What do we need

1
New Technological Developments in Diagnostic
Testing Covering Infectious Diseases, UC Berkeley
/ October 2, 2003

• What do we need?
• Where do we stand currently?
• What are the stumbling blocks?
• Where can/should/will we be in 5 years?

2
New Technological Developments in Diagnostic
Testing Covering Infectious Diseases, UC Berkeley
/ October 2, 2003

• What do we need?
• Where do we stand currently?
• What are the stumbling blocks?
• Where can/should/will we be in 5 years?

3
The New Yorker, January 29, 2001
4

• What do we need?

• Environmental Clinical
• Broad spectrum detection/dx tools
• Rapid, real-time (quantitative),
• automated, (pt-of-care) monitoring
• Standardized sampling methods
• Understanding of natural background
• Agent (variation genetic, antigenic, geo, temp)
• Setting (related agents, non-biological issues)

5
current Dx approaches
system indicators
6
New Technological Developments in Diagnostic
Testing Covering Infectious Diseases, UC Berkeley
/ October 2, 2003

• What do we need?
• Where do we stand currently?
• What are the stumbling blocks?
• Where can/should/will we be in 5 years?

7
Current procedures and technologies

• Detection
• Culture (reference, complete analysis)
• Immunoassays
• Solid-phase, hand-held (SMART), FACS, ELISA
• Nucleic acids
• Amplification PCR, SDA
• Capture magnetic, Ab, electrical
• Detection fluor, ECL, chromo, bDNA
• microarray, mass spectroscopy

(reliance on few antibodies!)
(reliance on type strains!)
8
Syndromes of suspected microbial origin success
in achieving microbiological diagnosis

• pneumonia 50-70
• encephalitis 30
• sepsis 10
• acute diarrhea 20-50

suspected on basis of response to antibiotics,
among other observations
9
Why might traditional approaches for pathogen
identification have failed?
Reliance on cultivation--insensitivity
Microbial phenotypic markers unreliable one can
be mislead when one asks a microbe to perform in
the laboratory! Serology delayed, or
impossible PCR not well deployed, problems with
low clinical sensitivity
10
New Technological Developments in Diagnostic
Testing Covering Infectious Diseases, UC Berkeley
/ October 2, 2003

• What do we need?
• Where do we stand currently?
• What are the stumbling blocks?
• Where can/should/will we be in 5 years?

11
Major challenges, obstacles

• Diversity of potential agents
• (including bioengineered, chimeric organisms)
• Variability, varying evenness of agents,
• in nature
• Defining relationship between detected agent
• and disease risk
• Complex biological background!
• Sampling, processing procedures
• (standardization, calibration in real world)

12
Bacteria
The Tree of Life
(based on rRNA sequences)
Archaea
Eukarya
Pace NR. A molecular view of microbial diversity
and the biosphere. Science 1997 276734
13
Bacteria
The Tree of Life
(based on rRNA sequences)
human-associated
Archaea
Eukarya
Pace NR. A molecular view of microbial diversity
and the biosphere. Science 1997 276734
14
Bacteria
15
Emergence of infectious diseases

• Acquisition of toxins,
• adhesins
• Antigenic variation,
• e.g. new capsule
• Broadened host range
• Improved growth
• or transmissibility
• Acquisition of drug R

• Societal events poverty, crowding, conflict,
  migration
• Globalization of food supply
• Environmental changes
• Human behavior sexual, recreational, diet,
  travel
• Impaired host defenses, antibiotic use
• Public health infrastructure

16
What does it mean for an infectious disease to
emerge?

• Evolution of new agent to cause disease
• (evolution of virulence)
• Previously-recognized agent causes disease
• with new features (clinical, epidemiological,
  geographical, histological)
• Pre-existing, but previously-unrecognized
• agent makes itself known ( new disease features)

17
Pathogen discovery basics
pathology
18
Pathogen discovery basics
pathology
microbe
host
19
Pathogen discovery seeking molecular signatures
Pathogen as source of signature

