That's an interesting fat emulsion, to say the least. The Tween and PEG will probably make the intestine quite leaky. Which is actually not a bad thing, since it may cause endotoxemia which would in itself decrease insulin sensitivity ("metabolic endotoxemia"). In my opinion, the reported IR after only 10 or 15 days is perhaps more due to endotoxemia due to a leaky gut because of Tween and PEG than to adiposity due to the fat. In that case, using Intralipid instead of the emulsion may not recapitulate the effects since Intralipid may not cause endotoxemia.

I would also try using a high-fat diet instead, as suggested above, and just wait a couple of weeks. I would recommend a diet with 40% KCal from fat, based on milk fat instead of sunflower or soybean oil. You could take small blood samples from the tail vein now and then to measure fasting glucose levels, to estimate occurence of IR.

By the way, if speed is the main reason for the model, and if genetic defects in the leptin pathway are of no concern, could you not take Zucker rats?

I don´t know the model but in the abstract the state that they have given it ig = intra gastric, meaning that the have been given it into the stomach. I would never use a model in which you give the animals 20 % Tween 80 and 30 % propylenglycol to the animals (stated in the paper). The so called "control" is that the animals have been given pure water ig. Thus, the effect might just as well (actually more likely) caused by the detergent exposure, rather than the fat. My advice is that you shall find another model. There are plenty of better models around.

I have not used lipid emulsiion diet but a carcinogenic agent.

If you want to control the amount administrated to the animals I would advise to use intra-gastric instillationm, as the authors write.

I agree with Lars Hellgren that it is not a good idea to instillate 20% Tween 80 in he stomach. At least, if you want to use this model in a valid way, one should use vehicle in the control grtoup (that is, everything unless lipids including detergent...) and hope that it does not kill or harm the animals so much that it invalidates your study.

Indeed the description in that article is incomplete. I would recommend that you simply try to retrieve the missing infomation by contacting the authors using the e-mail they provided. Mode of administration may be as important as the volume administered. Intragastric intallation allows better control of your delivered volume, but may not be acceptable by your animal experiment institutional review board, depending on their overall policy. Of note is that some early changes in insulin and inflammatory changes can be detected in rodents very early on during high fat diet, which is much simpler to manage. Depending on your experimental hypothesis and the response you get from the authors you might look into the latter option as well.

The authors used Tween 80 as a surfactant in the emulsion so that the fat goes along with the water (hence emulsion). They appear to have done so on top of the regular food consumption as the control group was only given water while the treated group given the emusion. The substance is probably not a problem especially at small amounts. You can estimate the volume they gave to each rat. The body fat gain in the treated group is 7 g greater than the control group for 10 days, so 0.7 g/day (Table 1). Based on metabolizable energy content of 90% for dietary fat, this means each rat received no more than 1 g of lard, i.e.

A constant volume of 100 mL fat emulsion containing 20 g lard, 1 g thyreostat, 5 g cholesterol, 1 g sodium glutamate, g sucrose and 5 g saccharose, 20 mL Tween 80, 30 mL propylene glycol was prepared by adding distilled water and stored at 4°C. The rat emulsion was given to the test group instead of the drinking water for 10 days (non-specified quantity, i. e., the animal is free to drink once need to drink).

Dear colleugues, thank you very much for you answers. Based on the works pulished and your answers we have decided to give animals 10 ml/day of fat emulsion (administration of 5ml ig twice a day because of the rat mass of 220 g).

Due to Tween 80 LD50 for rats being 34.5 g/kg (oral) and a large amount of data indicating the use of the model I suppose that 2ml of Tween 80 per day is not supposed to be toxic.

I have contacted the authors, but I am still waiting for their response.

you can try to give your rats high fat diet (40%, no more) from weaning to adult life (about 10 weeks). It depend also of the strain of rats you used. Th Wistar is being fat very quickly, and also use males. The female placed on food choice (P, fat, CHO) eat more CHO than fat.

If you are worried about the surfactant (Tween), why not use commercial soybean lipid emulsion 20% (very low concentrations of phospholipids as the surfactant).

As Ronald annd Zenkui say, try to use Intralipid emulsión of intravenously or intraperitoneal via. Do not use Twueen because it is an inhitbitor of Lipoprotein lipase and hepatic lipase, producig a huge variartion in lipoproteins profil. Use male rats to exclude in your model the influence of hormonal variation due to estrogen status and diet.

All the aforementioned answers are interesting; however, I think it is important to understand that give a rat 5ml of fat emulsion is difficult. Often you have to administer it slowly with a long canule in order to ascertain the amount of liquid. I think that Dr Wikinski suggestion is valid, do not use Tween in your protocol.

