Accurate cells counts w/o microscope...?

Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum

Help Support Homebrew Talk - Beer, Wine, Mead, & Cider Brewing Discussion Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

spenghali

Well-Known Member
Joined
Jan 7, 2009
Messages
457
Reaction score
16
Location
Corvallis
Is there a way to get an accurate cell count for pitching rates without the use of a microscope? I am looking for something other than a pitching rate calculator such as Mr. Malty.
 
On the flip side, thanks to the massive cuts in basic science research these last few years, there are some amazing deals to be had on high-end microscopes on eBay. Good luck! I don't bust mine out too often anymore, but it's a lot of fun.
 
I thought not...alas Christmas is near...

Don't do it. Once you cross that line, there is no coming back. No longer a mere beer nerd, he's a nerd with a microscope! Watch out! If I'd have known how 'sciencey' making beer was, I'd have killed less brain cells in school.
 
Is there a way to get an accurate cell count for pitching rates without the use of a microscope?

Yes, but they are probably not practical for the home brewer. Nephelometers and spectrophotometers can be used for cell counting as can various pieces of biomedical equipment which can, for example, distinguish and count RBCs and the various types of WBC's.

A nephelometer measures the amount of light scattered by yeast cells (and any other particle so trub, protein globs etc become a factor and a photometer measures light absorbed so that the color of the broth must be taken into account but once either instrument is calibrated it should be useable to the level of accuracy needed for this application.
 
I've never used a nephelometer, but I have used a spectrophotometer. How would you calibrate them if not with a microscope and hemocytometer? Plus, wouldn't recipe formulation and idiosyncratic brewhouse parameters make calibration a moving target?
 
Yes, but they are probably not practical for the home brewer. Nephelometers and spectrophotometers can be used for cell counting as can various pieces of biomedical equipment which can, for example, distinguish and count RBCs and the various types of WBC's.

A nephelometer measures the amount of light scattered by yeast cells (and any other particle so trub, protein globs etc become a factor and a photometer measures light absorbed so that the color of the broth must be taken into account but once either instrument is calibrated it should be useable to the level of accuracy needed for this application.

Why count through inference when you can actually count? I've seen these Amscope microscopes for sale cheap (~$200 or less?). Cell morphology isn't possible but simple counting certainly is. A capable hemocytometer can be bought new on Ebay for $30.
 
That's how I would do it (because I have a 'scope and hemacytometer) but you could use any particle counting system you trusted. It would be the same concept as in using a refractometer for measuring alcohol. You would develop calibration curves for individual beers in your portfolio in the lab and then use the nephelometer in the brewery when brewing that beer for a much quicker check. I might be able to come up with a few general curves e.g. one for lager yeasts and one for ale yeasts, which you an others could use that would give an approximate count. How accurate would you have to be?
 
I haven't used ImageJ for automated counting yet. But I use it for manual counting.

As for a practical alternative to counting cells, I would say pitching yeast by weight. Unfortunately there is some strain and obviously slurry dependency. But for well flocculating strains, where you get stable sediment, I found the cell count to be 3-4 Billion cells per gram in clean sediment.

I actually suggest that even brewers who have a microscope may want to establish a correlation between slurry weight and cell count since counting cells takes time and you may want to skip it at some point.

Kai
 
Image J for manual counts is very nice. Especially for documenting results and checking your work, but for most cell counts I just use a clicker. The automated cell counting looks like it would take quite a bit of tweaking to get consistent results. The automated bit might be useful if you had weeks worth of counts to perform. Maybe months.

Some of my slurries are 1 billion cells per gram, others are 250 million cells per gram. A white labs vial I counted recently was 2 billion.

I use a microscope and hemocytometer. It works great and has opened my eyes to a whole new part of the brewing process. The variation from one slurry to the next, but my day to day results are very consistent, so I feel I am getting accurate results. but of course, the proof is in the pudding, or in this case I suppose, beer.

Here is the microscope and some pictures of yeast:

http://woodlandbrew.blogspot.com/2012/11/amscope-binocular-compound-microscope.html
 
If time isn't an important factor, you could count CFUs using agar plates. But personally as much as I hate counting cells with a hemocytometer, I hate making serial dilutions, plating, and counting colonies even more.

There are good ImageJ macros out there for semi-automated cell counting. It works well as long as you don't have a lot of yeast-sized debris in your sample.
 
Why count through inference when you can actually count? I've seen these Amscope microscopes for sale cheap (~$200 or less?). Cell morphology isn't possible but simple counting certainly is. A capable hemocytometer can be bought new on Ebay for $30.

I agree. That's the setup I have and it's worked well for me. It's fairly fast too. I can get down to about 5% standard deviation in about 15 minutes of work including dilutions and staining.

Here is a side by side comparison of the $30 to a $260 hemocytometer:
http://woodlandbrew.blogspot.com/2012/12/hemocytometers-side-by-side.html
 
If time isn't an important factor, you could count CFUs using agar plates. But personally as much as I hate counting cells with a hemocytometer, I hate making serial dilutions, plating, and counting colonies even more.

