Go Back   Science Forums Biology Forum Molecular Biology Forum Physics Chemistry Forum > Molecular Research Topics Forum > Protein Science > ELISA Assay Forum
Register Search Today's Posts Mark Forums Read

ELISA Assay Forum ELISA Forum. Discuss and post questions including optimization, troubleshooting, protocols and theory of ELISA assays using monoclonal, and polyclonal antibodies or peptides.


Linearity-of-dilution concept

Linearity-of-dilution concept - ELISA Assay Forum

Linearity-of-dilution concept - ELISA Forum. Discuss and post questions including optimization, troubleshooting, protocols and theory of ELISA assays using monoclonal, and polyclonal antibodies or peptides.


Reply
 
LinkBack Thread Tools Display Modes
  #1  
Old 03-29-2011, 03:25 PM
Pipette Filler
Points: 124, Level: 2 Points: 124, Level: 2 Points: 124, Level: 2
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Mar 2011
Posts: 3
Thanks: 0
Thanked 2 Times in 2 Posts
Default Linearity-of-dilution concept



Hi everyone,

I'm working with a homemade sandwich ELISA to detect the capsid protein of HIV-1, p24. We have been using this ELISA in the lab for several years with good results. Recently, we began to concentrate the virus by ultracentrifugation, recovering the pelleted viruses in PBS. But suddenly, we are getting overvalued measurements when measuring the p24 in the concentrated virus preparation. We are getting signals even to dilutions up to 10e-8 ! Then if you multiply the results by factor dilution you get values of millions of ng/ml, which of course makes no sense at all. One would expect to have similar values whichever dilution factor you use...no ?

The standard curve in perfect, no background in the blanks. Furthermore, a mock preparation without viruses give no signals to the same levels as the blanks...very puzzling for us

I've been looking on the web and found this website from Thermo where
They talk about spike-and-recovery and linearity-of-dilution. I believe this could be our problem, especially for linearity-of-dilution. Has anyone encounter the same problem and could explain me or direct me toward some resources. Is there any excellent recent book on ELISA that anyone could suggest me ?

ELISA conditions:
coating 5µg/ml, carbonate buffer pH 9.5
biotin-secondary antibody, dilution 1:2000 of stock 1µg/ml
HRP-streptavidin, dilution 1:20 000
Substrate, TMB-S from neogen (could this substrate be too sensitive for our assay ?)

Many thanks for helping me
Reply With Quote
  #2  
Old 03-30-2011, 05:27 PM
danfive's Avatar
Nobel Laureate
Points: 8,612, Level: 64 Points: 8,612, Level: 64 Points: 8,612, Level: 64
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Jul 2007
Location: Houston TX
Posts: 969
Thanks: 12
Thanked 184 Times in 154 Posts
Default Re: Linearity-of-dilution concept

I'm not sure I understand everything, you don't mention what your starting material/concentration is.

But you seem to say you have an unrealistic value for detected capsid protein.

Perhaps it is a false positive signal. Do you have a hook effect, greater dilutions (less sample) have higher signal than lesser dilutions (more sample)??

Have you optimized for Dilution Factor and even Dilution Buffer??
A presence of non-ionic detergent can help disaggregate any clumped protein, that would require a greater dilution factor if using buffer alone.
Reply With Quote
The Following User Says Thank You to danfive For This Useful Post:
admin (05-04-2011)
  #3  
Old 04-04-2011, 06:54 PM
Pipette Filler
Points: 124, Level: 2 Points: 124, Level: 2 Points: 124, Level: 2
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Mar 2011
Posts: 3
Thanks: 0
Thanked 2 Times in 2 Posts
Default Re: Linearity-of-dilution concept

Quote:
Originally Posted by danfive View Post
I'm not sure I understand everything, you don't mention what your starting material/concentration is.

But you seem to say you have an unrealistic value for detected capsid protein.

Perhaps it is a false positive signal. Do you have a hook effect, greater dilutions (less sample) have higher signal than lesser dilutions (more sample)??

Have you optimized for Dilution Factor and even Dilution Buffer??
A presence of non-ionic detergent can help disaggregate any clumped protein, that would require a greater dilution factor if using buffer alone.
Thanks for the reply danfive,

Our starting material is cell culture supernatant, containing the antigen in the order of hundreds of ng/ml.

No, we do not have a hook effect as you describe. The greater the dilution, the less signal we have. But, within the dynamic range of the standard curve, we get divergent results for the different dilutions chosen.

