Protective effects of PACAP in ischemia

PACAP has been shown to be neuroprotective in vitro in different neuronal cultures against various toxic insults and in models of neuronal injuries in vivo [33, 34]. Numerous in vivo data have been published showing its protective actions in cerebral ischemia [33, 35]. The first proof for the in vivo neuroprotective effect came from a rat global ischemia study, where intravenous or intracerebroventricular (icv) PACAP administration reduced hippocampal neuronal loss [36]. This was achieved via suppression of JNK and p38, while stimulation of ERK activity [37‐39]. These observations were followed by studies demonstrating that PACAP was also effective in transient and permanent focal ischemia in rats and mice induced by middle cerebral artery occlusion (MCAO) [27, 40‐44].

Subsequent studies provided further details on the neuroprotective mechanisms. Anti-apoptotic and anti-inflammatory actions seem to be the main protective mechanisms in PACAP’s actions in rat and mouse models of cerebral ischemia. PACAP decreased apoptosis in the ischemic penumbra [45], inhibited expression of bcl-2-associated death promoter, caspase-3, macrophage inflammatory protein-1alpha, inducible nitric oxide synthase2, tumor necrosis factor-(TNF) alpha mRNAs and increased ERK2, bcl-2 and IL-6 [40, 41, 46]. Decreased inflammatory response was also found after post-stroke PACAP-producing stem cell transplantation, where numerous chemokines as well as TNF, NFkappaB and IL-1 decreased [47]. In brain cortical neurons subjected to oxygen-glucose deprivation and reoxygenation, PACAP induced neuronal protection by both direct actions through PAC1 receptor, and indirect pathways via neurotrophin release, activation of trkB receptors and attenuation of neuronal growth inhibitory signaling molecules p75NTR and Nogo receptor [41]. In addition, PACAP induced apurinic/apyrimidinic endonuclease APE1 in hippocampal neurons that can be an additional factor reducing DNA stress and hippocampal CA1 neuronal death in global ischemia [48]. In mouse MCAO, several genes were affected in the ischemic core and penumbra after PACAP treatment [49‐52]. Among the upregulated genes was IL-6, which was strongly induced during the critical first 24 h, suggesting a relationship between PACAP and IL-6 in accordance with previous findings by Ohtaki and co-workers [40]. Several other cytokines and growth factors were altered in a region-specific and time-dependent fashion after post-ischemic PACAP treatment, such as brain derived neurotrophic factor [50, 51]. Whether alterations of these factors are consequences of PACAP reducing infarct volume by other mechanisms or represent a causative factor is not known at the moment. Only in case of IL-6, it has been proven that PACAP failed to improve ischemic lesion in IL-6-deficient mice, showing the causative role of IL-6 in PACAP-mediated neuroprotection in mice [40]. Numerous further factors playing a role in neuronal defense, axonal growth and development were also modified after ischemia [52]. A relationship between hypoxia inducible factor (HIF) and PACAP was described in several studies in different experimental paradigms [53‐55]. Under in vitro and in vivo hypoxic conditions, HIF1-alpha activation upregulated PACAP, which in turn activated PAC1 receptor [56]. Although PACAP reduced HIF1-alpha expression in a model of diabetic retinopathy 2 weeks after the treatment, bone marrow-derived stem cells homing into the ischemic brain was also facilitated by a recently described HIF1-alpha-activated PACAP38-PAC1 signaling process [55]. A detailed time-dependent analysis of PACAP’s effect on cerebral HIF1 expression could clarify the role of this pathway in PACAP-induced neuroprotection in ischemia. Analogs of PACAP were also tested in focal ischemic models. In a study of ischemia/reperfusion injury, a potent metabolically stable PACAP38 analog [acetyl-(Ala¹⁵, Ala²⁰) PACAP38-propylamide] led to the same degree of protection as native PACAP38 [46]. This is an important finding, as one of the limitations of PACAP’s therapeutic use is its poor stability. However, according to these data enhancing its plasmatic half-life did not lead to an increase of its neuroprotective potential [46], but analogs might have less vasomotor side effects, as described in another study [57].

