Human intestinal ischemia and reperfusion

The intestinal mucosa is responsible for absorption of nutrients from the lumen
and for separation of luminal content (external environment) from the host
(internal environment). Disruption of this delicate balance at the
mucosal-luminal interface occurs during intestinal ischemia-reperfusion (IR), a
frequently observed phenomenon of normal physiology. Furthermore, intestinal IR
is considered as common pathway in several pathologies, where temporary flow
reduction results from vascular disease (thrombosis, embolism), surgical states
(aortic surgery, cardiopulmonary bypass) and other pathologies, including
necrotizing enterocolitis, pancreatitis and shock. Moreover, splanchnic
hypoperfusion is seen as a phenomenon involved in the etiology or perpetuation
of inflammatory bowel disease (IBD). Intestinal IR results in local tissue
damage with barrier function loss facilitating bacterial translocation,
triggering systemic inflammation and multiple organ failure (MOF). Despite high
morbidity and mortality rates associated with intestinal IR, no effective
preventive or therapeutic strategies exist.
In search for new interventional targets, it is crucial to comprehend the
underlying mechanisms associated with human intestinal IR.
For a long time only animal studies were reported on intestinal IR. This is
probably one of the main reasons that no effective clinical treatment for
intestinal IR was developed and the associated high morbidity and mortality
rates are unchanged the last 70 years. Therefore, a new human model enabling
the study of sequelae of jejunal IR was developed by our workgroup. This
allowed a better understanding of several physical (mucus-layer, epithelial
lining) and immunological (Paneth cells) barriers, preventing invasion of
foreign or endogenous threats from intestinal lumen to sterile interior
environment after 30 minutes ischemia. Prolonged (>45 minutes) ischemia
followed by reperfusion leads to irreversible damage and increasingly evident
endoplasmatic reticulum stress in Paneth cells, accompanied by apoptosis,
resulting in bacterial translocation and systemic inflammation. Furthermore,
using a newly developed human experimental colon-IR model, colonic epithelium
was shown more resistant to IR-induced damage than jejunum, although the high
colonic microbiota density is potentially more toxic than small intestine
intraluminal content. This may be explained by differences in mucus
organisation. Another explanation is the detrimental activity of intestinal
wall penetrating digestive enzymes present in the proximal small intestine.
Recent rat studies undergoing intestinal IR indicate that pancreatic enzymes
play a major role in acute inflammatory processes resulting from intestinal
ischemia, hemorrhagic and endotoxic shock. Serine proteases (e.g. trypsin,
chymotrypsin, elastase) are important pancreatic enzymes, which are stored in
secretory membrane-bound vesicles as zymogens. The trypsin zymogen is activated
by duodenal brushborder enterokinases, while activated trypsin activates more
trypsinogen and chymotrypsinogen. To avoid uncontrolled proteases activity,
enzymes-inhibitors are present in the circulation. Under normal physiological
conditions, proteases are compartmentalized in the intestinal lumen by the
mucosal barrier (mucus-layer and epithelial lining). The pathophysiological
role of proteases was discovered by adding protease-inhibitors intraluminally
in rats, showing abrogation of the usual sequelae of intestinal IR,
inflammation and shock. These results gave rise to the *autodigestion*
hypothesis suggesting that powerful proteases leak across the intestinal
mucosal barrier, initiating self-digestion of the intestinal wall and leading
to systemic inflammatory response (SIRS) with MOF. Protease activity in the
intestinal wall was accompanied by mucus disruption, tight-junction loss and
epithelial cell disruption after intestinal IR in animals. Recently, a report
showed that intraluminally administered protease-inhibitors decreased shock and
sepsis in a patient with Fournier*s gangrene.
Proteases, particularly serine proteases, also act as molecules that are able
to send specific signals to cells involved in intestinal inflammatory responses
through the activation of G protein-coupled protease-activated receptor-2
(PAR-2). Proteases cleave PAR at specific sites, unmasking a new N-terminal
sequence, acting as tethered ligand, which binds to the receptor to initiate
multiple signalling cascades. PAR-2 is present on epithelial and endothelial
cells and various types of immune cells with high expression in the
gastrointestinal and respiratory tracts. PAR-2 is activated by trypsin, mast
cell tryptase and coagulation factors (VIIa, Xa) and modulates several
gastrointestinal functions, including motility and secretion. Recent studies
using PAR-2 knockout-mice and specific PAR-2 agonists showed PAR mediated
neutrophil recruitment in mesenteric vessels, increased expression of
endothelial adhesion molecules, delayed gastrointestinal transit, and increased
histological damage. PAR-2 is reported to interact at signalling level with
Toll Like Receptor-4, resulting in enhanced NF-kappa B-mediated inflammatory
response, establishing a novel paradigm of receptor cooperativity. Furthermore,
two distinct pools of PAR-2 are present in intestinal epithelial cells: an
apical pool accessible from the lumen, activated by trypsin, and a basolateral
pool accessible from the interstitium and blood, activated by mast cell
tryptase or proteases released by recruited leukocytes. A recent study using
cell lines and mouse ileum suggests that the outcome of PAR-2 activation is
dependent on the specific receptor pool that is activated, because separate
signalling pathways are triggered. Epithelial PAR-2 activation directly affects
cytoskeleton contraction by triggering myosin light chain with subsequent
tight-junction permeability changes, probably via the basolateral PAR-2. Rodent
studies suggest an important role for PAR-2 in inflammation following
intestinal IR. PAR-2 mRNA and protein expression in intestinal mucosa is
upregulated after IR and inhibition of PAR-activating proteases is beneficial,
diminishing post-ischemic intestinal inflammation, such as myeloperoxidase
activity and chemokine and adhesion molecule expression. In line, IBD patients
have increased levels of PAR-2 activating proteases in lumen and colonic
tissue. A human PAR-2 antagonist is not available. However, soy extracts,
characterized by presence of serine protease-inhibitors, have recently
attracted attention because of their anti-inflammatory, PAR-2-mediated
properties in IBD animal models.

This study aims at (further) revealing the pathophysiology of intestinal IR in
man, with a specific interest for the role of proteases and protease-activated
receptor-2 (PAR-2), cellular and inflammatory changes, barrier function and
intestinal permeability, microscopic mucosal changes, gene expression patterns
and identification of targets for diagnostic, preventive and therapeutic
strategies.

Interventional study

During the procedure, a small intestinal segment will be isolated and
selectively exposed to 30, 45 or 60 minutes of ischemia, followed by up till
120 minutes of reperfusion. In a subgroup a protease inhibitor and/or MMP
inhibitor (or vehicle) will be administered intraluminally. At given time
points, tissue and blood is collected for further analysis.

The patients enrolled in this study will all undergo major upper abdominal
surgery. Because IR is applied to intestinal tissue which will be resected
anyway during the surgical procedure, this will not interfere with standard
surgical care. There are no specific benefits for the participating patients,
however in the future the results of our study will likely be useful for
patients suffering from intestinal IR. The additional risks for the patients in
this study are marginal and they will not increase the total operation risk. We
already performed similar studies in the human small intestine and these have
shown no negative effects or increased risk for the participating patients
compared to patients who undergo a PPPD (or whipple procedure) without
participation in the IR study (Hundscheid et al; article in preparation). We
refer to METC approval of our previous project *Complement activation after gut
ischemia-reperfusion injury in man* (MEC 06-3-044).
However, administration of a protease and MMP inhibitor are new components in
our ischemia-reperfusion studies. Nonetheless, these are substances that are
already administered very regularly in humans for other indications and it is
not to be expected that these substances cause any harmful effects.