Morphological characterization of anti-nociceptive effect of endogenous lipid palmitoylethanolamide in two murine models: peripheral mononeuropathy and diabetic polyneuropathy

Recent evidences suggest that mast cell activation and degranulation have a key role in the beginning and maintenance of a persistent pain, such as neuropathic one. Particularly mast cells are known to release NGF (Leon et al. 1994) and to express trkA (Horigome et al. 1993) receptors able to bind NGF. NGF loop may cause mast cell degranulation, leading to a further release of the neurotrophic factor NGF and many other pro-nociceptive and pro-inflammatory mediators. The release of NGF and other pro-inflammatory cytokines leads to a degeneration of nerve fibers. Therefore inhibition of mast cell degranulation could represent a good strategy in the cure of neuropathic pain. A class of molecules able to negatively modulate mast cell degranulation is represented by ALIAmides (from the acronym Autacoid Local Injury Antagonist), naturally-occurring lipid amide deriving from membrane fatty amides and structurally releted to endocannabinoids (Facci et al.1993, Mazzari et al., 1996). Palmitoylethanolamide (PEA), an endogenous lipid, is the most studied, in addition Petrosino and colleagues (2007) demonstrated that endogenous palmitoylethanolamide level decreased in the spinal cord of neuropathic mice. In our previous study (Costa et al., 2008) we demonstrated that palmitoylethanolamide (PEA), administered i.p. at the dose of 10 mg/kg for 7 days from the chronic constriction injury, a well established model of peripheral mononeuropathy, evoked a relief of both thermal hyperalgesia (increased sensitivity to thermal stimulus) and mechanical allodynia (pain due to a mechanical stimulus which does not normally provoke pain) in neuropathic mice (chronic constriction injury model) and significantly reduced the NGF levels in the sciatic nerve of neuropathic mice. In addition Costa and colleagues demonstrated PEA efficacy on allodynia which develops in diabetic mice and to partially increase NGF level. Starting by this assumption we wanted to characterize mast cell degranulation in the same animal model and after the same PEA treatment. In the chronic constriction injury model in addition to prolonged treatment (7 days), animals received PEA for a short period (2 days) because rapid mast cell activation was supposed. On 3rd and 8th day, 24h after the last administration, mice were sacrificed and the sciatic nerve and the spinal cord were removed from 3d and 8d group in each experimental group (sham/vehicle, CCI/vehicle and CCI/PEA). Sciatic nerves were processed in paraffin wax or Epon-Araldite resin to obtain respectively longitudinal (6 µm) or semi thin transversal (1 µm) sections, while spinal cord was in paraffin wax to obtain transversal (6 µm) sections. Another experiment was performed to collect spinal cords for a biochemical analysis. Spinal cord was removed because recent evidences suggest that microglia cells, in particular activated microglia, play a key role in the induction and maintenance of neuropathic pain. In the sciatic nerve longitudinal sections of control mice (3d and 8d group) arranged and consistent nerve morphology with a homogeneous localization of nuclei Schwann cells was observed, and limited number of intact and degranulated mast cells was present, indicative of physiological resident mast cells. Sciatic nerve of CCI mice (both groups) appeared oedematous without compactness of the fibers. In particular, in CCI mice of 3d group, the mast cells were localized in the surrounding tissue with a strong recruitment of intact mast cells, while in CCI mice of 8d group mast cells infiltrated in the inner central part of the sciatic nerve. In neuropathic mice treated with PEA of 3d and 8d group a mild inflammation and mast cell recruitment was observed. The evaluation of mast cell density (expressed as total mast cell/mm2) confirmed a time-dependent mast cell recruitmant after 3 and 8 days from the injury, indicating that mast cells degranulation plays a key role in the beginning and maintenance of neuropathic pain. The 2 days PEA treatment acted on mast cell recruitment; in fact mast cell density was comparable to that observed in sham mice. The 7 days PEA treatment modulated mast cell degranulation; in fact a significant reduction of average ratio of degranulated mast cells over intact mast cell compared to CCI mice was observed. The presence and implication of mast cell was confirmed by double immunostaining images obtained from longitudinal sections showing a co-expression of mast cell proteases