| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
INVITED COMMENTARIES |
Renal Research Manchester Royal Infirmary Manchester, United Kingdom
e-mail: angela.summers{at}cmmc.nhs.uk
Animal models in the field of peritoneal dialysis (PD) have in the past received mixed reviews (1). Criticisms with respect to the exact relevance to the human situation have always tainted rodent and murine studies. When looking for specific molecular pathways, it is the "needle in a haystack" problem when we think of the dialysis patient. There are so many complicated processes to consider, such as uremia, dysregulation of the immune system, malnutrition, chronic exposure to dialysis fluids, recurrent infection, all of which could contribute to complications such as encapsulating peritoneal sclerosis (EPS), which is a rare condition. How can animal models hope to mimic the real life scenario?
Researchers have tried to replicate the human condition in animal models of EPS. The most commonly reported model is that of the chlorhexidine gluconate infusion (2). This was initially relevant when EPS was found to be attributable to disinfectants but its clinical relevance has waned in the light of EPS occurring in the absence of such solutions.
Much, however, has been learned from the chlorhexidine models. The chlorhexidine gluconate model yielded extensive information about molecules involved in the pathogenesis of EPS (or at least its animal equivalent). The take-home message from these experiments has been that transforming growth factor-beta (TGF-β1) has emerged as a key player (3,4).
In this issue of Peritoneal Dialysis International, Margetts et al. describe a new animal model of EPS (5). This model utilizes a novel methodology: instead of using chemicals or fluids to damage the peritoneum, they have fast tracked to using TGF-β1, which has been shown to be a driver in many fibrotic scenarios, including EPS. This novel approach, which has previously been described by this group in animal models of lung fibrosis, cuts straight to the chase and avoids administration of fluids into the peritoneum altogether. By using helper-dependent first-generation adenoviruses expressing TGF-β1, delivered to the peritoneum in mice, the model results in features similar to fulminant human EPS, which characteristically includes extensive adhesions and "cocooning" or encapsulation of the bowel, involving the visceral and parietal peritoneum.
Is it possible that a single molecule can indeed be exclusively responsible for the development of such a severe condition, as this new model suggests? Clearly, long-term delivery of TGF-β1 activates a cascade of other molecular pathways that contribute to the progression of disease. The upregulation of other cytokines, such as plasminogen activator inhibitor type-1 (PAI-1) and matrix metalloproteinase (MMP), as a result of prolonged expression of TGF-β1 is consistent with findings in other animal models of EPS (6,7). There is one discrepancy with this model compared to previous findings: the lack of vasculopathy; the levels of vascular endothelial growth factor (VEGF) were not measured (8). Maybe future experiments will clarify the role of VEGF, if there is a role to be found.
There has been much debate over the difference between simple sclerosis and fulminant EPS, with some arguing that the two are separate entities and have different pathological mechanisms (9). Personally, I have always preferred the hypothesis that the latter is a progression of the former. This new model supports that notion and presents good evidence that EPS is the continuum of simple sclerosis, with short-term expression of adenovirus TGF-β1 leading to histological changes similar to the so-called simple sclerosis described in humans. Interestingly, these changes regress once the adenovirus ceases to express the TGF-β1. However, prolonged expression of TGF-β1 by the helper adenovirus ultimately culminates in the abdominal cocoon. This new model also adds credence to the two-hit hypothesis described by Kawanishi (10), in that the first stimulus, being PD itself, may increase TGF-β1 production, as described in a recent article by Yao et al. (11). This in itself may not cause progressive damage and this initial process appears to regress in the absence of further stimulation. So, another factor has to kick in to prolong the expression of TGF-β1. We can only speculate what the second hits may be. This model suggests that it may be something that stimulates further TGF-β1, which could be infection. It is, however, possible that genetic variants in the genes for TGF-β1 may mean that some people are high producers; hence, these patients for some reason cannot turn off the fibrotic process (12). Margetts makes the comment that, in his experience, mouse strains vary in their fibrotic response to the TGF-β1-producing adenovirus. Another explanation may be that genetic variations may occur in growth factors that drive TGF-β1 production. Certainly, it seems that once TGF-β1 production has been consistently high for a long period of time then the cocoon may be inevitable, and hence cessation of PD or renal transplantation may not stop the progression, implying that, once the progression is underway, PD fluids can no longer be blamed for the continuation of the whole process. The questions to ask next are, How much TGF-β1 is critical and over what time frame? Can the whole process be arrested or even reversed? And, if this whole process can be attributable to one molecule, can we not just block it with an antibody? Certainly, this is an attractive option and not unprecedented as anti-tumor necrosis factor antibodies have proven successful in the treatment of rheumatoid disease (13). Blockade of VEGF with antibodies administered in eye drops has been successful in preventing corneal neovascularization in patients with diabetic retinopathy (14).
Clearly, this new model of peritoneal sclerosis will provide scope for interventional studies. It will be possible to investigate factors that may arrest or even reverse the ongoing damage. Antifibrotic factors have indeed been demonstrated to be effective in the chlorhexidine model. The antifibrotic and anti-inflammatory pirfenidone reduces matrix deposition in the chlorhexidine gluconate model and has been proposed as a potential therapeutic option for the prevention of EPS (4). ONO-4817, an inhibitor of MMP-2, significantly reduces TGF-β1, smooth muscle actin, and collagen deposition (7). A range of experiments is required to elucidate which antifibrotic reagent is effective and then to investigate optimal dosage and timescales. The authors note that several investigations, such as peritoneal function, were not possible once the bowel had cocooned but there is scope in this model to progressively monitor such parameters before encapsulation occurs.
The role of infection needs further investigation as, in the human disease, EPS is often noted to occur following peritonitis. Also, histological markers of inflammation have been demonstrated with patients often presenting clinically with raised C-reactive protein. We know that infection can increase TGF-β1 production. In this new model, histological and cellular features of infection were not demonstrated. Consistently elevated levels of interleukin-6, however, were reported, but the exact role of this cytokine in this model is difficult to define as levels do not change significantly with severity of disease. It is possible that the superimposition of an infection may accelerate the process. This will require further investigation in this model.
So the questions remain, Can this model help us to understand the clinical picture of EPS in humans? and Does this really represent the pathogenic process in a human with end-stage renal disease on long-term PD?
Unfortunately, as yet we have no handle on whether TGF-β1 levels are increased in the peritoneum of PD patients that develop EPS. In this mouse model, however, levels of TGF-β1 in peritoneal fluid were not shown to be significantly elevated following long-term adenovirus expression of TGF, which questions whether TGF-β1 level in effluent is indeed reflective of tissue level. Longitudinal measurements of TGF-β1 levels in PD effluent of patients that subsequently develop EPS may provide insight. Hopefully, studies such as the Global Fluid Study that have been prospectively collecting PD fluids from large numbers of patients may soon reveal some clues as to whether we can indeed translate from mice to men.
DISCLOSURE
The author has no conflict of interest.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  A. M. Summers BIOMARKERS OF EPS: CAN WE GO "BACK TO THE FUTURE"? Perit. Dial. Int., March 1, 2010; 30(2): 158 - 160. [Full Text] [PDF]
|
|
|
|
|
|
S. L.S. Fan SHOULD WE USE BIOCOMPATIBLE PD SOLUTIONS FOR ALL PATIENTS? Perit. Dial. Int., November 1, 2009; 29(6): 630 - 633. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