• broad range PCR
• microbial/viral survey phyloarray
• subtractive/comparative methods
• representational difference analysis
• differential display
• expression or phage display library screening
• (using host antisera or T-cells)
• small molecule or protein detection (MS)

Host as source of signature

• host genome-wide transcript profiling
  (microarray, other)
• host protein profiling (microarray, MS)

20
Pathogen discovery seeking molecular signatures
Pathogen as source of signature

• broad range PCR
• microbial/viral survey phyloarray
• subtractive/comparative methods
• representational difference analysis
• differential display
• expression or phage display library screening
• (using host antisera or T-cells)
• small molecule or protein detection (MS)

Host as source of signature

• host genome-wide transcript profiling
  (microarray, other)
• host protein profiling (microarray, MS)

21
digest specimen, purify / concentrate DNA
22
Rhinosporidiosis

• slow-growing tumors of nasal mucosa, ocular conj.
• southern India, Sri Lanka prevalence 1-2
  children

23
Rhinosporidium seeberi a fungus?
endospore
sporangium
stain PAS reagent
24
(No Transcript)
25
Artemia (Brine Shrimp)
Xenopus (Frog)
Mytilus (Blue Mussel)
Tripedalia (Jellyfish)
Animals
Microciona (Sponge)
Diaphanoeca (Choanoflagellate)
Rosette Agent
95
Rhinosporidium seeberi
Dermocystidium sp.
DRIPs
Dermocystidium salmonis
Psorospermium haeckelii
Ichthyophonus hoferi
Aspergillus
Fungi
Chytridium (Chytrid)
Mucor (Bread Mold)
Protists
Acanthamoeba (Amoeba)
Zamia (Plant)
Chlorophytes
Rhodophytes
Porphyra (Red Algae)
Lagenidium (Oomycete)
Heterokonts
Labyrinthuloides (Slime Net)
Perkinsus (Protozoan Oyster Parasite)
Apicomplexa
Sarcocystis (Coccidian Protozoan)
.10
26
Other Animals
Plants
Protists
Jellyfish
DRIPs deepest branch of animals aquatic
parasites hostsfish Rs in humans water
exposure
Fungi
Sponges
DRIPs
Choanoflagellates
0.1
27
Bioterrorism future considerations

• Mining nature (unconventional agents)
• Improving upon nature
• Engineered pathogens cloning of
• known virulence factors/islands,
• deletion of inhibitory factors, host
• modifying factors (eg, cytokines),
• shuffled evolved vir factors
• Novel agents (pathogenic proteins,
• bioregulators, chimeric agents)

28
Major challenges, obstacles

• Diversity of potential agents
• (including bioengineered, chimeric organisms)
• Variability, varying evenness of agents,
• in nature
• Defining relationship between detected agent
• and disease risk
• Complex biological background!
• Sampling, processing procedures
• (standardization, calibration in real world)

29
4.3-6.2 healthy humans positive without history
of anthrax or anthrax exposure
30
Bacteria
31
Uncultivated TM7 in the human mouth
32
We know more about the tropical rain forest than
we do about the human endogenous microbial flora!
33
Major challenges, obstacles

• Diversity of potential agents
• (including bioengineered, chimeric organisms)
• Variability, varying evenness of agents,
• in nature
• Defining relationship between detected agent
• and disease risk
• Complex biological background!
• Sampling, processing procedures
• (standardization, calibration in real world)

34
The New Yorker, November 19, 2001
35
The realities of anthrax detection, 2001...

• Lack of standardized collection methods
• Low specimen analysis through-put
• Inadequate laboratory surge capacity
• Slow turn-around, late in disease course
• Inadequate delivery and implementation of
• state-of-the-art technologies in the field
• Data interpretation negatives and positives

36
New Technological Developments in Diagnostic
Testing Covering Infectious Diseases, UC Berkeley
/ October 2, 2003

• What do we need?
• Where do we stand currently?
• What are the stumbling blocks?
• Where can/should/will we be in 5 years?

