) = ∆∆Cq.

2-∆∆Cq is the relative expression of the target gene in cases compared to controls.

Next, we examined statistically significant differences in relative gene expression

between cases and controls using the Welch two-sample t-test implemented in the R

package. We generate boxplots to visualise the mean and media relative expression

in cases and controls separately, and the Welch two-sample t-test p-value to show

statistically significant differences in relative gene expression between these two

groups.

Results

Page 8 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

Transcriptional profiles among the cases and controls

In the microarray discovery cohort, there were 12 children with pneumococcal

meningitis (6 male, 6 female, median age 1.1 years) and 3 controls (2 male, 1

female, median age 7 years). The breakdown was as follows: HIV-infected survivors

(n=3), HIV-infected non-survivors (n=3), HIV-uninfected survivors (n=3) and HIV

uninfected non-survivors (n=3) (Supplementary Table 1). The RT-PCT validation

cohort had a median age of 3.09 years.

We examined whether global transcriptional profiles of peripheral blood from children

with pneumococcal meningitis (n=12) were distinct from those of healthy controls

(n=3) randomly selected from a larger dataset by microarray expression profile

analysis 10. In general, there was a marked distinction in differential expression

between cases and controls (Supplementary Figure 1). We identified 10 significantly

differentially expressed genes (Benjamini & Hochberg (BH) adjusted p-value < 1.5e

3 and > ±2-fold change) (Figure 1). We observed significant up-regulation of: S100

calcium binding protein A12 (S100A12); vanin-1 (VNN1); arginase, liver (ARG1);

matrix metallopeptidase 9 (gelatinase B, 92kDa gelatinase, 92kDa type IV

collagenase) (MMP9); annexin A3 (ANXA3); interleukin 1 receptor, type II (IL1R2);

CD177 molecule (CD177); S100 calcium binding protein A9 (S100A9), cytoskeleton

associated protein 4 (CKAP4); and glycogenin 1 (GYG1).

Differential expression, underlying HIV infection and disease outcomes

In the microarray discovery cohort, we did not find any significantly differentially

expressed genes for comparisons between HIV-infected cases and HIV-uninfected

cases or between survivors and non-survivors (Supplementary Figure 2).

RT qPCR validation

RT qPCR validation was performed for a set of 34 prioritised genes selected from

literature for cases (n=229) and controls (n=13). The RT-PCR results are in

agreement with our microarray analysis findings. All the genes are significantly

differentially expressed between the two groups with the exception of guanine

nucleotide-binding protein subunit alpha-13 (GNA13), protein numb (NUMB), and

presenilin-1 (PSEN1) (Figure 2). There was no significantly increased gene

Page 9 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyexpression in HIV-infected compared to HIV-uninfected cases (Supplementary

Figure 3). Interestingly, there was significant up-regulation of folate receptor 3

(FOLR3), NUMB and S100A12 in survivors compared non-survivors (Supplementary

Figure 4). There was wide variability in relative gene expression in cases, but not

controls.

Pathway analysis

Networks were reconstructed using IPA software for the genes differentially

expressed between cases and controls in the microarray experiment, and the

combined microarray and RT qPCR experiments (Figure 3A and 3B). For the genes

defined by the microarray experiments the networks were predominantly related to

granulocyte function, but also including antimicrobial and endothelial activation

responses (Figure 3A). The merged set demonstrated wider networks involving

immune cell activation as well as leucocyte migration and adhesion (Figure 3B). The

canonical pathways mapped by genes defined in the microarray experiment

demonstrated greatest changes in the arginine and granulocyte pathways. Canonical

pathways in the combined microarray and RT qPCR experiments demonstrated

greatest change in leucocytes, and especially neutrophil activation and migration,

Notch and glucocorticoid receptor signalling pathways (Supplementary Tables 3 and

4).

Discussion and conclusion

We have shown significant differences in RNA transcriptional profiles in children with

pneumococcal meningitis compared to controls. Children with pneumococcal

meningitis demonstrated increased expression in genes involved in the inflammatory

response, and glucose and L-arginine metabolic pathways. Dysregulation of these

pathways can lead to an impaired adaptive host response to pneumococcal

infection, thereby contributing to the increased morbidity and mortality. Our findings

in the initial cohort of pneumococcal meningitis were validated in the larger cohort of

all children with IPD, which includes presentations with pneumonia as well as

meningitis. These findings add validity to the initial results from the microarray

experiments, and suggest that the host response observed is systemic, and not

simply localised to the process of blood-brain barrier disruption per se. We observe

Page 10 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlywide variability in relative gene expression in cases, which perhaps reflects

differences in disease onset and robustness of host response within this group.

Elevation of S100A12, VNN, ARG1, MMP9, ANXA3, IL1R2, CD177, S100A9,

CKAP4 and GYG1 in pneumococcal meningitis supports previous findings that have

highlighted the important roles of some of these genes during host pathogen

response. These genes have important interconnected functions for cellular

interactions and response to infection (Figure 3A and Supplementary Table 3), and

play crucial roles during host immune response; cell regulatory processes such as

apoptosis and differentiation; and metabolic processes such as amino acid and

glucose metabolism.

Vanin-1 (VNN1) protein is expressed by human phagocytes, and involved in

leukocyte adhesion and migration 23. The VNN-1 knockout mice model has provided

clear evidence that VNN-1 modulates redox and immune pathways 24. Exposure of

human mononuclear cells to oxidative stress results in up-regulation of human VNN1

and down-regulation of PPAR 25. IL1R2 and IL1RN were up-regulated in cases,

which is consistent with our previous report of the IL-1Ra single-nucleotide

polymorphism rs4251961 playing a key role in the pathophysiology of IPD and in

other human infections 10. Recent reports also demonstrate IL1R2 expression up

regulation in sepsis, being more pronounced in Gram negative than Gram positive

infections 26.

During infection, host production of the cytokine nitric oxide (NO) after non-opsonic

phagocytosis, exerts microbicidal effects 27 28. ARG1 expression is inducible in the

lungs of mice in response to pneumococcal infection 29. Phagocytosis of

pneumococci by macrophages also results in increased production of nitric oxide

synthase 2 (NOS2) dependent production of NO and reactive nitrogen species 30.

