Chapter x summary

Chapter I – Literature Survey. Heme oxygenase (HO) is the rate-limiting enzyme in the
degradation of heme, converting heme to CO, biliverdin and free iron. Biliverdin is subsequently
reduced to bilirubin by biliverdin reductase (BVR). Three HO isoforms have been identified: the stress-
inducible HO-1 and the constitutive HO-2/HO-3 isoforms which are expressed under basal conditions.
The HO-2 isoform is widely expressed in the gastrointestinal (GI) tract and more specifical y in
myenteric neurons, interstitial cel s of Cajal (ICCs) and mucosal epithelial cel s; co-localisation of HO-2
with neuronal NO synthase (nNOS) has been reported in different GI tissue preparations. Based on
the observation that enteric non-adrenergic non-cholinergic (NANC) neurotransmission was impaired
in HO-2-/- mice, the suggestion was made that CO might function as an inhibitory (i)-NANC
neurotransmitter in the enteric nervous system (ENS). Furthermore, other studies reported a possible
interaction between the HO/CO and NOS/NO signal ing pathways. The adaptive response of HO-1 to
various stimuli suggests that HO-1 may play an important role in conferring protection against different
types of stress. Accordingly, HO-1-/- mice show an increased systemic inflammatory response and
decreased survival after lipopolysaccharide (LPS) chal enge and renal ischemia, whereas HO-1
induction has been shown to protect tissues against inflammatory and oxidative stress injury.
Emerging evidence reveals that CO can exert diverse biological and cytoprotective effects. In the GI
tract, CO inhalation has been shown to reduce ischemia/reperfusion (I/R) injury of intestinal grafts and
chronic colitis in IL-10-/- mice. In addition, it has been reported that CO inhalation protects against the
development of postoperative ileus and necrotising enterocolitis. The recent discovery that certain
transition metal carbonyls function as CO-releasing molecules (CO-RMs) in biological systems
highlighted the potential use of this class of compounds as stratagem to deliver CO for therapeutic
purposes.
Chapter II. The aim of our experimental work was to investigate the role of the HO/CO signal ing
pathway in the GI tract under normal and pathological conditions; the latter was investigated in a
murine model of postoperative ileus (POI).
Chapter III. In a first study, we investigated the role of the HO/BVR system in i-NANC
neurotransmission in murine gastric fundus and jejunum. Previous studies have shown that both HO-2
and BVR are expressed in ICCs and co-localized with nNOS in a large proportion of myenteric
neurons along the GI tract. Neither HO inhibition by chromium mesoporphyrin (CrMP) nor co-
incubation with CO or biliverdin/bilirubin affected nitrergic neurotransmission – i.e. relaxations induced
by NANC nerve stimulation or exogenous NO – under normal physiological conditions. However,
biliverdin/bilirubin reversed the inhibitory effect of the superoxide generator LY83583 on exogenous
NO-induced relaxations in both tissues. When gastric fundus muscle strips were depleted of the
endogenous antioxidant Cu/Zn superoxide dismutase (SOD) by the Cu-chelator DETCA, electrical y
induced NANC relaxations were also affected by LY82583; however, biliverdin/bilirubin could not
substitute for the loss of Cu/Zn SOD when this specific antioxidant enzyme was depleted. In jejunal
muscle strips, the combination DETCA plus LY83583 nearly abolished contractile phasic activity and,
hence, did not al ow studying nitrergic relaxation in these experimental conditions. In conclusion, our
data do not establish a role for HO/CO in enteric NANC neurotransmission under normal physiological
conditions. However, the antioxidants biliverdin/bilirubin might play an important role in the protection
of the nitrergic neurotransmitter against oxidative stress.
Chapter IV. Although we did not confirm a role for CO as an endogenous NANC neurotransmitter
in the GI tract, it is well known that CO relaxes a number of GI preparations when administered
exogenously. Similar to NO, CO has been reported to act via activation of soluble guanylyl cyclase
(sGC), leading to an increase in cGMP levels. Our study confirms the involvement of sGC in CO-
evoked responses, as 1/ the sGC inhibitor ODQ (10 µM) abolished CO-induced relaxations; 2/ the
sGC sensitizer YC-1 was able to potentiate CO-evoked inhibitory responses; and 3/ the relaxant
response evoked by CO (300 µM) was associated with a significant increase in cGMP. Remarkably,
CORM-2-induced relaxations were only partial y reduced by ODQ – even at 100 µM – suggesting that
CORM-2 mediates GI smooth muscle relaxation in both a sGC-dependent and sGC-independent
manner. The mechanism(s) of relaxation ‘downstream of sGC/cGMP’ involved KCa channel activation
as wel as other unspecified mechanisms. Remarkably, the NOS inhibitor L-NAME also significantly
reduced CO- and CORM-2-evoked relaxations in jejunum, but had no effect on the relaxations in
gastric fundus. Col ectively, these results indicate that CO and CORM-2 cause enteric relaxation via
sGC and the subsequent activation of KCa channels, but an additional sGC-independent mechanism is
required for CORM-2; in addition, a NO-mediated amplification of CO signal ing in jejunum was
suggested.
