RESEARCH POSTER PRESENTATION DESIGN © 2015

www.PosterPresentations.com

Early Onset Sepsis (EOS) – need for novel diagnostic methods

• 47% of EOS cases and 92% of EOS deaths occur in preterm infants1

• Primarily intrauterine route of infection in spontaneous premature birth

• Diagnostic difficulty due to nonspecific signs, inadequate sensitivity of blood culture and low PPV of blood count indices and inflammatory markers2-5

• Both delayed diagnosis of sepsis and prolonged early antibiotic exposure have serious adverse outcomes.6-8

16S rRNA gene amplicon sequencing of umbilical cord blood is a culture-independent method of pathogen detection that may improve EOS diagnosis.

But is cord blood sterile?

• Bacterial populations demonstrated in placenta and amniotic fluid9, 10

• Few reports of bacteria isolated from cord blood10-12

INTRODUCTION

CONCLUSIONS

REFERENCES1. Weston EJ et al. The Burden of Invasive Early-onset Neonatal Sepsis in the United States, 2005-2008. Pediatr Infect Dis J 2011;30:937–941.

2. Polin RA and the Committee on Fetus and Newborn. Management of Neonates with Suspected or Proven Early-Onset Bacterial Sepsis. Pediatrics 2012;129:1006-1015.

3. Connell TG et al. How reliable is a negative blood culture result? Volume of blood submitted for culture in routine practice in a children’s hospital. Pediatrics 2007;119:891–896.

4. Hornik CP et al. Use of the Complete Blood Cell Count in Early-onset Neonatal Sepsis. Pediatr Infect Dis J 2012;31:799-802.

5. Benitz WE et al. Serial C reactive Protein Levels in the Diagnosis of Neonatal Infection. Pediatrics 1998;102(4).

6. Kumar A et al. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006 Jun;34(6):1589-96.

7. Kuppula VS et al. Prolonged initial empirical antibiotic treatment is associated with adverse outcomes in premature infants. J Pediatr2011;159:720-725.

8. Cotton CM et al. Prolonged duration of initial empirical antibiotic treatment is associated with increased rates of necrotizing enterocolitis and death for extremely low birth weight infants. Pediatrics 2009;123:58-66.

9. Aagaard K et al. The placenta harbors a unique microbiome. Sci Transl Med 2014;6: 237-65.

10. Wang X et al. Comparative Microbial Analysis of Paired Amniotic Fluid and Cord Blood from Pregnancies Complicated by Preterm Birth and Early-Onset Neonatal Sepsis. PLoS ONE 2013;8(2):e56131.

11. Goldenberg RL et al. The Alabama Preterm Birth Study: umbilical cord blood Ureaplasma urealyticum and Mycoplasma hominis cultures in very preterm newborn infants. Am J Obstet Gynecol 2008;198(1):43-e1.

12. Jiménez E et al. Isolation of commensal bacteria from umbilical cord blood of healthy neonates born by cesarean section. Curr Microbiol. 2005 Oct;51(4):270-4. Epub 2005 Sep 20.

13. Green SJ et al. Deconstructing the polymerase chain reaction: understanding and correcting bias associated with primer degeneracies and primer-template mismatches. PLOS ONE 2015 May 21;10(5):e0128122.

14. Zhang J at al. PEAR: a fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics 2014 Mar 1;30(5):614-20.

15. Caporaso JG et al. QIIME allows analysis of high-throughput community sequencing data. Nature methods 2010 May 1;7(5):335-6.

16. McDonald D et al. The Biological Observation Matrix (BIOM) format or: how I learned to stop worrying and love the ome-ome. GigaScience 2012 Jul 12;1(1):1.

17. DeSantis TZ et al. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied and environmental microbiology 2006;72(7):5069-5072.

