the glue grant: inflammation and the host ... - sbi.org2005; 102:4501-6. healthy subjects trauma...

The Glue Grant: Inflammation and the Host The Glue Grant: Inflammation and the Host Response to InjuryResponse to Injury

Methods in BioengineeringMethods in Bioengineering

Ronald G. Tompkins, M.D., Sc.D.Ronald G. Tompkins, M.D., Sc.D.

John F. Burke Professor of SurgeryHarvard Medical School

Chief of StaffShriners Burn Hospital – Boston

Chief, Burn ServiceMassachusetts General Hospital

Burden of TraumaBurden of Trauma

•• Hundreds of thousands of Americans die each year and over Hundreds of thousands of Americans die each year and over 2.6 million are hospitalized from trauma, sepsis, and burns at a2.6 million are hospitalized from trauma, sepsis, and burns at asocietal cost estimated at over $260 billion societal cost estimated at over $260 billion

•• Morbidity and mortality rates of those surviving initial Morbidity and mortality rates of those surviving initial resuscitation have not improvedresuscitation have not improved

•• Clinical trials in trauma have been industryClinical trials in trauma have been industry--supportedsupported

•• Benefits from traditional research approaches have been limitedBenefits from traditional research approaches have been limited

SIRS and CARS in Burn PatientsSIRS and CARS in Burn Patients

•• Initial resuscitation followed by Initial resuscitation followed by the prothe pro--inflammatory response inflammatory response ((SIRSSIRS))•• Proportional to severity of Proportional to severity of

the injurythe injury•• Can lead to early MODSCan lead to early MODS

•• After a period of relative clinical After a period of relative clinical stability, a compensatory antistability, a compensatory anti--inflammatory response (inflammatory response (CARSCARS))•• Suppressed immunitySuppressed immunity•• Diminished resistance to Diminished resistance to

infectioninfection•• May be discharged May be discharged

uneventfully or develop late uneventfully or develop late MODSMODS

• PORC (R. Maier, D.Herndon)• Genomics (R. Davis)• PACB (L. Moldawer)• CAM (D. Schoenfeld) • DIC (R. Tibshirani, D. Donaho, W. Wong,

J. Storey, W. Xiao,)• Proteomics (R. Smith, C. Miller-Graziano)• IDDC• Administration

Our Core StructureOur Core Structure

PATIENT-ORIENTED RESEARCH CORE (PORC)

SOP ACUTE TRAUMA BURN INFLAMMATION UT HOUSTON MGH UMDNJ - RWJMC U COLORADO UT GALVESTON U WASHINGTON U WASHINGTON WASHINGTON U LOYOLA U PITTSBURGH UT SOUTHWESTERN NW U U ROCHESTER UT SOUTHWESTERN

INFLAMMATION AND THE HOST RESPONSE TO INJURY U54 GM062119

DELIVERABLES IN RED SAMPLES = THIN ARROWS INFORMATION = THICK ARROWS

PROTEIN ANALYSIS AND CELL BIOLOGY CORE SOP

SAMPLE COLLECTION AND COORDINATION SITE

U FLORIDA

CELL SIGNALING AND HUMAN CYTOKINE MURINE CYOKINE CYTOKINE INTRACELLULAR ELISA ELISA PRODUCTION REGULATION U MICHIGAN U FLORIDA U ROCHESTER NW U

LC/MS HIGH THROUGHPUT 2-D GELS PROTEOMICS MGH PNNL

IINNFFOORRMMAATTIIOONN DDIISSSSEEMMIINNAATTIIOONN AANNDD DDAATTAACCOOOORRDDIINNAATTIIOONN CCOORREE

TRAUMA-RELATED DATABASE MMGGHH

COMPUTATIONAL ANALYSIS AND MODELING CORE

BIOSTATISTICAL GENOMIC DATA MINING ANALYSIS ANALYSIS ALGORITHMS MGH U WASHINGTON MIT STANFORD U, MGH

GENOMICS CORE SOP

RNA ISOLATION (WASH U) AND

cDNA PREP (STANFORD U)

TRAUMA PORC MICROARRAY MVC BURN PORC SNP MICROARRAY MICROARRRAY STANFORD U WASHINGTON U U FLORIDA

MODEL VALIDATION CORE (ANIMAL CORE)

