Quantitative Assessment of Autocrine Cell Loops

Abstract

Regeneration of functioning tissue essentially involves recapitulating relevant aspects of organogenesis, so that the starting composite of cells, matrix, and molecular factors develops into the desired structure and physiology. A crucial aspect of development is local cell—cell communication; that is, molecular regulatory factors are more typically paracrine and autocrine than endocrine in nature. Autocrine loops were originally thought of predominantly as being involved in pathological behavior, but it is becoming increasingly clear that a large portion of normal physiological behavior—and a tremendous portion of development—is strongly regulated by autocrine factors (1). Thus, continuing progress of the field of tissue engineering will require increased understanding of how autocrine loops operate, so that they can be designed or manipulated systematically. We have made an effort in this direction, and some early experimental and modeling results can be found in the literature (2–5). In this chapter, we describe the methods we have used for creating autocrine cell loops and quantitatively assessing their operation.

Keywords

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

1 Introduction

Regeneration of functioning tissue essentially involves recapitulating relevant aspects of organogenesis, so that the starting composite of cells, matrix, and molecular factors develops into the desired structure and physiology. A crucial aspect of development is local cell—cell communication; that is, molecular regulatory factors are more typically paracrine and autocrine than endocrine in nature. Autocrine loops were originally thought of predominantly as being involved in pathological behavior, but it is becoming increasingly clear that a large portion of normal physiological behavior—and a tremendous portion of development—is strongly regulated by autocrine factors (1). Thus, continuing progress of the field of tissue engineering will require increased understanding of how autocrine loops operate, so that they can be designed or manipulated systematically. We have made an effort in this direction, and some early experimental and modeling results can be found in the literature (2–5). In this chapter, we describe the methods we have used for creating autocrine cell loops and quantitatively assessing their operation.

A crucial assessment issue is how much of the ligand synthesized by autocrine cells is captured by the secreting cells themselves, and how much is permitted to escape into the bulk extracellular medium. A perhaps counterintuitive principle of autocrine loops is that they might be providing the most useful physiological function when one cannot experimentally find any autocrine ligand in the bulk medium, because that is when the system should be most sensitively regulated by what is captured; when one can experimentally find substantial levels of autocrine ligand in bulk medium, that may be when pathological behavior arises. Another important principle of autocrine loops is that proper regulation can only occur when the ligand is synthesized in its transmembrane precursor form for proteolytic cleavage at the cell surface. When a purely secreted form of the ligand is synthesized, receptor binding can take place along the secretory pathway even before the ligand reaches the cell surface; this short-circuiting probably compromises aspects of receptor trafficking dynamics and interactions of ligand with extracellular matrix that affect cell regulation by autocrine loops.

Our assessment methods are ELISA, for determination of the amount of autocrine ligand escaping into bulk extracellular medium, and the Molecular Devices Cytosensor, for determination of the amount of autocrine ligand captured by secreting cells. The Cytosensor measures extracellular acidification rate (ECAR) in real time, which can be correlated to receptor-mediated signaling processes (6,7). Our model experimental system is of B82 mouse L-cells, transfected with the genes for human epidermal growth factor receptor (EGFR) and transforming growth factor alpha (TGF-α) using the two-plasmid tetracycline-controlled gene-expression system. This system allowed us to calibrate readouts from the Cytosensor with ligand binding, because we could quantify Cytosensor readouts from nonautocrine cells challenged with exogenous ligand, and to vary ligand synthesis rate by changing the concentration of tetracycline in the medium.

2 Materials

2.1 Creation of Autocrine Cell Loops

2.1.1 Reagents for Construction of pUHD 10.3-TGF-α Plasmid

• 1. pUHD15.1neo and pUHD 10.3 (first and second plasmids of tetracycline controlled, inducible expression system) obtained from M. Gossen and H. Bujard (8) (Heidelberg University, Germany).