18. Oksanen J et al. The vegan package. Community ecology package, 10 (2007).

- NUSeq Core Facility, Northwestern University

- Molecular Epidemiology Laboratory, NMH

- UIC DNA Sequencing Facility

- George Chlipala, UIC Biostatistics Core

- Juanita T Saqibuddin, NICU Cord Blood Study

- Lucy Minturn, NICU Cord Blood Study

Leena B. Mithal, MD1, Michael Malczynski, BS2, Stefan J. Green PhD3, Chao Qi, PhD4, Ram Yogev, MD1, and Karen Mestan, MD1. 1Pediatrics, Northwestern University Feinberg School of Medicine; 2Clinical Microbiology Laboratory, Northwestern Memorial Hospital;

3Center for Genomic Research, University of Illinois at Chicago; 4Pathology, Northwestern University Feinberg School of Medicine. Chicago, IL, United States.

Deep Sequencing of 16S rRNA Gene Amplicons to Screen Umbilical Cord Blood of Preterm Infants

Leena B. Mithal, MD, MSCIAnn & Robert H. Lurie Children’s Hospital

Infectious Diseases225 E Chicago Ave, Box #20

Chicago, IL, USA 60611lmithal@luriechildrens.org

Ph 312.227.4080 Fax 312.227.9709

Study Design

• Nested case control study within a longitudinal birth cohort

• Clinical data collection of maternal and infant variables: chorioamnionitis, delivery type, culture results, laboratory values, antibiotic treatment, etc.

• Cord blood obtained by sterile venipuncture of umbilical vein, stored at -80°C

• DNA extracted using Qiagen QIAamp DNA Mini Kit from archived RBC/buffy coat

• Identified cEOS cases. Frequency matched PS cases and controls within target gestational age and birth weight ranges.

METHODS RESULTS• Cord blood 16S rRNA gene amplification from genomic DNA coupled with deep

sequencing (NGS) detected the pathogenic organism in all cases of EOS. A complex microflora was often detected, which may indicate that other organisms (not just the primary pathogen) are present when infections occur.

• Pathogens were detected in some PS patients with negative postnatal blood culture. Sanger and NGS aligned in these cases.

• Alpha diversity was significantly higher in infants with infection than controls, and low diversity is a likely indication of the presence of a dominant pathogen.

• Bacterial taxa associated with vaginal, oral and gastrointestinal flora were detected in cord blood of controls. This may represent the presence of low or non-pathogenic and perhaps commensal bacteria.

In summary,

NGS detected bacterial taxa that were not identified through cultivation or with cultivation-independent PCR and Sanger sequencing. Direct Sanger sequencing (without cloning) can be negatively impacted when elevated microbial diversity is present. Deep sequencing can be used delineate the presence and role of rare or novel organisms in EOS, or to better characterize mixed microbial communities.

Limitations of this technology for identification of infection include:

- High sensitivity, potential for contamination

- Unclear significance of low levels of potentially pathogenic organisms

- Inability for species level identification

- Inability to detect viruses and microeukaryotic organisms

According to this preliminary data, cord blood does not appear to be sterile. The significance of bacterial DNA in these cord blood samples is unclear, though further investigation is warranted to study maternal-fetal origins of the neonatal microbiome.

ACKNOWLEDGEMENTS

Culture and sequencing results for cEOS and selected control and PS patients

Funding:

Thrasher Research Fund Early Career Award

Lurie Children’s Pediatric Physician-Scientist Award

NHLBI Grant K23 HL093302

Northwestern Memorial Foundation Friends of Prentice Grants Initiative

Background: Early onset sepsis (EOS) causes morbidity and mortality in preterm infants, yet diagnosis is inadequate. 16S rRNA gene amplicon sequencing of cord blood (CB) is a culture-independent method of pathogen detection that may improve EOS diagnosis. Ureaplasma and Mycoplasma spp. have been isolated from CB with uncertain clinical significance, and PCR is more sensitive. Is cord blood normally sterile? Our objective was to compare Sanger (SS) and next generation (NGS) 16S rRNA sequencing results of CB from preterm infants with confirmed sepsis (cEOS), presumed sepsis (PS) and no sepsis (control).