SOP ACUTE TRAUMA-HEMORRHAGE BURN INFLAMMATION UAB BWH U MICHIGAN

ADMINISTRATION CORE MGH

CLINICAL DATA VAILIDATION BY PORC

AND CAM CORES

• U54 program at MGH - currently in 5th year

• 22 performance sites

• Genome-wide expression profiling in humans and partial profiling in mice

• Phenotype-genotype link to clinical database of patients with trauma, sepsis and/or burns

Program HighlightsProgram Highlights

Current Accomplishments

Biological Accomplishments

• Demonstrated genome-wide changes in gene exp. after injury (~10,000 genes) in buffy coat

• Limited number of genes (~1,000) predicted a differential outcome towards MODS

• Identified over 3,500 distinct proteins in plasma with ~600 proteins over the first 7 days

• Demonstrated tissue-specific genome-wide expression with contrasts and commonalities to buffy coat

• Identified novel functional modules based on initial analysis of leukocyte subpopulations

Leukocyte RNA Isolation By Buffy Coat Leukocyte RNA Isolation By Buffy Coat and Lysis Techniquesand Lysis Techniques

•• ““Buffy coatBuffy coat”” isolation of RNA from leukocyteisolation of RNA from leukocyte--enriched enriched populationpopulation

•• Centrifuge to obtain interface between RBC and plasmaCentrifuge to obtain interface between RBC and plasma•• Elimination of residual RBC with lysing solution (Buffer EL, Elimination of residual RBC with lysing solution (Buffer EL,

Qiagen, Inc.)Qiagen, Inc.)•• Standard RNA isolationStandard RNA isolation

•• ““Lysis methodLysis method”” involves mixing whole blood first with involves mixing whole blood first with lysis buffer (ACK Buffer) to remove lysis buffer (ACK Buffer) to remove RBCsRBCs

•• Centrifuge to pellet Centrifuge to pellet unlysedunlysed WBCsWBCs..•• Wash Wash pelletedpelleted cells, and standard RNA isolationcells, and standard RNA isolation

S.D. From Mean

<-2.0 -1.0 0 1.0 >2.0

Buffy coat PAXgene™ In common with both methods

unst

imul

ated

SEB

stim

ulat

ed

unst

imul

ated

SEB

stim

ulat

ed

unst

imul

ated

SEB

stim

ulat

ed

•943* probe sets discriminated between SEB stimulated and unstimulated whole blood with Buffy coat

•303* probe sets discriminated between SEB stimulated and unstimulated whole blood with PAXgene™

• 254 probe sets in commonPhysiol Genomics 19:247-54, 2004

200

500

1000

2000

4000

6000

Buffycoat PAXgene™ Buffy coat

unstim SEBPAXgene™unstim SEB

• Blood sample was obtained froma single subject, and total RNA isolated byeither PAXgene™or Buffy coat methods.cRNA was generated from 10 μg of starting material using Affymetrix protocols, and detected using a 2% agarose gel and an Agilent2100 system.

Physiol Genomics 19:247-54, 2004

#G004 GenMAPP Analysis

Physiol Genomics 19:247-54, 2004

Pearson Correlation Coefficient

from cRNA hybridization (n=4)

from RNA starting material (n=4) 0.994 ± 0.002

Leukocyte gene expression from same healthy subject over 24 hrs (n=4 subjects, 4-6 time points, per subject)

0.991 ± 0.003

Leukocyte gene expression from individual healthy subjects (n=17)

0.955 ± 0.017

from individual leukocyte populations in different healthy subjects (n=6)

T cells0.974 ± 0.011

Monocytes0.968 ± 0.010

comparing different cell types from same healthy subjects (n=6)

MO vs T cells0.862 ± 0.016

T cells vs BC0.888 ±0.023

MO vs BC0.929 ±0.013

Leukocyte gene expression from individual trauma patients (n=14)

0.913 ± 0.037

0.997 ± 0.001

Cobb, et al. PNAS 2005; 102:4501-6.

Healthy SubjectsTrauma Patients

B.