As far as functional recovery is concerned, PACAP is able to improve functional deficits in association with the morphological amelioration in stroke models. In rat permanent focal cerebral ischemia, PACAP improved certain sensorimotor deficits, such as reaction times to body surface touch [27]. Another study further supported this in a transient MCAO, evaluating neurological impairment by degree of limb flexion, grasping and symmetry of movements [46]. In a permanent focal ischemia model, PACAP-producing stem cells transplanted icv 3 days after stroke promoted functional recovery even when given beyond the therapeutic window for structural recovery [47].

PACAP is known to cross the blood-brain barrier (BBB), but it is still questionable whether the rate is sufficient to lead to effects in the brain under physiological or pathological conditions [2, 38]. Although ischemic conditions change region-specific crossing, it is suggested that the passage is sufficient enough to induce neuroprotection in ischemic brains [58]. Antisenses inhibit efflux pumps of the BBB, and could inhibit PACAP27 efflux and reduce the infarct size in mouse focal ischemia [59]. Regarding changes in cerebral blood flow, in some studies PACAP increased cerebral blood flow in ischemic conditions, while in others no change or even decrease was found [27, 46, 60]. PACAP has potent vasodilatory effects, which can also be included in the pathomechanism of migraine [61‐63]. However, given the contradictory data on cerebral blood flow after PACAP treatment, it remains unknown at the moment whether this effect plays a role in post-ischemic neuroprotection.

The role of endogenous PACAP was suggested by upregulation of PACAP signaling in different ischemia models and from knockout studies (Table 1). In a gerbil model of global ischemia, decrease in PACAP expression was followed by an increase 5 days later. This was accompanied by increases in PAC1 receptor expression in the vulnerable CA1 region, in contrast to the more resistant CA3 area, where PACAP expression did not change [36, 64]. Upregulation of PAC1 receptor could also be observed after focal ischemia [65, 66]. A massive upregulation of PACAP was found in peri-infarct regions [67]. In a rat global ischemia model, moderate PAC1 mRNA decrease was observed throughout the hippocampus, while granule cells showed increased PACAP expression [68]. It was suggested that the altered PACAP and PAC1 receptor expression might play a role in regulated neurogenesis after stroke [68]. In mouse hippocampal astrocytes, PAC1 receptor expression was increased 7 days after stroke, suggesting an important role of PACAP in reactive astrocytes [69, 70]. Further evidence for the endogenous protection by PACAP came from studies using PACAP deficient mice. Hetero- and homozygous PACAP knockout animals had increased infarct volume with increased edema formation and more severe neurological deficits after MCAO, and these could be ameliorated by PACAP injection [40, 71]. Furthermore, cytochrome-c release was higher, while mitochondrial bcl-2 was lower in mice lacking PACAP. It was also suggested that these protective effects could be mediated in part by IL-6 [40]. Endogenous PACAP also promotes hippocampal neurogenesis after stroke, as proliferation of neuronal stem cells in the subgranular zone of the hippocampus was found to be increased in wild type mice, but not in PACAP heterozygous animals [72].

The few available human data also support that PACAP might play a role in ischemic neuronal conditions. It was hypothesized that the elevated blood PACAP levels may reflect an increased leakage into circulation or an overproduction of PACAP as a pathological response to the loss of neural tissue in the CNS and it might be associated with the neuroprotective effects of the neuropeptide [73]. Plasma PACAP concentrations were higher in patients after acute spontaneous basal ganglia and aneurysmal subarachnoid hemorrhages than in healthy control subjects [73, 74]. Positive association was shown between PACAP levels and neurological score, as well as with hematoma volume. Patients, who died within the first week after admission, had higher PACAP levels and overall survival times were shorter in individuals with high PACAP concentrations [73, 74]. It is suggested that PACAP could be a good prognostic predictor in hemorrhage patients. These studies suggest that PACAP can be an independent predictor of survival and a potential prognostic biomarker of brain hemorrhage.