• 2. pMTE4 (transmembrane TGF-α cleavable protein) obtained from R. Derynck (9) (University of California at San Francisco).

• 3. Bluescript II KS+ (Strategene, La Jolla, CA).

• 4. Restriction enzymes: HindIII, NcoI, RsaI, XhoI, EcoRI, BamHI, PstI (Boehringer Mannheim, Indianapolis, IN).

• 5. Modifying enzymes: Klenow, T4 ligase (Boehringer Mannheim).

• 6. Low melting point agarose (Sigma, St. Louis, MO) and β-agarasel (New England Biolabs, Beverly, MA)

• 7. Wizard Maxiprep (Promega, Madison, WI)

2.1.2 Reagents for Creation of Autocrine Clones

• 1. Histidinol resistence plasmid (pREP8) (Invitrogen, San Diego, CA). pREP is modified into pR8 with the removal of the EBNA-1 and OriP segments (SacI to XbaI deletion and religation) to prevent episomal replication, and to allow for plasmid incorporation into chromosomal DNA.

• 2. CaPO₄ transfection reagents:

  • a. Salt solution: dissolve 3.7 g KCl (Sigma), 10 g D(+) glucose (Sigma), and 1 g Na₂HPO₄ (Sigma) into 50 mL final volume ddH₂O, filter sterilize with 0.2-µm filter, and store at −20°C.

  • b. 2X HBS (HEPES buffered saline): Add 1 mL salt solution, 1 g HEPES (Sigma), and 1 g NaCl (Sigma) to 85 mL of ddH₂O, pH to 7.1 with 1 N NaOH, and bring volume up to 100 mL with ddH₂O. Filter sterilize with 0.2-µm filter; store in 20-mL aliquots at 4°C.

  • c. 2.5 M CaCl₂: dissolve 36.8 g of CaCl₂·2H₂O (Sigma) into 100 mL final volume H₂O. Autoclave sterile. Store at room temperature or 4°C.

• 3. 8×8 glass cylinders (Bellco, Vineland, N.J.)

2.2 Determination of Autocrine Ligand in Bulk Extracellular Medium

2.2.1 Reagents for Measurement of Autocrine Ligand

• 1. Monoclonal anti-EGFR antibodies 528 and 225 (hybridomas; ATCC, Rockville, MD). Antibodies purified from hybridoma supernatant, 50% (NH₄)₂SO₄ (Sigma)-ddH₂O cut at 4°C, spin at 7600g for 10 min at 4°C. Dialysis pellet overnight in PBS with 30,000 MWCO dialysis tubing (Spectrum, Houston, TX).

• 2. TGF-α ELISA kit (Calbiochem, San Diego, CA).

• 3. 1X PBS buffer; 0.2 g KCl (Sigma), 0.2 g KH₂PO₄ (Sigma), 8 g NaCl (Sigma), 1.15 g Na₂HPO₄ (Sigma) dissolved in 1 L ddH₂O, pH 7.0, sterile-filter through 0.2-µm filter and store at 4°C.

2.2.2 TGF-α Cellular Processing

• 1. Protein A-conjugated Sepharose beads (Sigma).

• 2. Amicon Concentrators (YM3000 membrane)/Centricons (3000 MWCO) (Amicon, Beverly, MA).

• 3. Protein standards: albumin, carbonic anhydrase, cytochrome, aprotinin (Sigma).

• 4. Glycerol (Sigma).

• 5. G-50 Sephadex (Pharmacia, Piscataway, NJ).

2.3 Determination of Autocrine Ligand Captured by Secreting Cells

2.3.1 Molecular Devices Cytosensor Material

• 1. Transwells: 12 mm diameter, 3 µm pore size (Corning, Cambridge, MA).

• 2. Transwell inserts: 8 mm diameter, 3 µm pore size (Molecular Devices, Menlo Park, CA).

• 3. Spacer: 50 µm height/6 mm id (Molecular Devices).

