[Polleke]

Has anyone here got some experience with the use of dyes to detect living organisms in samples?


Or any absorption spectrophotometry specialist?

I was wondering the following thing:

you use the dye to measure the viability of micro-organisms. Living micro-organisms will reduce the piant and thus you will measure another absorption.
But I was wondering: what if you have dead tissue in your samples, this dead tissue does influence what you measure , even if you do not have any colorchange in the dye, right?

Or ?

[ARGOS++]

Dear Polleke;

Don’t be surprised, when the “whole” answer will become quite long!

To overcome this problem we have finally to fit a few “Partial Pictures” together, and to think about, what for Simplifications may be allowed to keep still a certain Accurateness!

KEYWORDS for the Discussion:
(A Subject that may be expanded during Discussion!)
     Living Micro Organism (LMO),  Dead Micro Organism (DMO),  Culture Medium (CM)
     Initial Dye (ID),  Degraded Dye (DD),
     Simple Scattering Correction (SSC),  Multiple Scattering Correction (MSC),
     Multi Component Analysis (MCA),  Multivariate Data Analysis (MVDA)
     Measuring Techniques (MT),  Reflectance (RE),  Absorbance/Absorption (AU)
     Wavelength (WL),  Spectral Range (SR),  Wavelength Range (WR)
     Observation/Experiment Time/Duration (OETD)


Picture  0:  Hints for the Measuring Techniques (MT) and Cuvette (CV):
To avoid as much as possible Problems you should measure the System in a Cuvette (CV) and in Units of Absorbance (AU), because only Absorption is linear to the present Concentration!
Don’t use “Petri Dish(s)”, because you are not able to know either the Path-Length, nor are able to keep it constant during the Experiment or from Experiment to Experiment!
Maybe the “Best” will be, if you measure Spectra and not only a few  WL’s.
There will anyway enough Problems be left to solve!
More about in a later Picture!

Picture  1:  Living Micro Organism (LMO),  Dead Micro Organism (DMO):
As you already mentioned, it may be not enough to measure only how the ID Concentration will decrease in time; and even doing only that results already in one or more Problems.
The cause for that are the LMO’s and the DMO’s, but especial the LMO’s, because their Number will not be constant in time. But even (more) worse, I don’t believe the increase will be linear in time. I believe contrary that as long as enough CM is available, the Increase will be something like exponential!
Maybe this LMO model has to be studied separately without ID.
Also DMO’s are a Problem, but of another kind, and may also not be constant in time too.
DMO’s before start of the Experiment will only give a constant Amount to the Scattering Model.
DMO’s generated during the Experiment will give an increasing Amount to the Scattering Model, but may also still contain some ID and influence the real available and start Concentration.
(It could be very helpful, if the DD has a different, but still measurable Spectrum!)

Picture  2:  Initial Dye (ID),  Degraded Dye (DD):
Picture  3:  Simple Scattering Correction (SSC),  and MSC:
………
Picture  n:       Combined “Picture”:
Picture  n + 1:  Possible Simplifications:
Picture  n + 2:  APP  = Analysis (of) Potential Problems:
………
Picture  m:       Measuring Model:    Spectra, MCA, and/or MVDA:
Picture  m + 1:  Measuring Experiment
Picture  m + 2:  Data Analysis:
………

I hope you got already a miner Idea of what all will influence your results and your Accurateness.

Maybe some have already done some more or less reliable work, or Simplifications: "Almar Blue”.
(The Fluorescence method on the same page may have less such Problems,but knows others, but I don’t really know!)
I was not able to find a real good reference about the method yet.


Please tell me if you are still interested in this discussion.
(If Yes! it means that you also inform me, as soon as you will later stop the discussion!)


Good Luck!
                    ARGOS⁺⁺

[Polleke]

Thanks a lot ARGOS++ for your response.
I allready know a lot of the things you told me, but still the questions stays: how good is this method to figure out how viable your samples are.
(its good, but I wonder what kind of mistakes are possible to be made. By this I mean : can I measure reductions if samples are dead? Yes you can (I know this, it happened allready to me), but then what? Does this mean that below that value everything is false positive or ..? and.... so on)



I have been using this dye now for a few months and I have been asking me the same question over and over: what if my fungus is dead and I pipet dead tissue in my wells, I will then measure absorbance or not?
To me it seems logic that I will measure absorbance because the dead tissue does absorb light or prevents the light from passing.
I am using a simple absorption spectrophotometry.

