Hormones from endocrine pancreas are quite susceptible to proteolytic degradation over time. Lower temperature will offer better protection for hormones, hence storage of plasma in freezers!

Multiple freeze-thaw cycles also cause loss of hormone activity (possibly the amount), as judged by biological function or in some cases by antibody binding (such as ELISA). In case of larger amounts for storage, it is better to freeze aliquots that are sufficient for one or a set of one measurements. If a repeat measurement is necessary, one will use another aliquot that has not been subjected to more than one cycle of freeze-thaw.

A safer method is to add a 1/10th volume of 10x protease inhibitor mixtures; most common and simple one is EDTA+Trasylol (Aprotinin). 

Plasma samples for insulin assay can be stored at 4 degree C for a day or two without substantial losses (

Hello Bakya,

I am very interested also in the stability of insulin and insulin analogs. Here is my advice to you:

Store your insulin stock solution as per manufacturers recommendation. Generally, insulin preparations are intended to be kept at 2-8 degrees C but can tolerate some time at higher temps. It is assumed by manufacturers that the average diabetic patient may not have access to a refrigerator during the workday and may carry the insulin around at room temp.

It is safest best practice to store all reagents in the best way possible. That being said, my experience with Humulin R is that it will not lose potency (as measured by Insulin RIA) for more than two years past the expiration date if it is stored at 4 degrees C. I cannot give any advice on the analogs such as glargine or lispro other than to state what the manufacturer recommends.

As for samples containing insulin, provided that there is not microbial contamination, the insulin can remain stable at room temperature in cell culture supernates for a month in our hands (a curious coworker had me test samples once a week that had been left on his bench, but I did not know that I was testing the exact same samples) Key point is that there cannot be microbial contamination or all bets are off. We always freeze our serum insulin samples at -20. I have gone back and tested human samples from 10 years ago and got back the same value as I did the first time around.

Your assay kit contains preservatives which is one reason why the incubation can be carried out at room temperature without worry. Insulin is fairly stable, but will be affected by repeated freeze-thaw cycles. Also, if you have a hemolyzed sample, it is likely that you will have a lower insulin measurement for that sample than you would predict in a non hemolyzed sample. Always note if you have an orange or red sample so that if the results are low you will know why.

While I can't speak specifically about mouse serum, we did run a series of experiments to test the effect of a freezer failure on our canine samples.  With just one thaw then immediate refreeze we lost 10-15% of the insulin in the sample.  The longer the sample was stored at room temp, the greater the loss in insulin.  By three days at room temperature approximately 50% of the insulin detected, and this dropped only slightly more when kept at room temp for 7 days.  

I agree with Elizabeth's answer - it will likely depend on the form of insulin, and the preservative used in the storage tube - it is possible that humulin is more stable than our canine insulin.  Also, just FYI, our samples were stored in heparininzed tubes, and contained EDTA.  I suspect the human insulin vials that can be kept at room temperature so long may contain more preservatives than ours, but I don't know for sure.  

Hi Cathryn,

Our human samples were drawn in red top tubes, FYI. I am interested in the loss that you see with one freeze thaw. We typically did not see any change even with 2-3 freeze thaw cycles with the antibody we were using (an in house antibody). I do not usually see any decrease in insulin with one freeze thaw cycle on a mouse sample, but then, one does not have much opportunity to keep freezing thawing and testing a tiny mouse sample. Usually two chances is all I have! A couple of questions:

What kind of insulin levels did your samples have? That is, were they samples with a lot of insulin, basal insulin, or did it not seem to matter?

I imagine that there can be an effect of slight degradation on the ability of your antibody to pick up the insulin - might I ask whose kit you were using (or were you using an in-house antibody?)

Also, I am confused - you say you stored the samples in EDTA tubes, or you mean that you drew the samples in EDTA tubes and then subsequently stored the plasma?

We had a range of insulin concentrations up to 300pmol/L, which were made by spiking fresh canine plasma with the standards from the kit, I believe.  We had talked about repeating this with samples from an IVGTT for endogenous hyperinsulinemia, but I don't believe we ever carried that out.  We saw greater degradation with basal concentrations, but we were already pretty close to the lower limit of the assay.  

We were using a Linco kit adapted in-house for dog insulin - the kits were used appropriately, so our results are probably due to a degradation of insulin, or a change in the matrix - maybe we'd have different results if we had used serum, but we were trying to test whether our thawed samples were still valid.  

And you're right - the blood was drawn into hep-EDTA, then the plasma stored.  

Thanks Cathryn, Good to know. I suspected that you would have more trouble with degradation at the lower end of your concentration range.

Thanks for the info!

EO

Hormones from endocrine pancreas are quite susceptible to proteolytic degradation over time. Lower temperature will offer better protection for hormones, hence storage of plasma in freezers!

Multiple freeze-thaw cycles also cause loss of hormone activity (possibly the amount), as judged by biological function or in some cases by antibody binding (such as ELISA). In case of larger amounts for storage, it is better to freeze aliquots that are sufficient for one or a set of one measurements. If a repeat measurement is necessary, one will use another aliquot that has not been subjected to more than one cycle of freeze-thaw.

A safer method is to add a 1/10th volume of 10x protease inhibitor mixtures; most common and simple one is EDTA+Trasylol (Aprotinin). 

Plasma samples for insulin assay can be stored at 4 degree C for a day or two without substantial losses (

Good answer Tausif,

I would state more strongly that is a necessity to add proteolytic inhibitors if one wishes to measure glucagon (or GLP-1 and/or GIP). No almost about it.

I agree that in general it is true also that best practices must be followed if one wishes to have reproducible results, therefore one must consider carefully if it is worth running samples that you think have been compromised.

Cheers!

You are correct, Elizabeth.

Addition of protease inhibitors is necessary for reliable and consistently good results.

On occasions we have managed without them when downstream processing of samples was such that proteases couldn't survive, for example acid acetone extractions.

[On occasions we all find ourselves wondering about sub-optimally processed samples (or less than perfectly designed experiments) that if some useful information can be extracted from them. Over the years, I have learned to be fairly ruthless in discarding things that will at the best only produce numbers of questionable value. I prefer to start over, design and conduct experiments as cleanly as possible.]

Thank u all for your valuable answers and suggestions...

Adding glycerol in the samples may be beneficial if freezing and thawing is required in case insulin was to be measured by ELISA. The dilution factor has to be considered.