20.109(S08):Transcript-level analysis (Day5)

Protocols

Part 1: Day 1 of ELISA

Plan samples in previous FNT?? Either way, upload the 96-well planner document.

We will run this assay in a 96-well microtiter plate, as we did for the fluorescence titration curves in Module 2. In ELISA, we will be testing for absorbance at a particular wavelength, rather than emission.

1. The first step in indirect ELISA is to adsorb all your samples to the wells. You will also need to prepare standard samples in the same plate, which get treated just the same as your test samples. These standards will be used as a reference for protein concentration.
2. Students do doubling dilutions or we prep in bulk?
3. Add 50 μL of sample per well. Cover the the plate, wrap around it with parafilm to better prevent evaporation, and allow the samples to sit for 90 minutes. In the meantime, set up your agarose gel (Part 2).
4. After the incubation time has passed, you will wash and then block your plate.
   • First, flick the solutions in the plate into the sink.
   • Using the multichannel pipet and a reservoir, add 200 μL of Wash Buffer to each well, then gently swirl the plate (by hand) for a few seconds.
   • Flick the solutions out again, and then blot the plate against paper towels. You can smack the plates pretty hard, but it is possible to break them!
   • Repeat the wash one more time.
   • Finally, add 200 μL of Block Buffer to the plate. Wait another 60-90 minutes. In the meantime, work on your analysis (Part 3).
5. Repeat the wash step that you performed above, again with two rinses.
6. When you are ready, ask the teaching faculty for some primary anti-collagen antibodies (these should be diluted at the last minute). Add 100 μL of diluted antibody per well.
7. Your samples will be left overnight (?longer?) in antibody solution, then moved back to block buffer by the teaching faculty.

Part 2: Agarose gel of PCR fragments

1. Today we will run a 1.2% agarose gel. (Why do you think we would use a higher concentration of agarose than for the samples in Module 1?)
2. Prepare enough diluted loading solution for 9 samples: 2 μL loading dye + 10 μL sterile water per sample.
3. Mix 10 μL of each PCR product with 12 μL of diluted loading solution.
4. Load your samples according to the following table, being very careful to load the same amount in each well:

Lane	Collagen type	Sample	Volume to load (μL)
1	N/A	100 bp ladder	11
2	CN I	2D cult.	18
3	CN I	3D cult. 1	18
4	CN I	3D cult. 2	18
5	CN II	2D cult.	18
6	CN II	3D cult. 1	18
7	CN II	3D cult. 2	18
8	CN I	control rxn.	18
9	CN II	control rxn.	18

• The samples will be run for ~ 45 min at 125V. The teaching faculty will then assist you in getting pictures with as high contrast as possible, then scan the pictures for analysis. (The digital photos will be immediately transferred to your laptops on a flash drive, and they will also eventually be posted on the wiki.)

Part 3: Begin ImageJ Analysis

Today you will use an image analysis program called ImageJ. This is offered free of charge by the NIH (National Institutes of Health). Your goal is to determine the collagen II: collagen I cDNA ratio for each of your samples, based on the intensity of the bands. This is one piece of evidence for the overall question posed by this module, namely, what factors affect chondrocyte phenotype maintenance (vs. de-differentiation to fibroblasts), and to what extent? If you finish the transcript-level analysis, you can move on to quantifying your live/dead data.

Transcript intensity

1. Open your agarose gel image by selecting File → Open
2. Using the rectangle tool, select an area in the background of the gel, the hit ctrl-M to measure it. Copy the mean and max intensity to an Excel file, and save the file.
3. Now select each of your collagen bands in turn and repeat the process. Ultimately, you would like to compare the intensity of the CNII to the CNI band for each sample in an equivalent way. You have a lot of choices to make at this point, and there is no single right way to analyze your data.
   • As one option, you could select equal areas from the center of each band, to avoid edge effects (where the band is dim) on mean intensity.
   • On the other hand, if both bands are equally bright, a thicker band would be indicative of more DNA. So you might want to compare both areas and intensities, instead of just intensities. It’s up to you. Just try to make a good case for what you are doing.
   • However, there are certain best practices for taking a ratio. For example, you probably want to subtract the background intensity from each sample band value.
4. If you have time, you can explore the built-in gel analysis program in ImageJ.

Cell viability

Your goal for this section will be to compare the effort required for, and the resulting accuracy of, manually counting live and dead cells vs. doing so by semi-automated image analysis. After you are done, you might consider under what conditions you might prefer one method or the other.

1. Open your live cell image (green filter) by selecting File → Open
2. Choose the line tool, and draw a line across the diameter of a typical cell
3. Select Analyze → Set Scale, and put 10 μm in under Known Distance
   • or get actual info from camera if possible
4. Convert your image to grey scale using Image → Type → 8-bit
5. Now you must somehow select your cells out from the background
   • One way to do this is by choosing Process → Binary → Make Binary. However, you might find that clusters of cells are ‘read’ as a single cell.
   • An improved method is to use Image → Adjust → Threshold. You can set the upper- and lower-bound intensities that define objects in your image. Play with the intensity sliders until your cells are mostly filled in with a red colour, but not overlapping with other cells whenever possible.
6. Now you can count the objects. Choose Analyze → Analyze Particles, and select Show →Outlines, Display Results, Summarize, and Record Stats. Also choose a reasonable area (not diameter!) range for objects that are cell-sized.
7. Try to play around with this process for a bit – are there any further changes you can make so the automated algorithm is as good as your eye?
8. Record your final results (manual and/or best automated) in your notebook, for each sample that you have data for. Don’t forget to also count the dead cells, and subtract this number from the live cells (since the red colour usually bleeds into the green channel).

For next time

-oral: presentation outline (maybe skip this)

-essay: finish