Increased plasma arginase activity depletes L-arginine concentrations, the substrate

for NO synthesis, leading to vascular dysfunction during severe sepsis and

supressed NO-mediated microbicidal effects 31. Increased ARG1 activity may also be

a bacterial survival strategy to escape the NO-dependent host antimicrobial immune

response 30.

Page 11 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

Neutrophil-specific glycoprotein CD177 is expressed on a subset of human

neutrophils, and is involved in neutrophil transendothelial migration. A previous

microarray study of purified neutrophils from septic shock patients revealed CD177

mRNA has the highest differential expression between cases and controls 32.

Consistent with our data, that study also demonstrated increased expression of

ARG1, ANXA3, CKAP4, IL1R2, MMP9, and VNN1. ANXA3 promotes endothelial cell

junction integrity in animal models, and endothelial cell motility in vitro. ANXA3 is up

regulated following neuronal injury, which may explain the finding in pneumococcal

meningitis 33.

S100A12 plays a prominent role in the regulation of proinflammatory processes

and immune response by recruiting leukocytes, promoting cytokine and chemokine

production, and regulating leukocyte adhesion and migration 34. The S100A8/A9

heterodimer is expressed by myeloid cells, especially neutrophils, and has a

protective effect in the host response to pneumococcal infection by increasing

circulating neutrophils through increased granulocyte-colony stimulating factor

production 35. It is an antimicrobial peptide, but also plays an important role in

leukocyte migration 36. During infection, S100 proteins stimulate the proinflammatory

immune response through interaction with the immunoglobulin family

transmembrane pattern recognition receptors: Receptor for advanced glycation end

products (RAGE) and Toll-like receptor 4 (TLR4). This leads to NF-κB-mediated pro-

inflammatory response with production of proinflammatory cytokines. This

inflammatory response in turn leads to increased expression of S100 proteins, and

the start of a positive feedback loop. Although as antimicrobial proteins they protect

against infection, they can also have a negative detrimental effect on the host by

amplifying the destructive pro-inflammatory responses. The increased expression in

survivors may be explained by this positive feedback loop 37.

Glucocorticoids also play a significant role in immune response regulation, by

supressing immune and inflammatory responses, and modulating cytokines that

promote host immune responses 38. We speculate that these responses are

important in regulating immune responses to avoid host tissue damage. NUMB

negatively regulates Notch, which in turn attenuates proinflammatory cytokines and

Page 12 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyincreases anti-inflammatory cytokines, which may explain the increase in NUMB

expression in survivors 39 40.

Our results are consistent with previous studies on transcription profiling in other

bacterial diseases, suggesting that some of the mechanisms are not specific to IPD.

Although it could be argued that the lack of age matching in cases versus controls

could cause differential expression due to maturation of the immune system in older

children, the consistency of our results with other studies makes this unlikely 41 42. A

limitation of our study is that the microarray analysis was performed using the

Affymetrix Human Genome U133A array, and sample number was constrained at

the time by cost. The microarray dataset was not large enough to allow all possible

multifactorial models of co-morbidity and disease outcomes to be exhaustively

examined, but sought to validate our findings using RT-qPCR in a larger cohort of

children. Further evaluation is in a larger cohort using whole genome sequencing

would provide more insights on the mechanisms of host response.

In conclusion, this comparative study of gene expression provides mechanistic

insight for IPD in children, and demonstrates significant and widespread immune

activation, with oxidative stress, recruitment of neutrophils, leucocyte adhesion and

migration, activation of antimicrobial peptides, and preservation of endothelial cell

junction integrity.

(2736 words)

Competing interests

None declared

Author contributions

FM, PJRD, EDC conceived and designed the study; LM, EDC collected the clinical

samples; FM performed experiments; BWK, FM, OV, KN, SR, PJRD, EDC

performed the analysis; BWK, FM, PJRD, EDC wrote the first draft of the manuscript;

BWK, FM, SR, LM, KN, OV, MM, EM, PJRD, EDC contributed to writing the

manuscript.

Page 13 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyAcknowledgements

The IPD Study Group (C Antonio, M Chinamale, L Jere, D Mnapo, V Munthali, F

Nyalo, J Simwinga, M Kaole, DL Banda, A Manyika, K Phiri) recruited patients. We

thank the children included in this study, and their parents and guardians for giving

consent for them to participate in the study. We also extend thanks to the nursing

and medical staff, at the Malawi-Liverpool-Wellcome Trust Clinical Research

Programme (MLW), Research Ward, for their contribution to this study. EDC was

supported by a Wellcome Trust Career Development Grant (068026). BWK was

supported by a Wellcome Trust Major Overseas programme award (084679/Z/08/Z).

The funding bodies had no role in collection, analysis or interpretation of data or in

writing the manuscript.

References

1. O'Brien KL, Wolfson LJ, Watt JP, et al. Burden of disease caused by Streptococcus pneumoniae in

children younger than 5 years: global estimates. Lancet 2009;374(9693):893-902. doi: S0140-

6736(09)61204-6 [pii]

10.1016/S0140-6736(09)61204-6 [published Online First: 2009/09/15]

2. Madhi SA, Petersen K, Madhi A, et al. Impact of human immunodeficiency virus type 1 on the

disease spectrum of Streptococcus pneumoniae in South African children. The Pediatric

infectious disease journal 2000;19(12):1141-7.