I agree on you that fluorescence is better, more aquarate.


I use microwells, plates to measure.
(I guess you know that?)


I have to admit, that I did not program everything myself.
I only measured how long it would take to reduce the paint for 100% and I optimalised the concentration of the micro-organism I use.

So I know the incubation time and the concentration of the organism (+-)

The questions stays: what is a reduction of about 10% Can I get 10% reduction (measuring it) just because of dead tissue? Or not?

I have allready noticed that when my samples are a bit more acid , the dye will switch from blue to red itself because of the pH of the samples... thus I measure reduction, while this reduction is caused by the pH and not the living organisms. (or not completely)



Anyway, you made the problem even worse lol


The thing is that when you do a search about the dye, you will never find a true or complete explanation about how it works, what influences you have...
its always very basic how they say it : the dye will change color when organims are alive, thus you will maesure another absorbance thus another reduction (and reduction is a measurement to say something about how viabel they are).


I think the main problem for me is that I am working with micro-organisms and study them and I do not have all the chemical knowledge to fully understand and know about the absorbance technique.

The problem is that in most cases, we just do not ask questions and simply use the dye as told in the manual you have.
The manual itself is very basic and easy.

this is the manual I have:

http://www.ab-direct.com/catalog/datasheetfactPdf.ashx?ProductID=78698


As you can see, its good, but its not good enough to answer my questions.

The cause for that are the LMO’s and the DMO’s, but especial the LMO’s, because their Number will not be constant in time. But even (more) worse, I don’t believe the increase will be linear in time. I believe contrary that as long as enough CM is available, the Increase will be something like exponential!

you are offcourse right, the growth is exponential.



Maybe I forgot to mention: you do not measure the exact amount of organisms in your sample, you measure the viability.
And the reduction you measure is used as a guidance to find the viability.
(you search for the optimum incubation time and concentration to figure you when you have a 100% reduction, wich then is your 100% viabilty, based on that number you recalculate all the other measured reductions)

[ARGOS++]

Dear Polleke;

I don’t believe that I made it worse, because it is already!

Thank you! for the Link of the method you use!
It is nearly as bad as I expected.

The problem is that in most cases, we just do not ask questions and simply use the dye as told in the manual you have.

That's common to nearly all Labs, because thinking is a hard job!

I use microwells, plates to measure. (I guess you know that?)

That's why I wrote: Don’t use “Petri Dish(s) ”.

I have to admit, that I did not program everything myself.

I never had such in mind, and it's also not required!

I have allready noticed that when my samples are a bit more acid , the dye will switch from blue to red itself because of the pH of the samples... thus I measure reduction, while this reduction is caused by the pH and not the living organisms. (or not completely)

That's precisely why you have to use an adequate Buffer!


I realised from:

. , but still the questions stays: how good is this method to figure out how viable your samples are.
(its good, but I wonder what kind of mistakes are possible to be made. By this I mean : can I measure reductions if samples are dead? Yes you can (I know this, it happened allready to me), but then what? Does this mean that below that value everything is false positive or ..? and.... so on)

..

I have been using this dye now for a few months and I have been asking me the same question over and over: what if my fungus is dead and I pipet dead tissue in my wells, I will then measure absorbance or not?
To me it seems logic that I will measure absorbance because the dead tissue does absorb light or prevents the light from passing.
I am using a simple absorption spectrophotometry.

- and from your whole last Posting, that you don’t know where your Formula (Eq. 2 on page 18) is coming from, and so you are not able to estimate the different possible influences.
But we are even less able to do it, because we own no experimental data, contrary to you!

For explaining the Formula to you/all the following Ingredients are required:
   -   The Linear Equation System.
   -   The “Special Beer-Lambert Law”: "Corrected Beer-Lambert Law”.
and:
   -   The Spectra on the Figure 3b of page 15.

The thing is that when you do a search about the dye, you will never find a true or complete explanation about how it works, what influences you have...
its always very basic how they say it : the dye will change color when organims are alive, thus you will maesure another absorbance thus another reduction (and reduction is a measurement to say something about how viabel they are).

As it requires a little wile to prepare/write all explanations required for your Formula, you may also study:  "Resazurin”.

So: “See you in a wile   — ‘Crocodile’”, if you don’t have other Questions in between.