A. C.

Healthy SubjectsTrauma Patients

Pearson Correlation Coefficient

Among Individual Healthy Subjects

0.955 ± 0.017

Among Individual Trauma Patients

0.913 ± 0.037

Between Healthy Subjects and Trauma Patients

0.888 ± 0.037

Figure 4

0

500

1000

1500

2000

2500

3000

3500

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1

Coefficient of variation

Num

ber o

f pro

be s

ets

Pearson Correlation Coefficient

Subject #1

Subject #2 0.990 ± 0.002

Subject #3 0.993 ± 0.002

Subject #4 0.993 ± 0.002

0.990 ± 0.006

Coefficient of Variation

Replicate Healthy Subjects

Individual Healthy Subjects

Individual Trauma Patients

Mean ± S.D. 0.093±0.0003 0.182±0.0006 0.207±0.001

Median 0.086 0.162 0.167

90% 0.134 0.275 0.359

10% 0.057 0.103 0.100

A. B.

C.

Cobb, et al. PNAS 2005; 102:4501-6.

Lysis Monocytes T cells

LysisMonocytesT cells

A.

B.

C.Pearson Correlation

CoefficientLysis vs Monocytes 0.929 ± 0.013

Lysis vs T cells 0.888 ± 0.022

Monocytes vs T cells 0.862 ± 0.016

Human LPS Model

Cell

LPS

Gram-negativebacteria

LBP

TLR-4

Cytokineexpression

CR1 Blood Collection Schema

Tompkins, Nature, 2005

0

20

40

60

80

100

0 4 8 12 16 20 24

Hours After LPS

Perc

ent o

f WB

Cs

PMNs

Lymphs

Monos

Tompkins, Nature, 2005

A network of transcriptional factors

A virtual cell of innate immunity

time point 0h

time point 2h

time point 4h

time point 6h

time point 9h

time point 24h

• 0. RELA genes• 1. HLA-D• 2. TUB-A • 3. POLR-II• 4. NDUFS• 5. ATP-V • 6. CCT• 7. PSM• 8. RPS/RPL

Global Innate Immunity Network

Current Accomplishments

Infrastructure Development

• Guidelines for early management of severe trauma and burn patients

• Analytical protocols for cell isolation• Web-based fully relational database• Novel computational and bioinformational tools• Novel approaches for high throughput proteomics• Microfluidics for the isolation of enriched leukocyte

populations

www.gluegrant.org

55μμmm

Erythrocyte Lysis

“Bulk” Lysisτ ~ 15 to 20 min

Microfluidic Lysisτ ~ 1 sec

Lysis buffers: Ammonium Chloride, Deionized Water

Differentials Following Microfluidic Lysis

Microfluidic lysis retains normal subcellular populations

Sethu et al., Analytical Chemistry (in press)

Flow Cytometry

GranulocytesMonocytes

Lymphocytes

GranulocytesMonocytes

Lymphocytes

Ammonium Chloride lysis

Microfluidic lysis

0

1

2

3

4

5

6

7

8Whole Blood

Microfluidic Lysis

“Bulk” Lysis

TotalLeukocytes

Lymphocytes Monocytes Granulocytes

Cel

l Con

cent

ratio

n (x

106

cell/

mL)

Bulk lysis results in significant cell loss

Sethu et al., Analytical Chemistry (in press)

Cell Surface Activation Markers

Sethu et al., Analytical Chemistry (in press)

Website: www.Gluegrant.org

• Public access to educational information

• Consortium member access to • Clinical, analytical, and animal model

protocols• Data• Results• Publications• Experimental methods

• Participating investigator to GLIMS, GMDS, Clinical Databases and protocols under development

Genetics and Molecular MedicineGenetics and Molecular Medicine

The impact of genomics and computing on healthcare will acceleraThe impact of genomics and computing on healthcare will accelerateteand progress over the next 10 years.and progress over the next 10 years.

Phase I > 5 YearsPhase I > 5 Years•• Dramatic expansion in diagnostic tests for existing diseaseDramatic expansion in diagnostic tests for existing disease•• ID of discrete molecular pathologies in major diseases (right RxID of discrete molecular pathologies in major diseases (right Rx: :

right disease)right disease)

Phase II > 5Phase II > 5--10 Years10 Years•• Increasing no. of new Rx derived by rational analysis of moleculIncreasing no. of new Rx derived by rational analysis of molecular ar

basis of diseasebasis of disease•• ID of pharmacogenetic markers for patient responses to Rx ID of pharmacogenetic markers for patient responses to Rx

(pharmacogenetics)(pharmacogenetics)•• New imaging probes for dynamic assessment of body functionsNew imaging probes for dynamic assessment of body functions