3. Nyasulu P, Cohen C, De Gouveia L, et al. Increased risk of death in human immunodeficiency virus-

infected children with pneumococcal meningitis in South Africa, 2003-2005. The Pediatric

infectious disease journal 2011;30(12):1075-80. doi: 10.1097/INF.0b013e31822cca05

4. Wong HR, Cvijanovich N, Allen GL, et al. Genomic expression profiling across the pediatric

systemic inflammatory response syndrome, sepsis, and septic shock spectrum. Crit Care Med

2009;37(5):1558-66. doi: 10.1097/CCM.0b013e31819fcc08

5. Wong HR, Freishtat RJ, Monaco M, et al. Leukocyte subset-derived genomewide expression

profiles in pediatric septic shock. Pediatric critical care medicine : a journal of the Society of

Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care

Societies 2010;11(3):349-55. doi: 10.1097/PCC.0b013e3181c519b4

6. Khoo SK, Petillo D, Parida M, et al. Host response transcriptional profiling reveals extracellular

components and ABC (ATP-binding cassette) transporters gene enrichment in typhoid fever-

infected Nigerian children. BMC infectious diseases 2011;11:241. doi: 10.1186/1471-2334-

11-241

7. Hoffman O, Weber RJ. Pathophysiology and treatment of bacterial meningitis. Ther Adv Neurol

Disord 2009;2(6):1-7. doi: 10.1177/1756285609337975

8. Glennie SJ, Williams NA, Heyderman RS. Mucosal immunity in resource-limited setting: is the

battle ground different? Trends in microbiology 2010;18(11):487-93. doi:

10.1016/j.tim.2010.08.002

S0966-842X(10)00139-3 [pii] [published Online First: 2010/09/14]

9. Iwajomo OH, Finn A, Moons P, et al. Deteriorating pneumococcal-specific B-cell memory in

minimally symptomatic African children with HIV infection. The Journal of infectious diseases

2011;204(4):534-43. doi: jir316 [pii]

Page 14 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly10.1093/infdis/jir316 [published Online First: 2011/07/28]

10. Carrol ED, Payton A, Payne D, et al. The IL1RN promoter rs4251961 correlates with IL-1 receptor

antagonist concentrations in human infection and is differentially regulated by GATA-1. J

Immunol 2011;186(4):2329-35. doi: 10.4049/jimmunol.1002402

11. Kelly MJ, Benjamin LA, Cartwright K, et al. Epstein-barr virus coinfection in cerebrospinal fluid is

associated with increased mortality in Malawian adults with bacterial meningitis. The Journal

of infectious diseases 2012;205(1):106-10. doi: 10.1093/infdis/jir707

12. Benjamin LA, Kelly M, Cohen D, et al. Detection of herpes viruses in the cerebrospinal fluid of

adults with suspected viral meningitis in Malawi. Infection 2013;41(1):27-31. doi:

10.1007/s15010-012-0292-z

13. Carrol ED, Salway F, Pepper SD, et al. Successful downstream application of the Paxgene Blood

RNA system from small blood samples in paediatric patients for quantitative PCR analysis.

BMC Immunol 2007;8:20. doi: 10.1186/1471-2172-8-20

14. Gentleman RC, Carey VJ, Bates DM, et al. Bioconductor: open software development for

computational biology and bioinformatics. Genome biology 2004;5(10):R80.1-R80.16.

15. Irizarry R. From CEL files to annotated lists of interesting genes. In: Gentleman R. CV, Dudoit S.,

Irizarry R., Huber W., ed. Bioinformatics and Computational Biology Solutions using R and

Bioconductor. New York: Springer 2005:431-42.

16. Smyth GK. Linear models and empirical Bayes methods for assessing differential expression in

microarray experiments. Statistical Applications in Genetics and Molecular Biology

2004;3(1):3.

17. Smyth GK. Limma: linear models for microarray data. In: Gentleman R. CV, Dudoit S., Irizarry R.,

Huber W., ed. Bioinformatics and Computational Biology Solutions using R and

Bioconductor. New York: Springer 2005:397-420.

18. hgu133a.db: Affymetrix Human Genome U133 Set annotation data (chip hgu133a) [program]. R

package version 2.7.1 version, 2009.

19. annotate: Annotation for microarrays. R package version 1.34.1. [program]. R package version

1.34.1. version, 2008.

20. GO.db: A set of annotation maps describing the entire Gene Ontology [program]. R package

version 2.12 version, 2013.

21. KEGG.db: A set of annotation maps for KEGG [program]. R package version 2.7.1 version, 2009.

22. Vandesompele J, De Preter K, Pattyn F, et al. Accurate normalization of real-time quantitative RT-

PCR data by geometric averaging of multiple internal control genes. Genome Biol

2002;3(7):RESEARCH0034.

23. Meghari S, Berruyer C, Lepidi H, et al. Vanin-1 controls granuloma formation and macrophage

polarization in Coxiella burnetii infection. Eur J Immunol 2007;37(1):24-32. doi:

10.1002/eji.200636054

24. Ferreira DW, Naquet P, Manautou JE. Influence of Vanin-1 and Catalytic Products in Liver During

Normal and Oxidative Stress Conditions. Curr Med Chem 2015;22(20):2407-16.

25. Zhang B, Lo C, Shen L, et al. The role of vanin-1 and oxidative stress-related pathways in

distinguishing acute and chronic pediatric ITP. Blood 2011;117(17):4569-79. doi:

10.1182/blood-2010-09-304931

26. Lang Y, Jiang Y, Gao M, et al. Interleukin-1 Receptor 2: A New Biomarker for Sepsis Diagnosis and

Gram-Negative/Gram-Positive Bacterial Differentiation. Shock 2017;47(1):119-24. doi:

10.1097/SHK.0000000000000714

27. Ulett GC, Adderson EE. Nitric oxide is a key determinant of group B streptococcus-induced

murine macrophage apoptosis. The Journal of infectious diseases 2005;191(10):1761-70. doi:

10.1086/429693

28. Gross A, Spiesser S, Terraza A, et al. Expression and bactericidal activity of nitric oxide synthase in

Brucella suis-infected murine macrophages. Infection and immunity 1998;66(4):1309-16.