Good Luck!
                    ARGOS⁺⁺

[Polleke]

I have to read your answers again to fully understand it all, bit I allready want to say one thing:
a petridish is not the same as a multiwell.
http://www.nuncbrand.com/getbinary.ashx?id=10431&lang=en&filename=96-MicroWell-Plates-Plates-.jpg&filesize=17226 ==> microwell

But indeed for measuring it might be the same?

A multiwells is a collection of tiny petridishes in 1 plate.

And why is it better to work with a cuvet?
Because with a cuvet you do not send the light true the bottom and top and thus not true any sedimentation? or is there another reason?

(by sedimentation I mean , in this case,  growth of micro-organisms at the bottom of your wells. This growth will also infleunce the reduction you measure....)



and adding a buffer to my solution is almost impossible because the buffer will also influence my results , because the buffer will influence my micro-organisms and thus their growth..


anyway, its a very hard for me to fully understand it all.

The thing is that in most cases it doenst even matter (even now, its not that important for me to fullu understand it, if I write something about it, I do not need to explain all this, but I just want to understand it better so I can have a better insight on the results.)

[ARGOS++]

Dear Polleke;

Only on the fly:

Sorry! As it seems that I have you misguided  with my “Petri Dish(s) ”!

I know “MultiWell Plates” quite ‘well’, because I have not only used them, I did also some programming for “them”.
My message was only for telling that Petri Dish(s) are the worst case and “MultiWell Plates” the second. —  Why?
Maybe you have already read on the “Beer-Lambert Law” above, that d, the path-length is an important parameter for accurate measurements of the Absorption.
But not to know your d,  and only praying that it may be constant and not to different from well to well is never a big improvement for the Accuracy of the method!

About the Buffer:
If you follow ‘exactly’ the written Procedure you are already using an adequate Buffer.
(Re-read the “Indication for Use:” on page 2! = pH 6.8 till pH 7.4).

Best Advise I can give you:
  Keep your “Stomach Feelings”, and try to get ‘behind’ it, even if it will be hard at begin!
  (You can be secured, with time you will get trained!)

Good Luck!
                    ARGOS⁺⁺

[Polleke]

Oh ok,

then I misunderstood you, I was under the impression you were saying that microwells and cuvets were the same.


Anyway: I understand it, that the longer the light has to pass, the worser the results.

And indeed, measuring in a cuvet is then better, but its impossible in this case.

and remember what I said about sedimentation: in a cuvet this is no problem, in a microwell it is....


===> when I have growth in my microwell, I will influence the results, because the cells at the bottom of my plate will stop the light and thus change the results... (even worse: imagine they are dead, then you have a lot of absorption, but no reduction of the dye, so you measure a lot of growth (reduction), but there is no reductiuon at all!)


One remarks: does it matter that much? I mean: the microwells arent that deep, to be honest, I think that the distance true a cuvet and a microwell is almost the same... or am I that wrong here?


another thing: you say something about the path lenght, not being constant, I know for sure that the dept of the wells is the same... I mean by this that they are really build to have the same volume, same distance..

BUT, there is a but: I offcourse do not know if the volume (see also my next question)I pipet in the well is always the same (it should be , normally always 100µl, but well, you can make mistakes during the use of a pipet, its not 100% certain + when dead tissue in inserted, this will again have an influence... )

How big this influence is... no one knows.




Ok , and know a (I think) stupid question: the path length influence the measurements, but do they by this mean the thickness of the sample you measure or also the volume of your "microwell" itself.
To be more clear (and offcourse this only is important for when you use microwells, because in a cuvet, the cuvet itselfs is always filled from one side to another side) I mean this: imagine I have a well wich is 5 cm deep, I fill it for 1 cm  thus meaning that I have 4 cm of "air" , and another one I have 5 cm deep but I filled it for 4 cm...

Then I assume I will have different results, because 4 cm of sample will absorp more then 1 cm, but how does the machine know this? Does the machine recalculate this?? OR not?

Because I do not understand it that I did not had to enter the depth of my filled samples.

I mean: how does the spectrophotometer knows I filled it with 100µl , resulting in 2mm thickness  ?? and not by example 5mm ?


THis problem does not occur when using cuvets , because you can only put in 1 cuvet size in your machine... and the cuvet is always filled from one side to another side...


(in an ideal situation, this problem is irrelevant, because I will always have my wells filled with the same volume , resulting in the same "hight" of my sample.... but when is something ideal?)



You see, its getting even more complicated to understand...



About the buffer issue : I know its buffered from 6.8 till 7.4 , but even this wasnt enough for my tests...