• 4. DV-cyto buffer: On the day of a Cytosensor experiment, make 1 L of Dulbecco’s volt modified Eagle’s media (DMEM, 4500 mg/L glucose, l-glutamine, Sigma) without sodium bicarbonate. Instead, add 2.59 g of NaCl (Sigma), pH to 7.4, and add 100 mg of BSA (Sigma). Filter-sterilize with 0.2-µm filter. Sometimes, DV-cyto can be stored at 4°C, but re-pH to 7.4 and filter.

2.3.2 Calibration of Cytosensor Readout Reagents

• 1. hEGF (Gibco BRL): EGF at 1 mg/mL in 50 mM sodium phosphate, stored at −20°C.

• 2. 1X WHIPS buffer: 1 g polyvinyl pyrrolidone mol wt 40,000 (Sigma), 7.6 g NaCl (Sigma), 0.373 g KCl (Sigma), 0.102 g MgCl₂·6H₂O (Sigma), 0.147 g CaCl₂·2H₂O (Sigma), 4.76 g HEPES-acid (Sigma), dissolve in 1 L of ddH₂O at a pH of 7.4, store at 4°C.

• 3. I¹²⁵ EGF: Add together 1 iodobead (Pierce, Rockport, IL), 100 mM 80 µL Tris (Sigma), 30 µL 100 mCi/mL I¹²⁵ (NEN, Boston, MA), 10 µL of 10 ng/µL EGF (Gibco-BRL). Wait 15 min and remove protein to new tube. Made fresh day of use, add 40 µL of 12 mg/mL sodium metabisulfite (Sigma) in PBS to iodobead tube, remove, and add to EGF tube. Add 40 µL of BSA chase solution (20 mL ddH₂O, 100 mg BSA-RIA grade (Sigma), 1 crystal KI (Sigma), sterile-filtered with 0.2-µm filter, and store 1-mL aliquots at −20°C). Separate I¹²⁵-EGF from I¹²⁵ over a G-10 Sepharose column (Pharmacia, Piscataway, NJ) with PBS. Store I¹²⁵-EGF at 4°C.

• 4. D/H/B buffer: make 1 L of Dulbecco-V-modified Eagle media (DMEM, 4500 mg/L glucose, 1 mM l-glutamine, Sigma) without sodium bicarbonate. Add 5.95 g/L HEPES (Sigma) and 1 mg/mL BSA (Sigma), pH to 7.2.

2.3.3 Cytosensor Readout of Antibody Inhibition Reagents

• 1. Polyclonal anti-TGF-α antibody (R & D Systems, Minneapolis, MN).

• 2. Monoclonal anti-EGFR antibodies 528 and 225 (hybridomas-ATCC) (see Note 2 ).

• 3. Antibodies used in Cytosensor experiments.

  • a. Because of the sensitivity of Cytosensor measurements, it has been noted that noise and baseline changes occur when adding a high concentration of antibodies in glycerol or a buffer with high salt concentrations, such as PBS.

  • b. To avoid this problem, antibodies or any other additives used at a high concentration need to be dialyzed in DV-s (DV-cyto buffer without BSA added) overnight at 4°C. Aliquot protein into small volumes to prevent cyclic freeze-thaw problems. Suggestion for antibodies and small volumes is Slide-a-Lyzer <0.5 mL capacity, 10K MWCO (Pierce, Rockford, IL).

2.4 B82 Fibroblast Tissue Culture Medium

B82 cells are a mouse L-cell line that lacks the EGFR chromosomal segment and does not express detectable levels of epidermal growth factor (EGF) and TGF-α. B82 receptor minus and receptor positive cells were obtained from G. Gill (University of California at San Diego) (10). All B82 cells are grown at 37°C, 5% CO₂ and 98% relative humidity.