There are good ImageJ macros out there for semi-automated cell counting. It works well as long as you don't have a lot of yeast-sized debris in your sample.

Agreed and Agreed.

Is counting CFU's more accurate than a hemocytometer? I've considered it, but wasn't sure if it would really be worth it. With the Hemocytometer you are diluting 20:1 or 40:1 but with plating is serial diluting to something on the order of 1E6:1
 
Agreed and Agreed.

Is counting CFU's more accurate than a hemocytometer? I've considered it, but wasn't sure if it would really be worth it. With the Hemocytometer you are diluting 20:1 or 40:1 but with plating is serial diluting to something on the order of 1E6:1

I'm sure some microbiologists out there would disagree with me, but I think the hemocytometer plus something like methylene blue is as good or better than counting colonies for exactly the reason you brought up. There's some error rate with methylene blue, but I think it's lower than the error you introduce by diluting something 1E6:1 or more.

In order to get good numbers from counting colonies, your starting sample and serial dilutions have to be perfect. And even when you do it perfectly with lots of vortexing, perfectly calibrated pipettes, etc. there's still variability so people do everything in triplicate and average the results. :drunk: Not worth it for these purposes, IMO.
 
Is there a way to get an accurate cell count for pitching rates without the use of a microscope? I am looking for something other than a pitching rate calculator such as Mr. Malty.

Why? If you're using fresh yeast and pitching the amount that Mr. Malty recommends, you really can't go wrong. I haven't heard anyone yet complain that it didn't produce good results. You could always pitch a little more than what Mr. Malty recommends if it's a concern. Do you want a microscope just to see how many are actually there?

I'm just curious as to why you're looking for something other than a pitching rate calculator.


Edit: After reading this, the tone didn't come through - I really am just curious. No disrespect intended....
 
I'm not a microbiologist, but this is what I have gathered:

Plate counts:
- take time
- do not show you if cells are clumping together. Each clump will be one colony. You'll have to inspect the diluted sample to check for clumps
- only counts cells that are actuallt viable. You are able to establish a correct viability count by matching cell density from hemocytometer with growth colonies.

MB staining
- quick
- you are able to see clumps of cells since you have to look through a microscope
- cell viability can be grossly overestimated compared to pate counts. This is b/c methylene blue may not be able to enter a dead cell. I had old cultures that hardly stained with MB but nothing grew on a plate.

MB staining is useful to check the health of a culture. If you find that less than 10% stain, you can assume that the culture is fairly viable, If more than 10% stain you have an old culture that you should not pitch anyway.

Kai
 
Kai,

This is the first time I noticed your location. You're only about 40 minutes from where I work. Small world.

I recently plated some WLP004 that measured 10% viability with MB and got approximately the number of colonies I was looking for. This was a pretty rough experiment so I want to re-run it, but I wanted to mention it. I also fermented a sucrose wort with 10% viability EC-1118 recently without a problem.

What I have noticed about Methylene Blue is that different strains take different amounts of dye to stain well. EC-1118 works well with equal parts 0.1% solution and diluted cells, while WLP566 stains fine with only one tenth of that concentration at 0.01%. If there is significant protein material it may take a little more stain as well. Normally I start with a 2:1 dilution of cells by 0.03% MB and if the dead cells are just a little bit blue I'll measure what is left in the test tube and add that much of 0.1% MB. This gives me a 4:1 dilution of about 0.06% MB. If you have too much MB the slide will look dark under the scope.
 
Kai,
This is the first time I noticed your location. You're only about 40 minutes from where I work. Small world.
Cool. I'm in the Woburn area on a regular basis.

What I have noticed about Methylene Blue is that different strains take different amounts of dye to stain well. EC-1118 works well with equal parts 0.1% solution and diluted cells, while WLP566 stains fine with only one tenth of that concentration at 0.01%. If there is significant protein material it may take a little more stain as well. Normally I start with a 2:1 dilution of cells by 0.03% MB and if the dead cells are just a little bit blue I'll measure what is left in the test tube and add that much of 0.1% MB. This gives me a 4:1 dilution of about 0.06% MB. If you have too much MB the slide will look dark under the scope.

I have not played around with determining the reliability of MB staining or improving its accuracy. I'm citing from the literature sources and this is one of the charts I have come across in a German presentation from the Weihenstephan brewing school:

YeastVitalityMethods.jpg
 
That is very interesting. Your results and this citation have definitely brought this to a new light for me. This will certainly be in my mind when I measure low viability.
 
Why? If you're using fresh yeast and pitching the amount that Mr. Malty recommends, you really can't go wrong. I haven't heard anyone yet complain that it didn't produce good results. You could always pitch a little more than what Mr. Malty recommends if it's a concern. Do you want a microscope just to see how many are actually there?

I'm just curious as to why you're looking for something other than a pitching rate calculator.


Edit: After reading this, the tone didn't come through - I really am just curious. No disrespect intended....

That's just it, I don't buy very much fresh yeast. I wash a lot of yeast and top crop sometimes. I am also a very scientific person, and my studies are going in this direction anyways so I figured I might as well dive in head first. Great discussion so far, thanks for the input, I had forgotten about this thread.
 
Back
Top