[ARGOS++]

Dear Polleke;

Let’s start with your Formula (Eq. 2 on page 18), because it may also help you for quite a number of your other Questions:

• A.)   The Linear Equation System:
  We need only a very simple form of it:  Two Equations!
               a₁ * x    +   b₁ * y  = A₁
               a₂ * x    +   b₂ * y  = A₂
  You may know that the solution for x will be:
          (a₁ * b2 – a₂ * b₁) x   =  A₁ * b₂ – A₂ * b₁  resulting in:
          x  =  (A₁ * b₂ – A₂ * b₁)  /  (a₁ * b₂ – a₂ * b₁)
  and for y the solution will be symmetric.
  
  
• B.)    The “Special Beer-Lambert Law”:
  I hope you have in the meantime discovered what all is required for more or less accurate measurement of any Absorption (incl. the conversion from Transmittance).
  We will take use of that on the next.
  
  
• C.)   The Spectra on the Figure 3b of page 15:
  Have you already realised, that the Spectra don’t reach the “Baseline”?
  We can only guess that they have forgotten to subtract the Spectrum of the CM!
  Otherwise:  Oh dear, oh dear, it will be a big source of Error, because as you can read on the  WL of 700 nm, that the “Error” will be 0.05 AU’s, and that’s ~10% of the Signal at Lambda_max!
  So I hope you followed exactly the Method:
  

  Measure Absorbance at WL of 570 nm and 600 nm after required incubation. Use a blank of  CM only!

  
  

Now let’s see what you are measuring (Yes! through the Wells!) on both WL’s.
As you can read from the Figure 3b of page 15, that you have Signals from both forms of the Dye (ID & DD)!
With the help from Beer-Lambert we get:
      A₆₀₀ =  ε_ox600 * c_ox * d   +   ε_red600 * c_red  * d  +  A_CM600   +  A_SCM(MO)600 + A_Plate600

      A₅₇₀ =  ε_ox570 * c_ox * d   +   ε_red570 * c_red  * d  +  A_CM570   +  A_SCM(MO)570 + A_Plate570

If we now assume, that you used a ‘well’ prepared “Blank”, and that all your Plates have an identical thickness of their Bottom, where you measure through, then it is allowed to drop the terms:  A_CMxxx and A_Platexxx.
But they are still a source of Inaccurateness!

So now we stay with:
      A₆₀₀ =  ε_ox600 * c_ox * d   +   ε_red600 * c_red  * d  +  A_SCM(MO)600

      A₅₇₀ =  ε_ox570 * c_ox * d   +   ε_red570 * c_red  * d  +  A_SCM(MO)570

“Your” Method simply ignores the term A_SCM(MO)xxx, which you identify as a possible Source of Inaccurateness, because it contains both: The LMO’s and the DMO’s!
Unfortunately I cannot estimate its Influence without any experiment, as it depends on several unknown parameters!

And now our reduced Equation System looks:
      A₆₀₀ =  ε_ox600 * c_ox * d   +   ε_red600 * c_red  * d

      A₅₇₀ =  ε_ox570 * c_ox * d   +   ε_red570 * c_red  * d

The Method also assumes that changes in d will not significantly influence the Solution and drop it too!
This Assumption may/will hold, as long as you are very precise in 'voluminas' and the Diameter of the measuring beam is large compared with the Diameter of the Wells!

That results in:
      A₆₀₀ =  ε_ox600 * c_ox   +   ε_red600 * c_red

      A₅₇₀ =  ε_ox570 * c_ox   +   ε_red570 * c_red

This Equation System you can now compare with A.) above (A₁ = A_xxx, a₁ =  ε_oxxxx, etc.)!
If you now use the Solution for x from A.) and do the whole “Translation”, then you will end in a Formula that gives you the Concentration of your initial Dye (ID)!

That’s the simplest, most primitive, less accurate, and not validate-able Form of any MCA!

Finally you can do the identical for your 0.0 % “Viability” and calculate the %-age, and you will end in exactly the given Formula (Eq. 2 on page 18).
But in this moment you do already an add. Inaccurateness!

Have you not forgotten on the way till here to summarise and weight all sources of Inaccurateness?

I hope you will now have a little more complete “Picture”! from your Method in use!

Good Luck!
                    ARGOS⁺⁺

[Polleke]

Ok, thank you very much for your inisghts ARGOS++.

I'll need to take some time to check it out completely to be able to understand it correctly.

thanks a lot