• 1. R⁻ media (Parental cells (B82R⁻) tissue-culture medium):

  • a. DMEM: 4500 mg/l-glucose, 1 mM l-glutamine, Sigma); add 3.7 g/L sodium bicarbonate (Sigma), filter-sterilize with a 0.2-µm filter (Gelman Sciences, Ann Arbor, MI) under 12 psig positive pressure (N₂ gas), and store at 4°C until use.

  • b. l-glutamine 1 mM (Sigma). 100 mM glutamine stock solution; dissolve 3.65 g into 250 mL 1X PBS buffer, filter sterilize through 0.2-µm filter, and store at −20°C in 10-mL aliquots.

  • c. Penicillin—streptomycin 100 IU/mL-100 µg/mL (Sigma).

  • d. Bovine calf serum 10% (Sigma).

• 2. R⁺ media (B82 with pXER (EGFR, B82R⁺) tissue culture media).

  • a. DMEM (Sigma).

  • b. l-glutamine 1 mM (Sigma).

  • c. Penicillin—streptomycin 100 IU/mL-100 µg/mL (Sigma).

  • d. 10,000 MWCO dialyzed bovine calf serum 10% (Sigma).

  • e. 1 µM methotrexate (Sigma). 1 mM 1000X stock solution; dissolve 25 mg methotrexate into 55 mL of 0.1 N HCl, sterile-filter with 0.2-µm filter, and store at −20°C in 10-mL aliquots.

• 3. R⁺/1st plasmid media (B82R⁺ cells with pUHD15. 1neo (B82R⁺/1st plasmid) tissue-culture media).

  • a. R⁺ media plus 600 µg active G418/mL (Sigma). 60 mg/mL-100X geneticin sulfate (G418) stock solution; 5 g (70% active ingredient) dissolved in DMEM, final volume 58.3 mL, sterile-filtered using a 0.2-µm filter, store at −20°C in 10-mL aliquots.

• 4. Autocrine media (B82 EGFR cells with pUHD15.1neo, pUHD10.3/TGF-α, pR8 (B82R⁺/TGF-α) tissue-culture media).

  • a. R⁺/1st plasmid media plus 2.4 mM histidinol (Sigma). 2.4 M histidinol stock solution; 5 g dissolved in 9.7 mL ddH₂O final volume, sterile-filtered using a 0.2-µm filter, store at 4°C.

  • b. 1 µg/mL tetracycline (repression of TGF-α expression, Sigma). 20 mg/mL tetracycline superstock, dissolve tetracycline powder into ddH₂O, and store at −20°C. Dilute tetracycline 1∶40 in ddH₂O, sterile-filter using 0.2-µm filter, and store at −20°C in 1-mL aliquots.

3 Methods

3.1 Creation of Autocrine Cell Loops

3.1.1 Construction of pUHD 10.3/TGF-α

First, Bluescript II K.S+ (pBS)/TGF-α plasmid is constructed for ease of molecular cloning because of Bluescript’s multiple cloning region.

• 1. The entire 800 bp TGF-α sequence is removed from pMTE4 with a HindIII digest (37°C, 2 h) and gel purified (1% low melting point agarose, 2.5 hours, 50 mAmps). Isolate the TGF-α gene fragment from the agarose using β-agarasel and ligate into HindIII digested Bluescript II KS⁺ plasmid (T4 ligase, overnight, 16°C). Minipreps can be checked via NcoI (only cuts if insert is present) digests and large scale preps of pBS/TGF-α purified with Wizard Maxiprep. Orientation of TGF-α insert (need T7 to T3 orientation for part 2) in Bluescript can be determined by RsaI digests (if T7 to T3, bands are 577, 1754, 1430 bps; if T3 to T7, bands are 348, 1754, 1659 bps) and sequencing.