Page 15 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly29. Knippenberg S, Brumshagen C, Aschenbrenner F, et al. Arginase 1 activity worsens lung-

protective immunity against Streptococcus pneumoniae infection. Eur J Immunol

2015;45(6):1716-26. doi: 10.1002/eji.201445419

30. Marriott HM, Ali F, Read RC, et al. Nitric oxide levels regulate macrophage commitment to

apoptosis or necrosis during pneumococcal infection. FASEB J 2004;18(10):1126-8. doi:

10.1096/fj.03-1450fje

31. Darcy CJ, Woodberry T, Davis JS, et al. Increased plasma arginase activity in human sepsis:

association with increased circulating neutrophils. Clinical chemistry and laboratory medicine

: CCLM / FESCC 2014;52(4):573-81. doi: 10.1515/cclm-2013-0698

32. Demaret J, Venet F, Plassais J, et al. Identification of CD177 as the most dysregulated parameter

in a microarray study of purified neutrophils from septic shock patients. Immunol Lett

2016;178:122-30. doi: 10.1016/j.imlet.2016.08.011

33. Chong KW, Chen MJ, Koay ES, et al. Annexin A3 is associated with cell death in lactacystin-

mediated neuronal injury. Neurosci Lett 2010;485(2):129-33. doi:

10.1016/j.neulet.2010.08.089

34. Goyette J, Geczy CL. Inflammation-associated S100 proteins: new mechanisms that regulate

function. Amino Acids 2011;41(4):821-42. doi: 10.1007/s00726-010-0528-0

35. De Filippo K, Neill DR, Mathies M, et al. A new protective role for S100A9 in regulation of

neutrophil recruitment during invasive pneumococcal pneumonia. FASEB J 2014;28(8):3600-

8. doi: 10.1096/fj.13-247460

36. Raquil MA, Anceriz N, Rouleau P, et al. Blockade of antimicrobial proteins S100A8 and S100A9

inhibits phagocyte migration to the alveoli in streptococcal pneumonia. J Immunol

2008;180(5):3366-74.

37. Zackular JP, Chazin WJ, Skaar EP. Nutritional Immunity: S100 Proteins at the Host-Pathogen

Interface. The Journal of biological chemistry 2015;290(31):18991-8. doi:

10.1074/jbc.R115.645085

38. Sakai A, Han J, Cato AC, et al. Glucocorticoids synergize with IL-1beta to induce TLR2 expression

via MAP Kinase Phosphatase-1-dependent dual Inhibition of MAPK JNK and p38 in epithelial

cells. BMC molecular biology 2004;5:2. doi: 10.1186/1471-2199-5-2

39. Kueanjinda P, Roytrakul S, Palaga T. A Novel Role of Numb as A Regulator of Pro-inflammatory

Cytokine Production in Macrophages in Response to Toll-like Receptor 4. Scientific reports

2015;5:12784. doi: 10.1038/srep12784

40. Singla DK, Wang J, Singla R. Primary human monocytes differentiate into M2 macrophages and

involve Notch-1 pathway. Can J Physiol Pharmacol 2017;95(3):288-94. doi: 10.1139/cjpp-

2016-0319

41. Langley RJ, Tsalik EL, van Velkinburgh JC, et al. An integrated clinico-metabolomic model

improves prediction of death in sepsis. Sci Transl Med 2013;5(195):195ra95. doi:

10.1126/scitranslmed.3005893

42. Liu K, Chen L, Kaur R, et al. Transcriptome signature in young children with acute otitis media due

to Streptococcus pneumoniae. Microbes Infect 2012;14(7-8):600-9. doi:

10.1016/j.micinf.2012.01.006

Figure and Table legends

Page 16 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyFigure 1: Distribution plot of the differentially expressed genes. The significantly

differentially expressed genes are shown in red colour. The significance threshold

(P<1.5e-3) is indicated by a dashed red line, and a fold change threshold of more

than 2 is shown by the dashed vertical lines. The green line shows P<0.05. (A)

Shows results for un-adjusted p-values, (B) results for Benjamini and Hochberg (BH)

adjusted p-values, and (C) stringent Bonferroni adjusted p-values, which represents

an over-correction.

Figure 2: Validation of RNA transcription profile differential expression using

Real Time quantitative PCR. Relative gene expression in cases compared to

controls for 34 genes assessed. The black line shows the boxplot median. The red

dot shows the mean, and the Welch two-sample t-test p-value is shown on the top

right corner.

Figure 3: The gene network for significantly differentially expressed

genes in the cases. Networks reconstructed to support the most direct connectivity

between genes differentially expressed between cases and controls. IPA (Qiagen)

software's Core analysis application has been used to perform an automatic

graphical reconstruction of the network via utilisation of IPA's Knowledge base

database of protein interactions. The meaning of links and shapes is explained in the

inserted legend. Functional connections are presented in blue, information

connections to the associated categories in pink and grey. (A) Network reconstructed

only for the DE genes identified by microarray analysis. (B) Network reconstructed

for the merged datasets of DE genes identified via microarray (red) and PCR

analysis (green). White blocks correspond to nodes added by IPA's network editor

automatically to ensure network connectivity.

Supplementary Figure 1: Children with pneumococcal meningitis and healthy

controls have distinct transcriptional profiles. A heatmap of the top 100 (P<0.05)

differentially expressed genes compared in cases and controls. The cases and

controls were distinguished into two distinct clusters highlighting their distinct

transcriptional profiles.

Page 17 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlySupplementary Figure 2: Comparison of microarray transcriptional profiles of

HIV infected and HIV-uninfected and survivors and non- survivors. There were

no significantly differentially expressed genes for any of these comparisons.

Supplementary Figure 3: Comparison of RT-qPCT gene expression for HIV

infected and HIV-uninfected. Relative gene expression in cases compared to

controls for 34 genes assessed. There were no significantly differentially expressed

genes for all of these comparisons. Relative gene expression in cases compared to

controls. The black line shows the boxplot median. The red dot shows the mean, and

the Welch two-sample t-test p-value is shown on the top right corner.

Supplementary Figure 4: Comparison of RT-qPCT gene expression for

survivors and non-survivors. Relative gene expression in cases compared to

controls for 34 genes assessed. There were three significantly differentially

expressed genes: FOLR3, NUMB, and S100A12. Relative gene expression in cases

compared to controls. The black line shows the boxplot median. The red dot shows

the mean, and the Welch two-sample t-test p-value is shown on the top right corner.