37
Some near-term goals

• Library of high-affinity binding reagents
• More sensitive binding detection
• Extensive database of sequences,
• other signatures
• High-throughput labs, with surge capacity
• Standardized, automated specimen collection
• and processing procedures, technologies
• Further development of DNA microarray
• approaches

38
cold
RV-infected

• 1600 unique 70-mers
• 140 viral genomes
• Sensitivity100 viral particles

39
Phyloarray v2 10,000 rDNA oligo probes
Agilent Theo Sana, Paul Wolbert Mike Eisen
(LBL) Pat Brown (Stanford)
40
Unexplained Deaths Project(CDC EIP)
total pop 7.78 million
Acute, life-threatening illness in ages
1-49, previously healthy all routine diagnostic
tests (-) enhanced passive surveillance Seek
patterns, clusters, clues look for infectious
agents using molecular (research) tools
Nikkari et al, Emerg Infect Dis 8188-194, 2002
Hajjeh et al, Emerg Infect Dis 8145-153, 2002
41
Unexplained Deaths Project(CDC EIP)
137 cases fit definition (5/95-12/98)
0.5/100,000 syndromes neuro (29),
respiratory (27), cardiac (21), multisystem
(13)
42
Microbe or host relative advantages
Microbe (as target) agent specificity
(adjustable) signatures more easily defined
Host (as target) agent need not be present
early diagnosis? outcome predictions
43
Can one recognize and classify clinical (and
pre-clinical) states of infection by examining
host gene response patterns?

• Potential advantages
• broad range
• early
• microbe not required in specimen
• prognostic value ( targets)

44
37,632 spots/elements 32,494 cDNAs 10,250 named
genes 18,000 unique genes
Human cDNA microarray
Two-color, comparative hybridization format
Alizadeh A et al., Cold Spring Harb Symp Quant
Biol 6471, 1999 Nature 403503, 2000
45
Comprehensive gene expression profiles integrate
host genotype and environmental input
gene
transcriptional control
46
cDNA microarray procedures-1
(Relman Brown)
47
cDNA microarray procedures-2
Cy5 Cy3
Cy3
Cy5
Image analysis Data filtering Normalization
R/G ratio represents relative abundance of
transcripts

• Pattern recognition
• unsupervised (class discovery)-
• clustering, SOM, SVD (PCA), ICA
• supervised (class prediction)-
• SAM, support vector machines, t/f-test (DLDA,
  ANOVA), modeling (waveform, periodicity)

Genes
Experiments
48
2-Way Hierarchical Clustering Methodology
genes
cluster by genes
cluster by microbial stimuli
microbial stimuli
microbial stimuli
microbial stimuli
49
(No Transcript)
50
(No Transcript)
51
Issues involving a complex background gene
expression patterns from blood in healthy
individuals

• Variability--How noisy is the background?
• Must each individual serve as his or her own
  control? Do patterns provide insight into
  physiology and intrinsic-ness?
• How well do blood cells report on local
• processes? Other sources...?

52
48 PBMC samples from 19 individuals
clustered on the basis of genes with intrinsic
scores gt2 SD from mean (340) (mean square
pairwise difference between/ mean square pairwise
difference within individuals)
Whitney A et al, PNAS 2003 1001896-1901
53
Donor-intrinsic gene expression
males
48 PBMC samples from 19 individuals clustered on
the basis of genes with highest intrinsic scores
(340)
females
Whitney A et al, PNAS 2003 1001896-1901
54
Looking further into the future

• Biosensors remote (e.g. cells),
• endogenous (e.g. flora)
• Hyperspectral imaging, analysis

55
Conclusions

• Challenges associated with detection
• and diagnosis are significant, but worthy of
  major investment
• Current status platforms are quite
• promising real-world issues still need further
  attention
• Major pay-off may be achieved if we
• embrace larger aspects of biology and disease,
  and anticipate future threats

56
Acknowledgements
Stanford Pat Brown Trevor Hastie Ash
Alizadeh Jennifer Boldrick Mary Brinig Paul
Lepp Cleber Ouverney Stephen Popper Kate
Rubins Addie Whitney
CDC Jim LeDuc Marc Fisher Peter Dull
Duke Barth Reller Chris Woods David Murdoch
57
(No Transcript)