• 2. The construction of pUHD10.3/TGF-α. Remove TGF-α sequence from pBS/TGF-α by digesting the plasmid with XhoI and EcoRI (37°C, 2 h each). Before digesting the plasmid with EcoRI, however, the linear plasmid (cut with XhoI) must be blunted with Klenow enzyme. Prepare pUHD10.3 by digesting it with BamHI (37°C, 2 h), blunting with Klenow enzyme (same method as with pBS/TGF-α), and digesting the linear plasmid with EcoRI (37°C, 2 h). Gel purify the DNA fragments with 1% low-melting-point agarose, 2.5 h, 50 mAmps, and isolate the fragments from the agarose using β-agarasel. Ligate the two DNA fragments together (T4 ligase, overnight, 16°C); check resulting plasmid preps for the correct pUHD10.3/TGF-α via PstI linearization (PstI only in TGF-α insert, no sites in negative pUHD10.3 plasmids) and sequencing.

3.1.2 Creation and Selection of Autocrine Cells

The transfection of the B82 cells were accomplished using the CaPO₄/DNA precipitation method (11).

• 1. Day 1: Seed one million B82 R⁺/1st plasmid cells into 60-mm tissue-culture dishes in R⁺/1st plasmid media.

• 2. Day 2: Prepare a 500-mL aliquot of pUHD10.3/TGF-α and pR8-Ca₃(PO₄)₂ precipitate by mixing 30 µg of pUHD 10.3/TGF-α and 10 µg of pR8, up to 450 µL ddH₂O (depending on plasmid concentrations), 500 µL of 2X HBS, and 50 µL of 2.5 M CaCl₂. Add the precipitate to the cells’ media plated on d 1. On d 3, rinse the cells with PBS and add back fresh R⁺/1st plasmid media.

• 3. Day 4: Dilute cells 1∶4 into autocrine media, keeping all of the cells. Note: Sometimes, if a large number of clones continue to grow in the selective media, and isolating individual clones is difficult, dilute the cells serially 1∶2, 1∶10, 1∶100, 1∶1000, 1∶10,000. Refresh the selective media every third d until individual cell colonies appear (approx 200–1000 cells per colony), typically appearing 1–2 wk after beginning selection. Isolate the colonies using autoclaved Bellco 8×8 mm glass cloning cylinders dabbed into autoclaved Vaseline, and pass the colony into individual well of a 24-well plate containing selective media.

3.2 Determination of Autocrine Ligand in Bulk Extracellular Medium

3.2.1 ELISA Measurement of Autocrine Ligand

• 1. Plate autocrine cells in four sets (one for induced TGF-α expression and one for uninduced expression, plus and minus antireceptor blocking antibodies) at a cell concentration allowing for 2–3 d of growth before cell confluence is reached in normal autocrine growth media. For example, B82 cells double once a day, thus plate 125,000 cells into a 35-mm tissue culture dish and confluence is 3 d later at 1.5 million cells.

• 2. Two days before cell confluence, remove tetracycline containing autocrine media, rinse the cells with PBS, and, to half of the cell wells, add tetracycline-free autocrine media, inducing TGF-α expression. To the other cell wells, add tetracycline containing autocrine media, keeping TGF-α expression off. NOTE: Protein synthesis takes about 6 h before detection at the surface and in the media, thus TGF-α expression is induced 24 h before beginning the experiment to ensure a steady-state expression of TGF-α.

• 3. Next day, with TGF-α expression at steady state, rinse the cells with PBS and add fresh autocrine media (tetracycline-free for induced cells, and tetracycline-containing media for uninduced cells). To half of the induced and uninduced autocrine cells’ media, add 10 µg/mL anti-EGFR-blocking antibody 225 or 528. A comparison of TGF-α concentrations, plus and minus antibody, allows the analysis of TGF-α uptake by its receptor.

• 4. On the last day, remove 1 mL of conditioned media, spin at 13,000g for 10 min at 4°C to remove cell debris. Store the supernatant at −20°C, if measuring TGF-α concentration another day, otherwise on ice. Measure TGF-α concentration via TGF-α ELISA. Determine cell density by trypsinizing the cells, and count cell number using a hemocytometer or Coulter counter (see Note 1 ).