Supplementary Table 1: Clinical data on children with pneumococcal meningitis and healthy controls. Supplementary Table 2: Primer sequences and the UPL probe number for each RT qPCR assay. Supplementary Table 3: Canonical pathways of differentially expressed genes between cases and controls in microarray experiment Supplementary Table 4: Canonical pathways of differentially expressed genes between cases and controls in combined microarray and RT- qPCR experiments

Page 18 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

94x24mm (300 x 300 DPI)

Page 19 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

177x177mm (300 x 300 DPI)

Page 20 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

361x481mm (300 x 300 DPI)

Page 21 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

270x203mm (300 x 300 DPI)

Page 22 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

117x42mm (300 x 300 DPI)

Page 23 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

177x177mm (300 x 300 DPI)

Page 24 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nly

177x177mm (300 x 300 DPI)

Page 25 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



nlyTable S1: Clinical data on children with pneumococcal meningitis and

healthy controls

Sample ID

Age

(years)

Sex Case/control HIV status

Survival status

Previous antibiotics use

Duration of symptoms (days)

94 0.4 F Case Positive Died No 2


Died Yes 6


Died No 1


Alive No 3


Alive No 2


Alive Yes 3

82 13.0 M Case Negative Died No 7


Died Yes 3


Died No 5


Alive No 3


Alive No 3


Alive Yes 3


Alive N/A N/A


Alive N/A N/A

C13 7.0 F Control Negative

Alive N/A N/A

Page 26 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Table S2: Primer sequences and the UPL probe number for each RT qPCR assay


ACSL1 Acyl-CoA synthetase long-chain family member 1 ccactgtgctgacgttcct gcactctgtctgtccgtatcc 26 NM_001995

ANXA3 Annexin A3 ttgttaaggaatatcaagcagcat gccagagagatcacccttca 17 NM_005139

C4A Complement component 4A (Rodgers blood group) ggaccctgaagacgaaattg gaagccctcgttcacctg 75 NM_007293, NM_001252204

C3AR1 complement component 3a receptor 1

aagaaagcaaggcagtccatt

cgtgtgagctcctcactgaa

16

NM_004054

ILR2A ILR2A interleukin 1 receptor, type II tacgcaccacagtcaaggaa aagaaggccagtgaaagtgg 2 NM_004633, NM_173343

MAPK14



ccatgagatgggtcaccag

65

NM_139012, NM_139014, NM_001315, NM_139013

MMP9

matrix metallopeptidase 9 (gelatinase B, 92kDa gelatinase, 92kDa type IV collagenase)



6

NM_004994

CD177 CD177 molecule gcctctccctgatctcctaca caaggatcctgggtctgc 79 NM_020406

CD55 CD55 molecule, decay accelerating factor for complement (Cromer blood group) aataatgatgaaggagagtggagtg tggtgggaccttggaagtta 77 NM_000574

CEACAM1 Carcinoembryonic antigen-related cell adhesion molecule 1 (biliary glycoprotein) taggcctaagaggctttctcc tgagggaaagtatgcacagtccgtgtc 80

NM_001712, NM_001184816, NM_001184813, NM_001184815, NM_001205344, NM_001024912

GYG1 Glycogenin 1 ctgaaacagcacaggaccac atcgccactgtccaagacat 75 NM_004130

FOLR3 Folate receptor 3 (gamma) cacaaaggctggaattggac cttgaaggagtggctccaga 19 NM_000804

ITGAM Integrin, alpha M (complement component 3 receptor 3 subunit) tctgaagaccattcagaaccag ggagctgctacttcctgtctg 7 NM_000632

LTF Lactotransferrin ctaatctgaaaaagtgctcaacctc gccatcttcttcggttttacttc 79 NM_002343, NM_001199149

OLAH Oleoyl-ACP hydrolase accgagatcgctccactg ggagaatcaagtgagattttaatagga 19 NM_018324, NM_001039702

RETN Resistin tgcaggatgaaagctctctg catggagcacagggtcttg 45 NM_020415

Page 27 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from




S100A12 S100 calcium binding protein A12 tcatatccctggtagccattg acctactctttgtgggtgtggt 44 NM_005621

SERPINA1 Serpin peptidase inhibitor, clade A (alpha-1 antiproteinase, antitrypsin), member 1 aatggggctgacctctcc gtcagcacagccttatgcac 82 NM_001002235, NM_001002236,

NM_000295

SAMSN1 SAM domain, SH3 domain and nuclear localization signals 1 tgaaaatctgtctgacatggtaca tagttgggaatgcgtgttca 21 NM_022136, NM_001256370