Chapter V. As sGC was identified to be the main target of CO-mediated relaxations in the GI tract,
we wanted to investigate the role of the sGCα1β1 and sGCα2β1 isoforms in the relaxant effect of CO
and CORM-2 in murine gastric fundus using wild-type and sGCα -/-
(bolus)-induced relaxations were abolished by the sGC inhibitor ODQ, while CORM-2- and CO (infusion)-induced relaxations were only partial y inhibited by ODQ. In sGCα -/- responses to CO and CORM-2 were significantly reduced when compared to wild-type mice, but ODQ stil had an inhibitory effect. The sGC sensitizer YC-1 was able to potentiate CO- and CORM-2-induced relaxations in wild-type mice, but lost this potentiating effect in sGCα -/- 1 mice, CO-evoked relaxations were associated with a significant cGMP increase; however, basal and CO-elicited cGMP levels were markedly lower in sGCα -/- besides the predominant sGCα1β1 also the less abundantly expressed sGCα2β1 isoform plays an important role in the relaxant effect of CO in murine gastric fundus; however, the sGC stimulator YC-1 loses its potentiating effect towards CO in sGCα -/- 1 mice. Prolonged administration of CO – either by the addition of CORM-2 or by continuous infusion of CO – mediates gastric fundus relaxation in both a sGC-dependent and sGC-independent manner.
Chapter VI. A novel method for the evaluation of intestinal transit and contractility in mice using
fluorescence imaging and spatiotemporal motility mapping was introduced in order to facilitate the
study of GI motility in our fol owing experiments.
Chapter VII. Recent evidence indicates that a complex cascade of inflammatory responses within
the intestinal muscularis can be attributed as the root cause of POI fol owing intestinal manipulation
(IM). The anti-inflammatory mediator HO-1 is induced as part of the postoperative muscularis
inflammatory milieu and treatment of mice, rats, and pigs with inhaled CO can substantial y prevent
POI. In our study, IP administration of water-soluble CO-releasing molecules significantly reduced the
development of POI in mice; the inactive compounds (iCO-RMs) – which do not release CO – did not
provide any protection. When studying the mechanism(s) of action of the most effective compound –
i.e. CORM-3 – we demonstrated that CORM-mediated protection against POI was associated with a
down-regulation of pro-inflammatory mediators, iNOS activity as wel as a decrease in leukocyte
recruitment into the muscularis externa of the manipulated bowel. Importantly, these protective effects
appear to be, at least in part, mediated through induction of HO-1 – in a p38-dependent manner – and
a reduction of IM-induced ERK1/2 activation. Administration of CORM-3 was also shown to reduce the
early ‘oxidative burst’ in the mucosa fol owing IM, thereby preserving the mucosal barrier integrity.
Chapter VIII. Recent studies reported that CO might exert its anti-inflammatory effects through the
induction of PPARγ, a member of the nuclear hormone receptor superfamily. Besides its well-known
role in glucose metabolism, PPARγ has indeed been shown to play a pivotal role in the regulation of
inflammatory/immune responses. Surgical manipulation induced a rapid phosphorylation and
subsequent degradation of PPARγ within both the mucosa and muscularis of the manipulated colon.
Accompanying these modifications, there was a decrease in PPARγ DNA-binding activity which was
significantly restored by rosiglitazone treatment. The functional severity of POI was significantly
ameliorated in mice pre-treated with rosiglitazone; this was associated with a down-regulation of
inflammatory parameters, iNOS/COX-2 enzyme activity as well as a decrease in leukocyte recruitment
into the intestinal muscularis of both colon and jejunum. These anti-inflammatory effects were
preceded by a PPARγ-dependent inhibition of Egr-1, a key regulator of inflammatory gene expression.
In addition, rosiglitazone markedly reduced the early ‘oxidative burst’ in the mucosa fol owing colonic
manipulation; an effect that appeared to be independent of PPARγ. In conclusion, these data
demonstrate that PPARγ occupies a key role in the pathogenesis of POI and that rosiglitazone
prevents POI by suppression of the muscularis inflammatory cascade through a PPARγ-dependent
down-regulation of Egr-1.
In conclusion, our results do not establish a role for the HO/CO signal ing pathway in i-NANC neurotransmission in murine gastric fundus and jejunum under normal physiological conditions. However, both our in vitro and in vivo data suggest that the HO/CO pathway may be of benefit in the prevention and/or treatment of GI motility disorders caused by free radicals such as POI, septic ileus, intestinal I/R injury and diabetic gastroparesis. Future studies should further unravel the underlying molecular mechanism involved in HO/CO-mediated cytoprotection, with the main focus on heme-containing proteins (sGC, cytochrome c oxidase, Nox) and several transcription factors (PPARγ, Egr-1, HIF-1α).

Source: http://www.girem.be/phd1.pdf