3.2.2 Tetracycline Concentration Effect on TGF-α Secretion

• 1. Plate autocrine cells at a similar cell density at a cell concentration that allows 2–3 d before cell confluence is reached in normal autocrine growth media. Plate enough cell dishes for 6 or more different tetracycline concentrations, with replicates.

• 2. Two days before cell confluence is reached, rinse the cells with PBS and replace the media with fresh autocrine media containing a gradient of tetracycline concentrations. For example, TGF-α expression is mostly inhibited by 1 µg/mL tetracycline, thus a 10, 1, 0.1 µg/mL to 10, 1, 0.1, 0 ng/mL tetracycline concentrations is an appropriate range.

• 3. Next day, with TGF-α expression at steady state, rinse cells with PBS and replace the media with a fresh autocrine media containing the tetracycline gradient. Also, include in the media, 10 µg/mL antireceptor blocking antibody 225 or 528, to prevent receptor TGF-α uptake and accurate measurement of TGF-α expression.

• 4. On the last day, remove 1 mL of conditioned media, spin at 13,00g for 10 min at 4°C to remove cell debris. Store the supernatant at −20°C, if measuring TGF-α concentration another day, otherwise on ice. Measure TGF-α concentration via TGF-α-ELISA. Determine cell density by trypsinizing the cells and count cell number using a hemocytometer or Coulter Counter.

3.2.3 Determining Cellular Processing of Secreted TGF-α Protein

• 1. Plate a large number of either transfected B82R⁻/TGF-α cells or with autocrine B82R⁺/TGF-α cells into 100-mm dishes, T75 flasks, or roller bottles in tetracycline-free growth media. Once cells have grown confluent and expressing TGF-α, replace the media with serum/protein-free media. Allow the media to condition for 1–2 d.

  • a. If using autocrine cells, one must have a high TGF-α expressor, overcoming ligand uptake by EGFR, or use antireceptor-blocking antibodies 225 or 528. If using antibodies, they should be removed before adding the conditioned media to the sizing column, because they might effect the mol wt standards elution or mask protein detection.

  • b. The antibody can be removed using rabbit antimouse antibodies with protein A-conjugated Sepharose beads. Incubate the conditioned media with rabbit antimouse antibody, protein A-beads for a minimum of 2 h, at 4°C on a rocker. Spin beads down at 130g for 2 min and continue with the supernatant.

• 2. A high concentration of TGF-α (>100 ng/mL) will be required to run on the sizing column; thus, it may be necessary to concentrate the conditioned media using Amicon concentrators (large volumes, >100 mL) or centricons (small volumes, <10 mL) with 3000 mol wt cutoffs (MWCO) done at 4°C.

• 3. Mix 100 ngs of TGF-α with protein standards (albumin, 66 kDa, 2.5 mgs), carbonic anhydrase (29 kDa, 1 mg), cytochrome c (12.4 kDa, 1 mg), aprotinin (6.5 kDa, 3 mg), and 50 µL of glycerol, to a total volume of 1 mL.

• 4. Overlay the TGF-α/protein standards mixture on a 1-m G-50 fine Sephadex column (equalibrated with PBS) with a small buffer head (<2 mL). The glycerol in the protein mixture will help keep the proteins from diluting in the buffer head.

• 5. Elute the protein mixture into the beads, and then add a larger buffer head. Elute the proteins through the column, collecting 5 minute fractions.

• 6. Measure the fractions’ 280 nm to determine when the protein standards eluted off the column. Measure TGF-α concentrations in the elute fractions by a TGF-α ELISA kit.

3.3 Determination of Autocrine Ligand Captured by Secreting Cells

3.3.1 Molecular Devices Cytosensor Operation/Setup

• 1. Plate cells in normal growth media at 250,000 cells/transwell the day before an experiment (4 transwells/experiment), 1 mL of cells/media in the top portion of the transwell and 2 mL of media in the bottom.