VNN1 Vanin 1 tcctgaggtgttgctgagtg agcgtccgtcagttgacac 80 NM_004666

CD59 CD59 molecule, complement regulatory protein caaggagggtctgtcctgtt

agcactgcaggct[a or g]tgacc 66

NM_000611, NM_203331, NM_203330, NM_203329

CFLAR CASP8 and FADD-like apoptosis regulator gcaatccaaaagagtctcaagg

tgagcgccaagctgttcc 7 NM_003879

KLRF1 Killer cell lectin-like receptor subfamily F, member 1

tgatctccttgatcctgttgg

tcttcttgtgccattattcacttt 47 NM_016523

LCN2 Homo sapiens lipocalin 2 tcacctccgtcctgtttagg

aggtaactcgttaatccagggtaa 61 NM_005564

NUMB Numb homolog (Drosophila) gaggaagactgatttccccatt

gcacagggagctgatgctc 16

NM_001005745, NM_003744, NM_001005744, NM_001005743

Page 28 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Supplementary Table 3: Canonical pathways of differentially expressedgenes between cases and controls in microarray experiment Ingenuity Canonical Pathways -log(p-value) Ratio MoleculesLXR/RXR Activation 2.77 0.0165 IL1R2,MMP9Arginine Degradation I (Arginase Pathway) 2.69 0.25 ARG1Urea Cycle 2.51 0.167 ARG1Arginine Degradation VI (Arginase 2 Pathway) 2.51 0.167 ARG1Glycogen Biosynthesis II (from UDP-D-Glucose) 2.51 0.167 GYG1Granulocyte Adhesion and Diapedesis 2.51 0.0121 IL1R2,MMP9Hepatic Fibrosis / Hepatic Stellate Cell Activation 2.43 0.011 IL1R2,MMP9Airway Pathology in Chronic Obstructive Pulmonary Disease 2.38 0.125 MMP9Citrulline Biosynthesis 2.38 0.125 ARG1Role of IL-17A in Psoriasis 2.17 0.0769 S100A9Superpathway of Citrulline Metabolism 2.14 0.0714 ARG1Inhibition of Angiogenesis by TSP1 1.79 0.0312 MMP9Inhibition of Matrix Metalloproteases 1.71 0.0263 MMP9Neuroprotective Role of THOP1 in Alzheimer's Disease 1.69 0.025 MMP9IL-10 Signaling 1.46 0.0147 IL1R2Glioma Invasiveness Signaling 1.45 0.0143 MMP9Bladder Cancer Signaling 1.36 0.0116 MMP9PPAR Signaling 1.34 0.0111 IL1R2HIF1α Signaling 1.25 0.00885 MMP9p38 MAPK Signaling 1.23 0.00855 IL1R2Pancreatic Adenocarcinoma Signaling 1.23 0.00847 MMP9Atherosclerosis Signaling 1.21 0.00806 MMP9IL-6 Signaling 1.2 0.00787 IL1R2Ovarian Cancer Signaling 1.15 0.00699 MMP9Relaxin Signaling 1.13 0.00671 MMP9Hepatic Cholestasis 1.11 0.00637 IL1R2PPARα/RXRα Activation 1.08 0.00602 IL1R2Agranulocyte Adhesion and Diapedesis 1.06 0.00571 MMP9NF-κB Signaling 1.06 0.00565 IL1R2Regulation of the Epithelial-Mesenchymal Transition Pathway 1.03 0.00535 MMP9ILK Signaling 1.02 0.00521 MMP9IL-8 Signaling 1.01 0.0051 MMP9Leukocyte Extravasation Signaling 0.996 0.00488 MMP9LPS/IL-1 Mediated Inhibition of RXR Function 0.99 0.00481 IL1R2Role of Osteoblasts, Osteoclasts and Chondrocytes in Rheumatoid Arthritis 0.954 0.00441 IL1R2Colorectal Cancer Metastasis Signaling 0.928 0.00413 MMP9Glucocorticoid Receptor Signaling 0.862 0.00352 IL1R2Role of Macrophages, Fibroblasts and Endothelial Cells in Rheumatoid Arthritis 0.84 0.00333 IL1R2Axonal Guidance Signaling 0.686 0.00226 MMP9

Page 29 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Ingenuity Canonical Pathways -log(p-value) Ratio MoleculesComplement System 6.35 0.111 CD55,CD59,ITGAM,C3AR1Granulocyte Adhesion and Diapedesis 5.02 0.0303 IL1R2,GNAI3,ITGAM,IL1RN,MMP9LXR/RXR Activation 4.24 0.0331 IL1R2,IL1RN,SERPINA1,MMP9IL-10 Signaling 3.64 0.0441 IL1R2,MAPK14,IL1RNLeukocyte Extravasation Signaling 3.36 0.0195 GNAI3,MAPK14,ITGAM,MMP9p38 MAPK Signaling 2.95 0.0256 IL1R2,MAPK14,IL1RNAtherosclerosis Signaling 2.88 0.0242 IL1RN,SERPINA1,MMP9IL-6 Signaling 2.85 0.0236 IL1R2,MAPK14,IL1RNInhibition of Angiogenesis by TSP1 2.84 0.0625 MAPK14,MMP9Glucocorticoid Receptor Signaling 2.83 0.0141 IL1R2,MAPK14,BAG1,IL1RNIL-17A Signaling in Fibroblasts 2.77 0.0571 MAPK14,LCN2Notch Signaling 2.72 0.0541 NUMB,PSEN1Acute Phase Response Signaling 2.5 0.0179 MAPK14,IL1RN,SERPINA1Amyloid Processing 2.46 0.04 MAPK14,PSEN1Agranulocyte Adhesion and Diapedesis 2.45 0.0171 GNAI3,IL1RN,MMP9NF-κB Signaling 2.44 0.0169 IL1R2,LCK,IL1RNMolecular Mechanisms of Cancer 2.42 0.0109 GNAI3,MAPK14,CFLAR,PSEN1IL-8 Signaling 2.31 0.0153 GNAI3,ITGAM,MMP9LPS/IL-1 Mediated Inhibition of RXR Function 2.24 0.0144 IL1R2,IL1RN,ACSL1CCR5 Signaling in Macrophages 2.21 0.0299 GNAI3,MAPK14Chemokine Signaling 2.2 0.0294 GNAI3,MAPK14Caveolar-mediated Endocytosis Signaling 2.16 0.0282 CD55,ITGAMArginine Degradation I (Arginase Pathway) 2.15 0.25 ARG1Acetate Conversion to Acetyl-CoA 2.15 0.25 ACSL1Toll-like Receptor Signaling 2.14 0.0274 MAPK14,IL1RNRole of Osteoblasts, Osteoclasts and Chondrocytes in Rheumatoid Arthritis 2.14 0.0132 IL1R2,MAPK14,IL1RNIL-15 Signaling 2.1 0.0263 LCK,MAPK14Rapoport-Luebering Glycolytic Shunt 2.06 0.2 BPGMPEDF Signaling 2.02 0.0238 MAPK14,CFLARUrea Cycle 1.98 0.167 ARG1Arginine Degradation VI (Arginase 2 Pathway) 1.98 0.167 ARG1Glycogen Biosynthesis II (from UDP-D-Glucose) 1.98 0.167 GYG1PPAR Signaling 1.96 0.0222 IL1R2,IL1RNIL-1 Signaling 1.95 0.022 GNAI3,MAPK14Cholecystokinin/Gastrin-mediated Signaling 1.87 0.0198 MAPK14,IL1RNAirway Pathology in Chronic Obstructive Pulmonary Disease 1.86 0.125 MMP9Citrulline Biosynthesis 1.86 0.125 ARG1Corticotropin Releasing Hormone Signaling 1.82 0.0187 GNAI3,MAPK14Role of Macrophages, Fibroblasts and Endothelial Cells in Rheumatoid Arthritis 1.81 0.01 IL1R2,MAPK14,IL1RNPaxillin Signaling 1.8 0.0182 MAPK14,ITGAMHIF1α Signaling 1.78 0.0177 MAPK14,MMP9Role of Tissue Factor in Cancer 1.73 0.0168 LCK,MAPK14FXR/RXR Activation 1.69 0.016 IL1RN,SERPINA1CCR3 Signaling in Eosinophils 1.69 0.0159 GNAI3,MAPK14