For exemple:
Let say we have a sample with 12 molecules of antigen. If we make a two fold dilution, we would expect to measure 6, then multiply by dilution factor 2 = 12. For a fourfold dilution, expect to measure 3, times dilution factor = 12 again logically.
With the same example of 12 molecules, in our hands we would get at dilution ½ a measurement of 8, so final concentration of 8 * 2 = 16 and at a dilution of 1/4 a measurement of 5, so final concentration of 5 *4 = 20. Our final calculated concentration increases as the dilution factor increases. But, no, it is not a hook effect.

We add our samples and standard curve on the plate, and then add 25µl of 2.5% Triton X-100 (final 0.5%). The purpose of this is to disrupt the lipididic envelope of the virus and release the p24 protein inside (our antigen). So yes we add a non-ionic detergent. I tried adding the detergent to the sample before making dilutions, but ended up with the same results.

We have never optimized for dilution factor nor dilution buffers. This did pique my curiosity and think it could help us. How do you achieve that ? We are currently using PBS 1X + 1%BSA both for diluting the standard and the samples.

Thanks for your help
Reply With Quote
The Following User Says Thank You to Sharpie For This Useful Post:
admin (05-04-2011)
  #4  
Old 04-06-2011, 07:25 PM
danfive's Avatar
Nobel Laureate
Points: 8,612, Level: 64 Points: 8,612, Level: 64 Points: 8,612, Level: 64
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Jul 2007
Location: Houston TX
Posts: 969
Thanks: 12
Thanked 184 Times in 154 Posts
Post Re: Linearity-of-dilution concept

Quote:
Originally Posted by Sharpie View Post
Thanks for the reply danfive,

Our starting material is cell culture supernatant, containing the antigen in the order of hundreds of ng/ml.

No, we do not have a hook effect as you describe. The greater the dilution, the less signal we have. But, within the dynamic range of the standard curve, we get divergent results for the different dilutions chosen.

For exemple:
.....
With the same example of 12 molecules, in our hands we would get at dilution ½ a measurement of 8, so final concentration of 8 * 2 = 16 and at a dilution of 1/4 a measurement of 5, so final concentration of 5 *4 = 20. Our final calculated concentration increases as the dilution factor increases. But, no, it is not a hook effect.
We add our samples and standard curve on the plate, and then add 25µl of 2.5% Triton X-100 (final 0.5%). The purpose of this is to disrupt the lipididic envelope of the virus and release the p24 protein inside (our antigen). So yes we add a non-ionic detergent. I tried adding the detergent to the sample before making dilutions, but ended up with the same results.

We have never optimized for dilution factor nor dilution buffers. This did pique my curiosity and think it could help us. How do you achieve that ? We are currently using PBS 1X + 1%BSA both for diluting the standard and the samples.

Thanks for your help
I see what you mean about the no hook effect.
Triton X-100 is great.

Do you believe that you are preparing a homogenous mixture before adding to plate? I ask because incongrous mixing during dilution can create weird effects.
You may consider comparing the following and see which works best.
1. Hand mixing via pipette (5-8 times)---limitation-not 100% homogenous mix accomplished.
2. Vortexing (medium, 5-10 sec)--limitation-some proteins will clump, change structure. Usually not recommended, but ideal for some v. small proteins/molecules.
3. Microtiter plate shaking (5-600 rpm 1.5-3min)--depending on the solution tends to create excellent homogenous mixture with tight CV's; limitation-none really, just need to optimize for time and rpms.
4. Inversion of tubes (10-12 times)--gentle, thorough mixing with larger # inversions--limitation-tough to standardize for inter-day runs etc.

As to optimizing for dilution factor and dilution buffer:
Dilution factor is for diluting past the hook in the "hook effect"--you may not have this need. but normally you would run the assay with samples set up as dilution series (DF = 2, 4, 8, 16 etc).

You may benefit from optimizing for dilution buffer/detergent type and concentration.

You use PBS 1X + 1%BSA.

So you could try PBS vs TBS (round 1, on rare occassion the buffer salt combination makes a big difference).

Assume PBS is the better buffer for you application.