• 2. On the day of the experiment, add the spacer and transwell insert to each transwell. To avoid air bubbles, add the spacer or insert into the media at an angle.

• 3. Purge the Cytosensor’s lines with DV-cyto and equilibrate the cells for approx 2 h on the Cytosensor at a 100 µL/min flow rate, 50% pump speed.

• 4. Pump cycle for all experiments is 1 min cycles to measure cells’ quick response to EGF. Thus, the pump cycle is 30 s at 100 µL/min (50% pump speed); 8 s, pumps off delay; 20 s, pumps off, get extracellular acidification rate (ECAR); 2 s, pumps off delay.

3.3.2 Calibration of Cytosensor Readout to Ligand/Receptor Binding

A correlation of ECAR to EGF/TGF-α-receptor complexes can be found by taking the best fit line determined in Subheading 3.3.2.1. , step 4 and Subheading 3.3.2.2. , step 6, entering in a free-ligand concentration, and plotting the resulting number (ECAR and complexes, respectively) as EGF-receptor complexes vs ECAR. Thus, ECAR output from the Cytosensor can be converted into receptor—ligand complex numbers.

3.3.2.1 Cytosensor Readout to Exogenous Ligand

• 1. Plate and set up the Cytosensor as described in Subheading 3.3.1. with B82R⁺/1st plasmid cells.

• 2. After equilibrating the cells with DV-cyto, do a stepwise EGF concentration gradient exposure to the B82R⁺/1st plasmid cells. For example: one EGF gradient series may be 0.05, 0.1, 0.5, 1, 2, 5, 10, 20, and 100 ng/mL EGF. Since the Cytosensor runs four transwells (channels) per day’s experiment, a suggestion would be to leap-frog EGF additions in each channel to minimize previous EGF additions effects. Thus, the run would be Channel A: 0 ng/mL EGF run continuously as blank/control; Channel B: 0.05, 1, 10 ng/mL EGF; Channel C: 0.1, 2, 20 ng/mL EGF; and Channel D: .5, 5, and 100 ng/mL EGF. EGF is diluted in the running buffer, DV-cyto.

• 3. Expose cells for 10 min (or until peak ECAR is reached) to desired EGF concentration, before switching cells back to DV-cyto media and allowing the cells to re-equilibrate their ECAR to baseline. Once cells have re-established their baseline, a new EGF concentration can be introduced to each channel. NOTE: There is a 300-µL dead volume between the inlet and line switch; thus, the EGF solution must be flowing at least 6 min before switching the lines to EGF (Line 1, DV-cyto; Line 2, EGF solution).

• 4. Analyze data by recording the peak ECAR, usually occurring between 6 and 10 min, subtract the baseline, and plot as a function of EGF concentration (free EGF). A best-fit equation can be determined as:

  

  where K _d is the EGF/EGFR disassociation rate constant. Use ECAR coefficient^✻ receptor number combined as one parameter, such as Receptor-ECAR, for the purposes of obtaining a parameter fit.

3.3.2.2 Binding of Exogenous Ligand

• 1. Plate R+/1st plasmid cells into 35-mm tissue-culture dishes in normal growth media and grow cells to confluence.

• 2. Two h before beginning the experiment, switch the cells’ media to D/H/B.

• 3. Add desired I¹²⁵ EGF gradient for 10 min at 37°C. Suggested EGF concentrations are 0.1, 0.3, 0.6, 1.2, 2.5, 5, 10, 20, 30, 40, 50 ng/mL.

• 4. After 10 min, immediately move the cells to 4°C, remove, and save the media. Rinse the cells with 1X WHIPS, and add the rinse to the corresponding media fractions. Count the combined solution fractions on a gamma counter as free EGF. Lyse the cells with 1 M NaOH for 10 min at 4°C. Add 1 mL rinse of 1X WHIPS, and count the lyse cells/WHIPS rinse in a gamma counter as bound ligand—receptor complexes.