Supplementary Table 4: Canonical pathways of differentially expressed genes between cases and controls in combined microarrayand RT- qPCR experiments

Page 30 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



GNRH Signaling 1.68 0.0157 GNAI3,MAPK14Role of IL-17A in Psoriasis 1.65 0.0769 S100A9Fatty Acid Activation 1.65 0.0769 ACSL1Superpathway of Citrulline Metabolism 1.61 0.0714 ARG1IL-12 Signaling and Production in Macrophages 1.58 0.0139 MAPK14,SERPINA1Parkinson's Signaling 1.56 0.0625 MAPK14Relaxin Signaling 1.55 0.0134 GNAI3,MMP9γ-linolenate Biosynthesis II (Animals) 1.53 0.0588 ACSL1Mitochondrial L-carnitine Shuttle Pathway 1.53 0.0588 ACSL1Oxidative Ethanol Degradation III 1.53 0.0588 ACSL1Hepatic Cholestasis 1.51 0.0127 IL1R2,IL1RNPPARα/RXRα Activation 1.47 0.012 IL1R2,MAPK14Tec Kinase Signaling 1.46 0.0119 GNAI3,LCKEthanol Degradation IV 1.44 0.0476 ACSL1Differential Regulation of Cytokine Production in Intestinal Epithelial Cells by IL-17A and IL-17F1.4 0.0435 LCN2Role of NFAT in Regulation of the Immune Response 1.4 0.0111 GNAI3,LCKHepatic Fibrosis / Hepatic Stellate Cell Activation 1.4 0.011 IL1R2,MMP9Dendritic Cell Maturation 1.39 0.011 MAPK14,IL1RNEndothelin-1 Signaling 1.39 0.0109 GNAI3,MAPK14IL-22 Signaling 1.38 0.0417 MAPK14Glycolysis I 1.38 0.0417 BPGMRegulation of the Epithelial-Mesenchymal Transition Pathway 1.37 0.0107 MMP9,PSEN1IL-17A Signaling in Gastric Cells 1.37 0.04 MAPK14Role of JAK family kinases in IL-6-type Cytokine Signaling 1.37 0.04 MAPK14Gluconeogenesis I 1.37 0.04 BPGMRole of NFAT in Cardiac Hypertrophy 1.36 0.0106 GNAI3,MAPK14Production of Nitric Oxide and Reactive Oxygen Species in Macrophages 1.35 0.0104 MAPK14,SERPINA1Clathrin-mediated Endocytosis Signaling 1.33 0.0102 NUMB,SERPINA1Thrombin Signaling 1.32 0.01 GNAI3,MAPK14Fatty Acid β-oxidation I 1.29 0.0333 ACSL14-1BB Signaling in T Lymphocytes 1.28 0.0323 MAPK14G Protein Signaling Mediated by Tubby 1.28 0.0323 LCKEthanol Degradation II 1.26 0.0312 ACSL1Systemic Lupus Erythematosus Signaling 1.26 0.00922 LCK,IL1RNCoagulation System 1.22 0.0286 SERPINA1Cardiac Hypertrophy Signaling 1.2 0.00862 GNAI3,MAPK14April Mediated Signaling 1.19 0.0263 MAPK14Inhibition of Matrix Metalloproteases 1.19 0.0263 MMP9Role of PKR in Interferon Induction and Antiviral Response 1.17 0.025 MAPK14B Cell Activating Factor Signaling 1.17 0.025 MAPK14Neuroprotective Role of THOP1 in Alzheimer's Disease 1.17 0.025 MMP9Stearate Biosynthesis I (Animals) 1.17 0.025 ACSL1Role of Hypercytokinemia/hyperchemokinemia in the Pathogenesis of Influenza 1.16 0.0244 IL1RNUVC-Induced MAPK Signaling 1.15 0.0238 MAPK14iNOS Signaling 1.14 0.0233 MAPK14Primary Immunodeficiency Signaling 1.13 0.0227 LCKGraft-versus-Host Disease Signaling 1.13 0.0227 IL1RN

