Home
Hi,
I'm a brewer with a passion for culturing my own yeast – which I store on agar slants with some success. When it comes to time to use the propagated yeast, it is important for me to be able to do accurate cell counts to assess the viability of the culture.
I bought a 400x stereo microscope by salter, and a haemocytometer, which work together pretty well and I'm able to do some counts. The problem is with sample prep – the yeast is too flocculent (clumps together readily) so I find that the sample I count isn't representative of the entire sample in my glass flask.
How do cell counters create a sample to count which is an accurate representation of the entire batch? Errors at the cell count stage seem to make my viabilty estimates utterly useless.
I realise this might be rather specialist for the forum, I guess I'm looking to learn more about cell counting in general, and slide sample preparation.
Many thanks in advance.
Hi jat147, I googled "counting yeast cells using hemocytometer" and got a number of results including some PDF files with description of methods, but I don't know if any would specifically answer your questions.
01:29
December 29, 2011
Offline
I realize that this topic is a few months old, but hopefully this can be of some help anyhow. It seems to me that your main problem is that your yeast mixture is heterogeneous. I have a limited knowledge about brewing, but as far as I understand it it is common practice to inoculate yeast cultures in a "shaker" that constantly stir the mixture. Would it not be possible to stir the mixture gently to make it more homogenous and then take a sample?
When it comes to the actual cell counting, I have had some (very basic) training as part of my bachelor in biology. I will try to put what I have learned into words here.
If we want to know the concentration of cells or microorganisms in a suspension, we can use a counting chamber. A counting chamber is a glass slide with a grid engraved into the glass and a profile that give the suspension a certain depth under the cover slip. When we know the size of the grid and the height under the cover slip, we can calculate the volume of the sample we are about to count.
There are several different counting chambers out there, and they differ in the size of the grid and the depth. Counting chambers with a large depth and a large grid such as "Fuchs Rosenthal" are suitable for counting large cells such as algae, while counting chambers with a smaller grid and a narrower depth such as "Thoma" are more suitable for counting bacteria.
The following table show the relationship between different counting chambers and the number of cells/mL at 1 cell/square:
Table 1 – Counting chamber data
|
Counting chamber
|
Depth, mm
|
Grid, mm2
|
Volume, mL
|
1 cell/square =
|
|
Thoma
|
0.02
|
0.0025
|
5*10^-8
|
2*10^7 mL-1
|
|
Bürker
|
0.1
|
0.04
|
4*10^-6
|
2,5*10^5 mL-1
|
|
Fuchs Rosenthal
|
0.2
|
0.0625
|
1,25*10^-5
|
8*10^4 mL-1
|
Put a drop of your suspension on the grid and cover it with a cover slip. Make sure you don't get any in the "pits" on the side of the cover slip as this will affect the precision of your calculation. Decide what square size you'll do the count in. If there are more than ten cells in one of the small squares or more than 70 in the large, it is easy to miss some of them and get less than accurate results. Use the data from the table above or the data supplied with your counting chamber along with the cell count to calculate the number of cells/mL. If we assume that counting cells with a counting chamber is a Poisson distribution we can calculate the counting error (Δn) with the following formula:
Where n = all the counted cells
So, if you count 100 cells, the counting error is 10, which = 10%. If you count 1000 cells, the counting error is 31,6, which ≈ 3,2%.
This counting error only say something about how accurate the counting is. Other factors affecting the accuracy of your results can be uneven distribution of cells in the mixture, measuring errors, etc, and these need to be dealt with by taking several samples (at least 3) and calculating standard deviation. It recommended to count 2-300 cells, this give a counting error of 6-7%. Count random squares, and squares at different locations on the sample.
Example:
I want to count the number of yeast cells in a random sample. I use a Bürker counting chamber, and I count 30 squares. I get the following data:
|
Square #
|
number of cells
|
Square #
|
number of cells
|
Square #
|
number of cells
|
| 1 |
12
|
11 |
7
|
21 |
3
|
| 2 |
2
|
12 |
10
|
22 |
11
|
| 3 |
8
|
13 |
7
|
23 |
8
|
| 4 |
7
|
14 |
9
|
24 |
8
|
| 5 |
10
|
15 |
4
|
25 |
10
|
| 6 |
3
|
16 |
6
|
26 |
11
|
| 7 |
8
|
17 |
14
|
27 |
7
|
| 8 |
6
|
18 |
9
|
28 |
7
|
| 9 |
8
|
19 |
9
|
29 |
1
|
| 10 |
6
|
20 |
4
|
30 |
4
|
|
Total:
|
70
|
Total:
|
79
|
Total:
|
70
|
This gives us on average 7,3 cells/square. Using the data from table 1 I get the following:
Bürker: 1 cell/square = 2,5*105 mL-1
(2,5*10^5 mL^-1) * 7,3 = 1825000 cells/mL
Counting error: Δn = √n = √219 = 14,8
Δn = 14,8/219*100 = 6,75799 ≈ 6,76%
I would then take two more samples and calculate the standard deviation.
A large part of this text is translated from the compendium "BIO 101 Organismebiologi 1, Laboratorieøvelser, Mikrobiologi & protister Ver. 2013" by Gunnar Bratbak, Institute for Biology, Universitety of Bergen.
Most Users Ever Online: 149
Currently Online:
8 Guest(s)
Currently Browsing this Page:
1 Guest(s)
Member Stats:
Guest Posters: 31
Members: 593
Moderators: 0
Admins: 1
Forum Stats:
Groups: 2
Forums: 18
Topics: 1403
Posts: 5585
Newest Members: mariareese, Habibie, EoinFirestorm, Peniko177, Microseeshirt, SteveRichfield
Moderators:
Administrators: Oliver (201)
