There are several ways to induce endothelial dysfunction in animals, both in vivo and in vitro.
I have recently started to induce ED with nicotine in male Wistar rats. Literature data show that ED could by induced with nicotine in dose 2mg/kg/day. To dissolve nicotine use 0.9% NaCl (5 mg of nicotine in 1 ml 0.9% NaCl). Nicotine is administered each day, internationally, for a period of 4 weeks. If you can I would also suggest you to determine some markers of ED (ADMA, ICAM-1, vWF...) from rat plasma or serum at the end of this period to be sure that ED is induced.
For endothelial denudation you can use a stainless steel wire which is small in diameter so that it can go thru lumen of blood vessel, and then you roll the wire by gently rubbing the blood vessel to surface. You do this for 10 times (back and forth). Be careful because you can completely destroy your blood vessel. Do not push to hard!!! By doing this you will mechanically remove endothelium.
If you have any additional question be free to contact me.
I support the denudation protocol submitted by Marko.
The endothelial dysfunction question is however is more complicated.
First, you need to decide if this ED will be your model or your "positive" control.
Second, do you need an acute ED (induced in minutes) or chronic model is Ok?
Finally, you review the feasibility of any potential models.
You can induce ED by acute oxidative stress using menadione (20 uM, 15 min) as a "positive" control. Use LPS as a sepsis model ex vivo (1-4 hours). Use TNFa (10 ng/ml, 30 min to 24-hours) as an inflammatory model. You also can block eNOS by specific inhibitors L-NIO since inhibition of NO production contributes to ED.
Note, you can use DMEM for aorta tissue culture for 24-48 hours.
For endothelium denudation, I recommend you gently rub the organ with cotton strip. For reference, I attach a file and quote another one for you:
Caricati-Neto, A., Pupo, A.S., Wanderley, A.G., Nuñez-Vergara, M., Koh, I.H., Jurkiewicz, N., Jurkiewicz, A., 1995. Role of the epithelium in the release of contractile agents from the rat vas deferens by clonidine. Annals of the New York Academy of Sciences 763, 463–469.
How to induce endothelial dysfunction is a tricky question because it all depends on what you are trying to mimic. As Sergey alluded to above, you can inhibit eNOS (which I guess would mimic the accumulation of ADMA), but you can also deplete tetrahydrobiopterin (to uncouple eNOS), or go down the route of inducing ROS, which a number of stimuli would do, as others previously said. I don't think that endothelial dysfunction is something you just simply induce, because it can take many forms. For example, one cardiovascular risk factor may activate Nox2 to induce ROS, another maybe mitochondria, or xanthine oxidase. Also, it's severity would determine whether eNOS becomes uncoupled, or maybe whether there is cross talk between sources of ROS (Sergey Dikalov I believe is an expert on this latter point).
What is the specific purpose of you inducing endothelial dysfunction?