Page 31 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Role of Cytokines in Mediating Communication between Immune Cells 1.06 0.0192 IL1RNRetinoic acid Mediated Apoptosis Signaling 1.01 0.0172 CFLARMSP-RON Signaling Pathway 1.01 0.0169 ITGAMCalcium-induced T Lymphocyte Apoptosis 0.999 0.0167 LCKIL-2 Signaling 0.973 0.0156 LCKUVB-Induced MAPK Signaling 0.96 0.0152 MAPK14Role of IL-17A in Arthritis 0.954 0.0149 MAPK14EGF Signaling 0.948 0.0147 MAPK14Melatonin Signaling 0.942 0.0145 GNAI3Glioma Invasiveness Signaling 0.936 0.0143 MMP9Role of MAPK Signaling in the Pathogenesis of Influenza 0.936 0.0143 MAPK14ErbB4 Signaling 0.936 0.0143 PSEN1GPCR-Mediated Integration of Enteroendocrine Signaling Exemplified by an L Cell 0.93 0.0141 GNAI3Role of PI3K/AKT Signaling in the Pathogenesis of Influenza 0.919 0.0137 GNAI3Ephrin B Signaling 0.919 0.0137 GNAI3STAT3 Pathway 0.919 0.0137 MAPK14BMP signaling pathway 0.913 0.0135 MAPK14CD40 Signaling 0.897 0.013 MAPK14IL-17A Signaling in Airway Cells 0.897 0.013 MAPK14ATM Signaling 0.887 0.0127 MAPK14FLT3 Signaling in Hematopoietic Progenitor Cells 0.872 0.0122 MAPK14Communication between Innate and Adaptive Immune Cells 0.872 0.0122 IL1RNAltered T Cell and B Cell Signaling in Rheumatoid Arthritis 0.867 0.012 IL1RNGPCR-Mediated Nutrient Sensing in Enteroendocrine Cells 0.862 0.0119 GNAI3α-Adrenergic Signaling 0.857 0.0118 GNAI3IL-17 Signaling 0.857 0.0118 MAPK14Neuregulin Signaling 0.852 0.0116 PSEN1LPS-stimulated MAPK Signaling 0.852 0.0116 MAPK14NF-κB Activation by Viruses 0.852 0.0116 LCKBladder Cancer Signaling 0.852 0.0116 MMP9TGF-β Signaling 0.848 0.0115 MAPK14G Beta Gamma Signaling 0.843 0.0114 GNAI3Factors Promoting Cardiogenesis in Vertebrates 0.838 0.0112 MAPK14FGF Signaling 0.834 0.0111 MAPK14Death Receptor Signaling 0.83 0.011 CFLARReelin Signaling in Neurons 0.825 0.0109 LCKErbB Signaling 0.804 0.0103 MAPK14CTLA4 Signaling in Cytotoxic T Lymphocytes 0.8 0.0102 LCKCDK5 Signaling 0.8 0.0102 MAPK14RANK Signaling in Osteoclasts 0.796 0.0101 MAPK14Antioxidant Action of Vitamin C 0.796 0.0101 MAPK14UVA-Induced MAPK Signaling 0.788 0.0099 MAPK14Virus Entry via Endocytic Pathways 0.784 0.0098 CD55SAPK/JNK Signaling 0.78 0.00971 LCKMouse Embryonic Stem Cell Pluripotency 0.772 0.00952 MAPK14Type I Diabetes Mellitus Signaling 0.769 0.00943 MAPK14T Cell Receptor Signaling 0.761 0.00926 LCK

Page 32 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Androgen Signaling 0.754 0.00909 GNAI3p53 Signaling 0.751 0.00901 MAPK14Axonal Guidance Signaling 0.741 0.00452 GNAI3,MMP9iCOS-iCOSL Signaling in T Helper Cells 0.73 0.00855 LCKPancreatic Adenocarcinoma Signaling 0.726 0.00847 MMP9Fc Epsilon RI Signaling 0.723 0.0084 MAPK14Renin-Angiotensin Signaling 0.723 0.0084 MAPK14Natural Killer Cell Signaling 0.72 0.00833 LCKfMLP Signaling in Neutrophils 0.72 0.00833 GNAI3Gαi Signaling 0.72 0.00833 GNAI3Sphingosine-1-phosphate Signaling 0.713 0.0082 GNAI3Type II Diabetes Mellitus Signaling 0.704 0.008 ACSL1CD28 Signaling in T Helper Cells 0.701 0.00794 LCKPKCθ Signaling in T Lymphocytes 0.698 0.00787 LCKCdc42 Signaling 0.692 0.00775 MAPK14p70S6K Signaling 0.689 0.00769 GNAI3Th1 Pathway 0.689 0.00769 PSEN1Role of Pattern Recognition Receptors in Recognition of Bacteria and Viruses 0.686 0.00763 C3AR1HMGB1 Signaling 0.686 0.00763 MAPK14P2Y Purigenic Receptor Signaling Pathway 0.686 0.00763 GNAI3Synaptic Long Term Depression 0.66 0.00714 GNAI3Ovarian Cancer Signaling 0.652 0.00699 MMP9Th2 Pathway 0.644 0.00685 PSEN1Regulation of eIF4 and p70S6K Signaling 0.624 0.00649 MAPK14Dopamine-DARPP32 Feedback in cAMP Signaling 0.614 0.00633 GNAI3CXCR4 Signaling 0.6 0.0061 GNAI3Mitochondrial Dysfunction 0.598 0.00606 PSEN1Gap Junction Signaling 0.595 0.00602 GNAI3Germ Cell-Sertoli Cell Junction Signaling 0.589 0.00592 MAPK14RhoGDI Signaling 0.582 0.00581 GNAI3Ephrin Receptor Signaling 0.582 0.00581 GNAI3Sertoli Cell-Sertoli Cell Junction Signaling 0.58 0.00578 MAPK14Th1 and Th2 Activation Pathway 0.565 0.00556 PSEN1CREB Signaling in Neurons 0.561 0.00549 GNAI3B Cell Receptor Signaling 0.561 0.00549 MAPK14RAR Activation 0.551 0.00535 MAPK14AMPK Signaling 0.549 0.00532 MAPK14NRF2-mediated Oxidative Stress Response 0.545 0.00526 MAPK14ILK Signaling 0.541 0.00521 MMP93-phosphoinositide Biosynthesis 0.539 0.00518 LCKBreast Cancer Regulation by Stathmin1 0.523 0.00495 GNAI3Integrin Signaling 0.505 0.00472 ITGAMcAMP-mediated signaling 0.492 0.00455 GNAI3Superpathway of Inositol Phosphate Compounds 0.481 0.00441 LCKPhospholipase C Signaling 0.47 0.00427 LCKColorectal Cancer Metastasis Signaling 0.458 0.00413 MMP9Signaling by Rho Family GTPases 0.453 0.00407 GNAI3

Page 33 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from



Protein Ubiquitination Pathway 0.441 0.00394 BAG1Xenobiotic Metabolism Signaling 0.422 0.00372 MAPK14G-Protein Coupled Receptor Signaling 0.42 0.0037 GNAI3Protein Kinase A Signaling 0.313 0.00266 GNAI3

Page 34 of 34



123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960



j.com/


jpo-2017-000092 on 31 August 2017. D

ownloaded from


Documents

BMJ Paediatrics Open is committed to open peer review. As ... · Confidential: For Review Only 2 33 Abstract 34 Introduction: Invasive pneumococcal disease (IPD), caused by Streptococcus