Introduction Methods Barrett’s esophagus is a condition in which the tissue lining the esophagus is replaced by tissue that is similar to the lining the intestine. In many cases, esophageal adenocarcinoma is not identified until later stages when treatments are not always effective. Although the causes of Barrett’s esophagus are not known, a condition known as GERD, acid reflux, is a risk factor. Estimates show that approximately 10 to 20% of Americans experience GERD on a weekly basis. Furthermore, due to its well characterized normal-to-cancer progression path, the Barrett’s esophagus can be used as a model system for cancer induction and development studies. By studying the respiration rates and other phenotypes of Barrett’s esophageal cells from different pre-cancerous and cancerous developmental stages at the single cell level, our objective at the Center for Biosignatures Discovery Automation is to identify and characterize early aberrant transformations and carcinogenesis.

• We have successfully measured oxygen consumption rates of human esohageal epithelial cells•Our single cell gene expression analysis technique could detect up to 10 genes in one cell with high reproducibility and specificity. •Using this technique, related gene transcription responses to hypoxia were analyzed at both bulk and single cell-levels. •The measurements of gene expression levels in single cells showed significant cell-to-cell variability in response to hypoxia, compared with results from bulk cells. • Ongoing efforts in system development is devoted to further optimization towards higher throughput and reliability with minimal user interaction.

This poster summarizes the invaluable contributions of all colleagues from MLSC at Arizona State University and the collaborators from the University of Washington, the Fred Hutchinson Cancer Research Institute, and the Brandeis University. The authors would like to thank all of these persons for their help and fruitful collaboration. This work is supported by NIH NHGRI 5P50HG002360-11 and P50HG002360-10S1

Measuring phenotypes of Barrett’s esophageal cancer cells at the single cell level

Motivation

Conclusions and Future Directions

Our Approachg

gPartially Adhered

Cell

Capillary tip(B)(A)

Medium

Adhered Cell

Trypsin

Glass Glass

PCR Cap

Detached Cell

(C)

Results

1. SortingFACS to obtain cell cycle-sorted cells

2. Cell collectionPick up single cells using micropipette and place into cap

3. RNA isolation

4. Reverse Transcription

5. qPCR 6. Quality assessmentAgarose Gel

Electrophoresis and Sequencing

7. Data analysis

Immediately precede to

step 4

G1

Acknowledgements

A. Mohammadreza1, J. Zeng1, J. Wang1, W. Gao1, S. Merza1, Y. Anis1, L. Kelbauskas1, W. Zhang1, C. Youngbull1, S. P. Ashili1, H. Zhu1, J. Houkal1 ,Y. Tian1, D. Smith1, M. Hupp1, P. Senechal-Willis1, T. Paulson2, L. Burgess3, B. Reid2, L. Wangh5, M. Holl1, R. Johnson1 and D. Meldrum1

1Center for Biosignatures Discovery Automation, Biodesign Institute, Arizona State University, Tempe, AZ, 2Fred Hutchinson Cancer Research Center, Seattle, WA, 3Microscale Life Sciences Center, University of Washington, Seattle, WA, 4Department of Laboratory Medicine, University of Washington, Seattle, WA , 5Department of Biology, Brandeis University, Waltham, MA

• Current cellular research is predominantly based on bulk analyses.

•The results are expressed as population averages thereby masking the significance of intrinsic cellular heterogeneity.

•Many diseases, including cancer, originate in few aberrant progenitor cells.

Figure 1. Automated approaches for single-cell loading, manipulation and multi-parameter analysis.

Figure 2.* Variation in transcript distribution shows the importance of the single cell approach.

• Measuring the concentrations of various metabolites by means of extracellular optical sensors in a hermetically sealed microchamber containing the cell.

• We produce the hermetic seal by placing a lipped lid containing the sensor on top of the microwell with the cell.

a) A cell is aspirated into the micropipette tip. b) The tip is lifted and the reservoir with the microwell substrate at the bottom is moved to align with the objective. c) The cell is dispensed from the micropipette tip into a microwell.

Step 1: Cell Loading

Step 2: Drawdown Experimental Setup

Lid

Lip

Cell Medium

Figure 3. Outliers may be responsible for adverse health affects.

Brightfield migraph of microwells with single

cells

Epi-fluorescence micrograph showing oxygen sensor

inside the lids

a) The hermetically sealed microchamber containing the cell is produced by placing the lid on top of the well and exerting a force of about 40-70 Newtons on the lid through the piston.

b) A compliant layer is placed between the lid and the piston to ensure equal force distribution across the lid.

Single cell and micropipette tip

Microwell array of 9 single cells

in wells

Single cells after 16-24 hours of

incubation

Step 3: Harvesting and PCR Analysis

Part A)a) The cells are first treated with low

concentration trypsin solution to induce partial detachment from the substrate.

b) The cells are aspirated into a microcapillary.c) The cells are transferred into a PCR cap.

Part B)As the final step of our analysis, we conduct transcription level profiling experiments at the single-cell level.

Single-cell

loading

Metabolism

phenotype measurem

ent

Single-cell

harvesting

Single-cell whole

transcriptome

analysis

Single-cell qRT-PCR

Develop single-cell qRT-PCR method

Develop single-cell

whole transcriptomeamplification

method

Multiple cells in a single well

(A combination of different cell

types or from the same cell line)

Data analysis

Data analysis

Gene expression changed

significantly

Characterize transcriptiona

l profiles

•The two graphs below show oxygen consumption time courses of single cells of human esophageal epithelial cell lines, CP-A is a metaplastic and CP-C is a dysplastic cell line.

• This graph shows the distribution of oxygen consumption rates in metaplastic CP-A and dysplastic CP-C esohageal epithelial cells.

PTGES5 Angptl4 MT3

GAPDH AMPKA1 VEGF

• Below is a graph of measured transcription levels of 6 genes in G1 sorted CP-A cells in response to hypoxic conditions.

Sensor