You could try with Chronic Umpredictable Stress (CUS). Here are a few examples:

. 

http://www.nature.com/npp/journal/v33/n2/full/1301410a.html

http://journals.plos.org/plosone/article?id=10.1371/journal.pone.0063302

http://www.jneurosci.org/content/26/14/3813.full

~ Thank you, your answer lead me to another question. The main differences of CUS and CMS are on it stressors severity. Thus, does the term 'mild' in CMS referring to decrease stressors severity compare to CUS? I was asked on the definition of the word 'mild' in CMS during my animal ethic application.

I think that the stressors are mild in both protocols. The main advantage of CUS over CMS is that the animals don´t develop tolerance to the stressors. 

You could check out the Wistar-Kyoto Rat (WKY) from Charles River Laboratories. They are hyper-responsive to stressors and tend to exhibit a depressive phenotype, both behaviorally and neurochemically. Check out some of the papers in this search: http://www.ncbi.nlm.nih.gov/pubmed/?term=wistar+kyoto+rat+depression

Learned helplessness is a good model for depression in rats.

Hi Qamarul

In my experience, in both mouse and zebrafish (the second attached paper with CUS from my lab) CUS worked nicely. I have experience of both the paradigms (CUS and CUMS) and can say that CUMS takes longer duration to give expected behaviour phenotype. BTW, you did not mention how many days you exposed for those stressors? I found in mice 6-7 days of CUS (longer episodes for shorter duration) induced depression in only females.  Pl. see our paper published in Biological Psychiatry. 2009;65(10):874-80 in your spare where as CUMS paradigm (shorter & milder episodes for 21days) could induce depression in both the sexes (unpublished lab data). Hope it helps to plan further rat experiments.

all the best to your work

In my opinion CUS has advantage over CM.  It is very difficult to fix a specific criterion for CMS in animal experiments.  The duration of stress in CUS also affects the outcome.  But 6-10 days of stress is sufficient to induce behavioural and biological changes but then ethical issues will have to be involve. There are two models of depression available that are reliable  and valid - learned helplessness in a jumping box and behavioural despair in swim test. 

Thank you all for the answer.

Although I kinda have guess the problem with my experiment. Apparently my control rats are more 'depressed' than the stress model rats.

There are many reasons for it some that I can think of are :- control rats holding area was in proximity of the stress rats holding area and the stress procedure room....and I didn't played with the control rats much..these are I think were rookies mistake on my behalf.

Anyway, I'm trying to establish a suitable model for my department to copy also. I will definitely following some of the advise and suggestion given above.

We use learned helplessness as one of the main models for stress/depression in our lab and get very consistent and robust results. In addition to looking at escape deficits we use other behavioral measures, like decreased sniffing of urine cues from a rat of the opposite gender as a measure of sexual-related anhedonic-like behavior.

Qamarul,

Manipulation (handling, cleaning cages), lighting conditions, environmental temperature, should be the same for control and treated rats. Avoid noise! check the pdf

Light- deprived rats is a non-stressful and genetically intact model of depression (Gonzalez and Aston-Jones 2008). Light deprivation is an inexpensive and easy method to produce anatomical, physiological and behavioral depression-related alterations considered as clinical benchmarks of depression, in diurnal and nocturnal rodents.

If you want to induce an acute and intense stress in your animals...I would recommend IMMOBILIZATION for 1 h (check on the pdf )