The composition is given in the Materials and Methods section: "a constant volume of 100 mL fat emulsion containing 20 g lard, 1 g thyreostat, 5 g cholesterol, 1 g sodium glutamate, 5 g sucrose and 5 g saccharose, 20 mL Tween 80, 30 mL propylene glycol was prepared by adding distilled water and stored at 4°C." I never used such a system to induce insulin resistance. It seems very interesting since insulin resistance takes place in only 10 days in young rats (200g), which is spectacular (although the difference in GIR is not so impressive).

Considering that insulin resistance in man is associated with both chronic overnutrition and with the postprandial response to carbohydrate intake - the question raises some concern about the general comparability of metabolic/nutritional studies in animals vs. those in man.

An oral gavage or pure oil (vegetable oil, etc) works well, but small volumes are required to prevent vomiting. Pilot this procedure to optimize the composition of fatty acids you require (type of oil) and volume that can be tolerated by rat. 1ml often works well in adult rats

The model is attractive for its short duration needed to create an insulin resistant state (10 days). By orally feeding high fat diets, you will need at least 2 months to get ~20% body wt excess over the control as we routinely do (Sprague Dawley rats; there may be species differences). So the gavage model saves time and worths a try provided that it will not introduce unwanted side effects. Another way for rapid introduction of insulin resistance is i.v. infusion of lipids such as Intralipid that has been done in humans where several hours of infusion will bring about IR. I would guess it may be even faster in rats.

I would be a little concerned for all of the above reason and others... Also, this procedure has received only 3 citations since 2005...See Shulman for intralipid infusion.

Dear Dr. Guo, the short period of 10 (or 15) days needed for IR modeling was the one of the main advantages of the model. However, according to your and the others advice I suppose to replace Tween with Intralipid. In accordance with this I would also like to ask if it is better to introduce Intralipid enterally or parenterally?

That's an interesting fat emulsion, to say the least. The Tween and PEG will probably make the intestine quite leaky. Which is actually not a bad thing, since it may cause endotoxemia which would in itself decrease insulin sensitivity ("metabolic endotoxemia"). In my opinion, the reported IR after only 10 or 15 days is perhaps more due to endotoxemia due to a leaky gut because of Tween and PEG than to adiposity due to the fat. In that case, using Intralipid instead of the emulsion may not recapitulate the effects since Intralipid may not cause endotoxemia.

I would also try using a high-fat diet instead, as suggested above, and just wait a couple of weeks. I would recommend a diet with 40% KCal from fat, based on milk fat instead of sunflower or soybean oil. You could take small blood samples from the tail vein now and then to measure fasting glucose levels, to estimate occurence of IR.

By the way, if speed is the main reason for the model, and if genetic defects in the leptin pathway are of no concern, could you not take Zucker rats?

Dear Dr. Eckhardt, the increase in gut permeability as well as alteration of gut microbiome is also the one of the parameters to be estimated in our planning research. The estimation of LPS as well as characterization of gut flora will show whether IR in the model at least partly mediated by metabolic endotoxemia.

The intralipid model has problems as well. Heparin is included, allegedly as a lipoprotein lipase inhibitor but it obviously affects hemostasis and other factors. I agree that increased inflammation due to effects in the portal circulation are probably also taking place. This is an extremely complex intervention and interpreting all its effects through the lens of insulin resistance is problematic.

Hi Dr. Tinkov, I assume your main goal is to study the effects of IR on something. So you may want to minimize the confounders as much as possible so that IR is the only or main variable modified. Infusing Intralipid i.v. is an acute (hours) way to manipulate the lipid metabolism and I'd assume lots of other parameters are also disturbed in addition to IR. By comparison, by feeding orally you give the animal more time to gradually adapt to and manage the exogenous lipids over a longer period of time so that the confounders/collateral effects are less of an issue and thus you get a less-confounded IR. I think you would create a nice model by modifying the paper you originally cited, i.e. to replace the 20% of lard emusion with Tween in it with Intralipid. If you choose the i.v. infusion paradigm, however, you could either use heparin, in which case will actually infuse fatty acids into the circulation (the lipase stripped off the endothelial sufface by heparin hydrolyzes triglycerides to fatty acids), or without heparin but Intralipid alone in which case you will infuse triglycerides and therefore the system will handle it as such instead of fatty acids. So you can decide which way is more suited for your study goal.

I have received the authors response on my question regarding the dose of fat emulsion given tp experimental animals. The authors say that the dosage was 1 ml of fat emulsion per 100 g of rat mass.

I would like to thank all of the colleagues for such an interesting discussion on the model.

You might also consider diets and different fat emulsions for your rat model. I have included links to two studies that used different diets and different fats to show effects on insulin resistance and insulin secretion. We used SD rats and stayed away from TWEEN. Changes in insulin resistance and insulin secretion were determined by clamp studies.

http://diabetes.diabetesjournals.org/content/50/1/123.long

http://diabetes.diabetesjournals.org/content/51/6/1825.long

Hi Alexey,

Just for your information, I attached the link for our fat-free total parenteral nutrition-induced IR model in SD rats.

Article Soybean fat supplementation controls insulin resistance caus...