Salvatore Salomone

Drug discovery efforts in Alzheimer's disease (AD) have been directed in the last ten years to develop "disease-modifying drugs" able to exert neuroprotective effects in an early phase of AD pathogenesis. Unfortunately several candidate disease-modifying drugs have failed in Phase III clinical trials conducted in mild to moderate AD for different methodological difficulties, such as the time course of treatment in relation to development of disease as well as the appropriate use of validated biological and neuropsychological markers. Mild cognitive impairment (MCI) has been considered a precursor of AD. Much effort is now directed to identify the most appropriate and sensitive markers which can predict the progression from MCI to AD, such as neuroimaging markers (e.g. hippocampal atrophy and amyloid positron emission tomography imaging), cerebrospinal fluid markers (i.e. association of elevated tau with low levels of amyloid β -peptide(1-42) and neuropsychological markers (i.e. episodic memory deficits and executive dysfunction). Recent studies demonstrate that the combination of these different biomarkers significantly increases the chance to predict the conversion into AD within 24 months. These biomarkers will be essential in the future to analyze clinical efficacy of disease-modifying drugs in MCI patients at high risk to develop AD. In the present review we analyze recent evidence on the combination of neuropsychological and biological markers in AD as a new tool to track disease progression in early AD as well as the response to disease-modifying drugs.

Diabetic retinopathy, a major cause of vision loss, is currently treated with anti-VEGF agents. Here we tested two hypotheses: (i) high glucose damages retinal pericytes, the cell layer surrounding endothelial cells, via VEGF induction, which may be counteracted by anti-VEGFs and (ii) activation of PLA2/COX-2 pathway by high glucose might be upstream and/or downstream of VEGF in perycites, as previously observed in endothelial cells. Human retinal pericytes were treated with high glucose (25mM) for 48h and/or anti-VEGFs (40μg/ml aflibercept, 25μg/ml bevacizumab, 10μg/ml ranibizumab). All anti-VEGFs significantly prevented high glucose-induced cell damage (assessed by LDH release) and improved cell viability (assessed by MTT and Evans blue). High glucose-induced VEGF-A expression, as detected both at mRNA (qPCR) and protein (ELISA) level, while receptor (VEGFR1 and VEGFR2) expression, detected in control condition, was unaffected by treatments. High glucose induced also activation of PLA2/COX-2 pathway, as revealed by increased phosphorylation of cPLA2, COX-2 expression and PGE2 release. Treatment with cPLA2 (50μM AACOCF3) and COX-2 (5μM NS-392) inhibitors prevented both cell damage and VEGF-A induced by high glucose. Finally, challenge with exogenous VEGF-A (10ng/ml) induced VEGF-A expression, while anti-VEGFs reduced VEGF-A expression induced by either high glucose or exogenous VEGF-A. These data indicate that high glucose directly damages pericytes through activation of PLA2/COX-2/VEGF-A pathway. Furthermore, a kind of feed-forward loop between cPLA2/COX-2/PG axis and VEGF appears to operate in this system. Thus, anti-VEGFs afford protection of pericytes from high glucose by inhibiting this loop.

Nitric oxide (NO), carbon monoxide (CO) and hydrogen sulphide (H2S) modulate vascular tone. In view of their therapeutic potential for ocular diseases, we examined the effect of exogenous CO and H2S on tone of isolated rabbit ophthalmic artery and their interaction with endogenous and exogenous NO. Ophthalmic artery segments mounted on a wire myograph were challenged with cumulative concentrations of phenylephrine (PE) in the presence or absence of NG-nitro-L-arginine (LNNA) to inhibit production of NO, the CO-releasing molecules CORMs or the H2S-donor GYY4137. The maximal vasoconstriction elicited by PE reached 20-30% of that induced by KCl but was dramatically increased by incubation with LNNA. GYY4137 significantly raised PE-mediated vasoconstriction, but it did not change the response to PE in the presence of LNNA or the relaxation to sodium nitroprusside (SNP). CORMs concentration-dependently inhibited PE-induced constriction, an effect that was synergistic with endogenous NO (reduced by LNNA), but insensitive to blockade of guanylyl cyclase by 1H-[1,2,4]oxadiazolo[4,3,-α]quinoxalin-1-one (ODQ). In vascular tissues cyclic GMP (cGMP) levels seemed reduced by GYY4137 (not significantly), but were not changed by CORM. These data indicate that CO is able per se to relax isolated ophthalmic artery and to synergize with NO, while H2S counteracts the effect of endogenous NO. CO does not stimulate cGMP production in our system, while H2S may reduce cGMP production stimulated by endogenous NO. These findings provide new insights into the complexities of gas interactions in the control of ophthalmic vascular tone, highlighting potential pharmacological targets for ocular diseases.

Homocysteine has been associated with extracellular matrix changes. The diabetic retinopathy is a neurovascular complication of diabetes mellitus and it is the leading cause of vision loss among working adults worldwide. In this study, we evaluate the role of homocysteine in diabetic retinopathy analyzing the plasma levels of homocysteine in 63 diabetic type 2 patients with nonproliferative retinopathy (NPDR), 62 patients with proliferative diabetic retinopathy (PDR), 50 healthy subjects used as control group, and 75 randomly selected patients.

To assess the relative bioavailability of a new subcutaneous (SC) diclofenac hydroxypropyl b-cyclodextrin (HPbCD) formulation administered to three body sites: quadriceps, gluteus, and abdomen. This was a pilot, single-dose, randomized, three-way crossover relative bioavailability study. A total of 12 healthy subjects received a single SC injection of diclofenac HPbCD 50 mg/1 mL in the quadriceps, gluteus, or abdomen. The AUC was comparable after SC diclofenac HPbCD in the quadriceps, gluteus, and abdomen. The Cmax was comparable after SC administration in the quadriceps or abdomen, and ~ 17% higher in the gluteus. The absorption was rapid (30 minutes) after administration of the treatment at any site. The treatment was well tolerated. The relative bioavailability of SC diclofenac HPbCD was comparable when administered to the quadriceps, gluteus, and abdomen. The new diclofenac formulation can therefore be administered subcutaneously to any of these sites without clinically significant differences. A further adequately powered study would be necessary to reveal any differences among injection sites in terms of peak plasma concentration.