In our laboratory, for islet isolation, the modified method of Lacy and Kostianovsky [Lacy PE, Kostianovsky M. Method for the isolation of intact islets of Langerhans from the rat pancreas. Diabetes. 1967; 16(1):35-9] with slight further modification was used.

The rats were anesthetized (60 mg/kg sodium pentobarbital i.p), laparotomized, and sacrificed by heart incision. The pancreas was inflated through the bile duct with an injection of 10 ml ice-cold Hanks’ balanced salt solution (HBSS) containing 0.5 mg/mL of collagenase P, then removed, minced with scissors, and digested for 15–17 min at 37°C. Digestion was terminated by adding 30 ml ice-cold HBSS, and the tube was shaken for 1 min. The suspension was filtered through a 500 μm plastic mesh to discard any undigested tissue. After three washes with cold HBSS, islets were hand-picked under a stereomicroscope. The islets were used fresh for studies related to insulin secretion or stored frozen at -80°C until further use.

Do you want to count in the intact pancreas or after islet isolation?

I used a protocol very similar to the one that Fahimeh described. We also injected the collagenase through the bile duct to inflate the pancreas. The counting  of the pancreatic islets was performed manually  under a stereomicroscope, but is  time-consuming

Dear Josh wilhelm, I want to count the sum of islet after hematoxyillin-eosing staining so i did not do isolation or anything, but the organ has been extracted and prepared for hematoxylin-eosin staining.

ear Colleague,

With our pathologists we have used the same method as Khoirotin Nisak. To have an good probability of objectivity we havecount in 5 slices (to have a basic control count > 20 islets) obtained in a determined part of the pancreas (head or/and tail but always the same part) of measurable dimensions (light field of a amicroscope with the same Ocular/objective valuesto have, for instance G x5 or 10) the number and the size of the islets (for the size, there are grids either in the objective or placed on the preparation covering glass). The results are to be compared between intact mice and STZ mice of the same age and managed in the same conditions. You need patience, but the results are statistically valuable,  correct enough and exclude the loss of islets during the manipulations such as collagenase digestion, collection of islets, centrifugation and so on.

Good luck,

V.Coulic

Khoirotin,

To perform semiquantitative measurements of remaining islets (beta cells) after streptozotocin treatment, there are a few ways, but none of them are very simple or very accurate.

If you are willing to use a destructive method where entire pancreas can be allocated for insulin measurements, total insulin content can be used. This is done by removal of pancreatic tissue followed by acid-acetone extraction, recovering the insulin (acid soluble peptides remain in liquid phase), and measuring the contents.

As you can imagine, this method is not accurate because after STZ treatment, the remaining beta cells try to compensate for the need of insulin by secreting more insulin and therefore their remaining insulin contents can be far below the levels of normal islets.

You could include insulin mRNA quantization, which would also be incorrect, but in this case, the error will be in the opposite direction. In order to compensate for lack of proper amount of insulin, the remaining beta cells will overproduce insulin and the increase compared to normal islets can be 2.5-4.5 fold.

If you can only work with fixed tissue sections, you may have to perform immunocytochemical detection of beta cells and measure area of beta cells. The more islets you can score per section, the better for getting a more useful quantitative number. However, the caveat of degranulation of beta cells still applies, which makes for poor quality staining when using anti-insulin antibody.

There are a couple of alternatives to improve the quality of data - you can combine staining for alpha cells (glucagon) with insulin staining. In rodents, alpha cells are mostly on the outer edge of islets, which conveniently delineates the core where the bulk of beta cells reside. Keep in mind that even after a partial destruction of beta cells by STZ, depending on how many days post treatment beta cell detection is done, the islet size will be reduced - hence the necessity to measure the area.

The other alternative to score beta cells in sections is to stain for Glut-2 (this is a membrane protein), and with some experience, it would find beta cells that degranulated and difficult too stain for insulin.

As I mentioned in the beginning, relatively meaningful measurements are not simple and require effort. One has to be mindful of inherent limitations and pitfalls of any method chosen for a task.

If your STZ treatment is designed for the total destruction of beta cells, this simplifies things somewhat - glucagon stain will show very small sized islets with very few unstained cells at the core and insulin staining will be negative.

Best, 

thank you so much for the information, i really appreciate it.

dear professor very coulic, could you please share the paper or the link of method that you have explained about islet count using HE staining. i really appreciate it. thank you.

Hi!

This is also a cool brochure to study about the topic. https://iidp.coh.org/docs/CountingBrochure.pdf

I personally think that (for human islet counting) the only really reliable counting method includes automated or semi-automated computer based counting. In our islet lab, we use the Saisam Software from Microvision (http://www.microvision.fr/en/softwares/life-science/morphometry/)

Regards,

Patrick