• 5. Count an extra, similarly plated cell dish, using a Coulter counter or hemocytometer to determine cell concentration.

• 6. Analyze data by converting free EGF (cpm to ng/mL) and bound EGF—receptor complexes (cpm to complexes/cell, EGF’s MW 6045). Plot as complexes vs free EGF. A best-fit equation can be determined for complexes = Free EGF^✻ receptor number/(Free EGF + K _d ), where K _d is the EGF/EGFR disassociation rate constant.

3.3.2.3 Calibration Curve of B82 EGFR Positive Cells

• 1. Use plots of ECAR vs EGF and Complexes vs EGF, as described in the two previous subheadings, to obtain the best-fit equations and parameter values. Obtain an expected value for ECAR and complex numbers as a function of the same EGF gradient concentration. Plot the expected values as complex numbers vs ECAR. Thus, an ECAR reading from the Cytosensor can be directly correlated to receptor—ligand complex number. See Fig. 1 .

Fig. 1.

Calibration of receptor—ligand complex numbers with ECAR.

3.3.3 Cytosensor Readout of Autocrine Ligand

• 1. Plate and set up the Cytosensor as described in Subheading 3.3.1. with autocrine cells in normal growth media containing tetracycline. Also, equilibrate the cells in DV-cyto buffer containing 1 µg/mL tetracycline.

• 2. Upon obtaining steady-state ECAR, impose the following conditions (one condition per Cytosensor channel-cell):

  • a. DV-cyto buffer.

  • b. DV-cyto buffer, 1 µg/mL antireceptor antibody 225 or 528.

  • c. DV-cyto buffer, 10 µg/mL tetracycline and 1 µg/mL anti-EGFR antibody 225 or 528.

  • d. DV-cyto buffer, 10 µg/mL tetracycline.

• 3. Run these conditions for 24 h and measure TGF-α’s expression induction by the increase in Channel B’s ECAR.

3.3.4 Cytosensor Readout of Antibody Inhibition

• 1. Plate cells and set up the Cytosensor as described in Subheading 3.3.1. with autocrine cells in tetracycline-free (initiates TGF-α expression) growth media.

• 2. Check to make sure cells are responding similarly by exposing the cells to a 10-ng/mL EGF pulse. After cells’ ECAR peaks, remove the EGF and allow the cells to re-equilibrate their ECAR to baseline.

• 3. Leaving one Cytosensor channel as control (DV-cyto buffer only), introduce a desired antibody concentration (anti-TGF-α decoy antibody or anti-EGFR blocking antibodies) to each channel. Steady-state ECAR response should occur within 30 min. Analysis of data can be performed by recording the minimum ECAR value on addition of antibody, or integrating the ECAR response for a 30-min period. For the conversion of ECAR to complexes, use the highest antibody concentration (antibody concentration at which further increases in antibody concentration did not decrease the cells’ ECAR) as a baseline equal to zero receptor—ligand complexes. Complexes in the presence of other antibody concentrations can then be determined from this baseline (see Note 3 ).

4 Notes

• 1. Cell density will have an effect on ligand concentrations in the extracellular media. Thus, all TGF-α concentrations must be measured at similar final cell densities. If necessary, because of slight differences in growth rates, dilute the initial cell concentrations accordingly.

• 2. When using high concentrations of an additive to DV-cyto (i.e., antibodies), accurate results will only be obtained by dialyzing the additive in DV-cyto to remove any additional components (i.e., salts, glycerol, and so on).

• 3. Theoretical predictions indicate that antiligand decoy antibody concentrations will need to be 100–1,000 times higher than antireceptor blocking antibody’s because of transport phenomena. Suggested concentrations for blocking antibodies is in the low µg/mL concentration range (i.e., 0.1–10 µg/mL).