Methods: This is a nested case-control study within a prospective cohort with archived CB collected by sterile venipuncture at birth. Infants (31±3 wks gestation) were selected using the following criteria: 1) cEOS [+ blood culture <72 hours of life, antibiotics], n=8; 2) PS [– culture, ≥2 abnormal labs, antibiotics], n=8; 3) control [– culture, no antibiotic course, did well], n=8. After DNA extraction, 16S rRNA gene sequencing using SS and NGS was performed. BLAST and QIIME were used for annotation of sequences and microbe identification.

Results: Postnatal cEOS organisms were E. coli and S. agalactiae. SS identified 4 of 8 cEOS species in CB; NGS identified 7 of 8. One E. coli cEOS case had G. vaginalis identified by SS and Gardnerella, Lactobacillus and Ureaplasma by NGS. Three PS had a predominant taxon: Enterobacteriaceae, Sneathia and Actinomyces. Other PS and all controls had a variety of bacteria identified: Enterobacteriaceae, Streptococcus, Lactobacillus, Ureaplasma, Mycoplasma, Prevotella, Bacteroides and Clostridia spp., among others.

Conclusion: CB 16S rRNA gene amplicon sequencing correctly identified pathogens in most cEOS cases. A predominant bacterial taxon was identified in some cEOS and PS, while others had negative SS and multiple organisms by NGS. NGS detected bacteria associated with vaginal, oral and gastrointestinal flora in controls. According to this preliminary data, CB does not appear to be “sterile”. The significance of bacterial DNA in neonatal disease is unclear. Further investigation of CB molecular diagnostics for EOS and possible maternal-fetal origins of the microbiome is warranted. Deep sequencing may be a tool to delineate the presence and role of rare or novel organisms in CB.

ABSTRACT

OBJECTIVETo compare Sanger (SS) and next generation (NGS) 16S rRNA sequencing results of cord blood from preterm infants with confirmed sepsis (cEOS), presumed sepsis (PS) and no sepsis (control).

Assess potential diagnostic value of pathogen identification for EOS Investigate bacterial presence in cord blood of preterm infants

Sanger sequencing

• PCR using V3-V4 region primers

• Gel electrophoresis for positive band

• Sequencing of PCR product

• Analysis of sequence at NUSeq Core Facility

• National Center for Biotechnology Information (NCBI) database BLAST

16S rRNA Gene Amplicon Sequencing

Deep sequencing

• Targeted amplicon sequencing (TAS) approach13, employing V3/V4 or V4 region primers

• Amplicons prepared for Illumina sequencing using Fluidigm AccessArray barcoding system

• Sequencing performed using an Illumina MiSeq

• Custom bioinformatics pipeline employing PEAR14 and QIIME15. Biological observation matrices (BIOMs)16

generated using the Greengenes 13_8 reference database17.

• α- and β- diversity analyses in the software package ‘vegan’ implemented in the “R” software18.

Column plot of bacterial relative abundance12 most abundant microbial taxa across the dataset are shown.

RESULTS

Gestational age: mean ± SD = 30.5 ± 2.7 weeksBirth weight: median (IQR) = 1460 (1108-1770) grams

Alpha diversity analyses

- Shannon Index (SI; genus level) values are significantly different across groups

(Kruskal-Wallis test, p < 0.003)

- Pairwise comparison of SI between cEOS and controls is also significant (Mann-Whitney test, p < 0.002)

Beta diversity analysesA nonmetric multi-dimensional scaling (nMDS) plot was constructed using the Bray-Curtis dissimilarity between microbial community composition of samples at the genus level: 46 samples and controls. Ellipses were calculated for the 95% confidence interval of each group of samples based on a multivariate t-distribution of the data points.

Global Analysis of Similarity (ANOSIM) R=0.367, p=0.001 (999 permutations)

Pairwise ANOSIM valuesBlank cEOS Control PS

Blank 0.489 0.495 0.325

cEOS 0.006 0.641 0.131

Control 0.001 0.001 0.108

PS 0.016 0.076 0.072

R values

p-values (999 permutations)

Blank cEOS Control PS

* NGS relative % abundance of identified bacteria

*