Then try (round 2):
PBS 1% BSA
PBS 0.5% Triton X100 (BSA optional)
PBS 1.0% Triton X100 (BSA optional)

Here if the percentages don't show any difference, I'd keep the lower one 0.5% standard, and try different detergents:
(round 3)
PBS 0.5% Triton X100
PBS 0.5% Tween 20
PBS 0.5% Igepal CA 630
PBS 0.5% Brij 35

You would run sample (3-5 to keep it simple) concurrently in all the above solutions and compare them head to head, to see which underlying buffer solution/mix is better.

I'm sure some things will need clarification. Post any Q's when the time comes.

Last edited by danfive; 04-06-2011 at 07:30 PM.
Reply With Quote
The Following User Says Thank You to danfive For This Useful Post:
admin (05-04-2011)
  #5  
Old 04-20-2011, 11:34 PM
Volunteer
Points: 1,621, Level: 24 Points: 1,621, Level: 24 Points: 1,621, Level: 24
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Apr 2011
Posts: 36
Thanks: 0
Thanked 10 Times in 10 Posts
Default Re: Linearity-of-dilution concept

How did you get on with the detergents?

Sounds like there may be matrix effects, with the curve diluent having a different composition compared to the sample diluent. How pure is the concentrate? What other factors have been co-concentrated that might contribute to a matrix effect.

Would be interested to know
Reply With Quote
The Following User Says Thank You to Elisadeveloper For This Useful Post:
admin (05-04-2011)
  #6  
Old 04-28-2011, 03:10 PM
Pipette Filler
Points: 124, Level: 2 Points: 124, Level: 2 Points: 124, Level: 2
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Mar 2011
Posts: 3
Thanks: 0
Thanked 2 Times in 2 Posts
Default Re: Linearity-of-dilution concept

Hello,

I was in vacation for the couple last weeks.
Many thanks for your thorough explanations danfive. I haven't tried the different conditions that you have proposed yet, but I will in the very near future.

Quote:
How did you get on with the detergents?

Sounds like there may be matrix effects, with the curve diluent having a different composition compared to the sample diluent. How pure is the concentrate? What other factors have been co-concentrated that might contribute to a matrix effect.
Elisadeveloper, as mentionned I haven't tried yet, but I will let you know the results.
I agree there must be some matrix effect and I will evaluate with a standard diluent that better match the sample diluent.

HIV viruses are produced by transfecting the whole HIV genome (~9,5kb) of in 293T eukaryotic cells. The virus produced is released in the culture media (RPMI 1640 + 10% FBS). We harvest the culture media, spin down the few cells floating,then filter on 0,45µm to get rid of cellular debris. Finally we pellet the virus by ultracentrifugation and resuspend the pellet in either PBS or RPMP without serum to achieve a 10X concentration. So in our 10X concentrate, our sample, I assume we can have residual FBS, protein clumps, cellular microvesicles, other chemicals from the RPMI.

Difficult to say which of the components can cause the matrix effect.

Thanks
Reply With Quote
The Following User Says Thank You to Sharpie For This Useful Post:
admin (05-04-2011)
  #7  
Old 05-04-2011, 03:25 AM
Volunteer
Points: 1,621, Level: 24 Points: 1,621, Level: 24 Points: 1,621, Level: 24
Activity: 0% Activity: 0% Activity: 0%
 
Join Date: Apr 2011
Posts: 36
Thanks: 0
Thanked 10 Times in 10 Posts
Default Re: Linearity-of-dilution concept

preparation of the curve, or spike in experiments in the 'mock preparation' should have the same matrix issues as the sample, so you will end up backcalculating like with like.
Reply With Quote
The Following User Says Thank You to Elisadeveloper For This Useful Post:
admin (05-04-2011)
Reply

Tags
concept , linearityofdilution


Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are On
Pingbacks are On
Refbacks are On

Forum Jump

Similar Threads
Thread Thread Starter Forum Replies Last Post
Taq Dilution Bassaml7 PCR - Polymerase Chain Reaction Forum 7 09-06-2009 10:06 PM
Homéopathie : Dilution korsakovienne contre dilution Hahnemannienne Julien Arlandis Forum De Chimie 161 06-14-2008 11:36 AM
"A Basic Look at the Aether" Ralph Hertle Physics Forum 0 03-29-2008 06:44 PM
"A Basic Look at the Aether" Ralph Hertle Physics Forum 0 03-29-2008 06:38 PM


All times are GMT. The time now is 09:37 PM.


Powered by vBulletin® Version 3.8.4
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright 2005 - 2012 Molecular Station | All Rights Reserved
Page generated in 0.17757 seconds with 16 queries