I (MMCP-I) and trkA receptor. In the tranversal sections of sham mice of both groups no histological abnormalities were present: typically myelinated axons are packed in the sciatic nerve in an orderly, parallel arrangement with minimal interaxonal space. In transversal sections of CCI nerves of 3d and 8d group a lot of myelinated fibers underwent a Wallerian-like degeneration, suggested by a dense and flocculent axoplasmic matrix. In mice treated with PEA for 2 and 7 days a mild degeneration of fibers and compact arrangement like sham transversal section was observed. The quantitative analysis confirmed that the intact axons density (expressed as intact axons/mm2) and the myelin thickness (expressed as µm2) in CCI mice (both groups) significantly decreased respect to sham mice, and PEA treatment partially reduced fiber degeneration at both times, emphasizing its role role to prevent the reduction of myelin thickness at both 3 and 7 days after the injury. In order to verify the presence of activated microglia in neuropathic mice treated with PEA, the expression of F4/80 protein in transversal sections of the dorsal horn of the lumbar (L4-L5) spinal cord was investigated. This region was considered because the afferent fibers from the periphery (sciatic nerve) arrive here. Representative images show that in CCI mice a time-dependent increase of activated microglia was observed in the dorsal horn of L4 and L5 spinal cord compared to sham mice. In the contralateral horn of the spinal cord a milder activation of microglia was detected in L4 and L5 only in the 8d group. These results were confirmed by the densitometric analysis of F4/80 expression obtained by western blotting. The treatment with PEA for two consecutive days attenuated the limited microglia activation, restoring to physiological level the expression of F4/80 protein. In the third part of this project we evaluated the involvement of mast cell in the anti-allodinic effect of PEA in diabetic and neuropathic mice. Diabetes was induced by a single intraperitoneal injection of streptozotocin (STZ, 120 mg/kg). Mice were divided into two groups: 17d and 14d group. In the 17d group mice developed allodynia 14 days after STZ injection. Before the beginning of the treatment and at the end of experiment, hyperglycaemia was evaluated. On 14th day, PEA treatment started and three experimental groups were present: non diabetic, diabetic mice treated with vehicle and diabetic mice treated with PEA. PEA was administered i.p. at the dose 10 mg/kg for three consecutive days. 24 hours after the last administration PEA had an anti-allodinic effect and the following organs were removed: sciatic nerves (right and left), kidneys and pancreas. In 14d group PEA treatment started on 7th day, when mice were diabetic but non allodinic, and 24 hours from the last administration pancreas was removed. Since Costa and colleagues demonstrated that a mild insulin level in the blood was present in PEA treated mice of 14d group but non that of 17d group, morphological analysis and a subsequant quantative analysis was performed on pancreas tissue. Longitudinal and transversal sections of sciatic nerve were stained with toluidine blue in order to examine morphology of mast cell and axons. In longitudinal sections intact mast cell, degranulating and degranulated mast cells were uniformely distributed throughout the nerve of the three experimental groups and the count of total mast cell (total number of mast cell/mm2) didn’t show any difference among the three experimental groups. In transversal sections axons undergoing degeneration were present in diabetic mice treated with vehicle or PEA respect to non diabetic mice. The count of intact axons revealed that at the end of treatment diabetic mice treated with vehicle showed a significant decrease of intact axons and that this decrease remained constant in diabetic mice treated for three consecutive days with PEA. Nephropathy is another consequence of diabetes; therefore renal transversal sections were stained with periodic acid Shiff (PAS) reagent to investigate the presence of an expansion of mesangium in the glomeruli. Diabetic mice showed an increase of mesangial matrix and subsequently an increase of the glomeruli area compared to non-diabetic mice. In mice treated with three consecutive injections of PEA appeared similar to the images observed in diabetic mice treated with vehicle. The evaluation of the glomeruli area (expressed in µm2) confirmed the increase of mesangial area. In fact in diabetic mice treated with vehicle a significant increase of the area was recorded compared to non diabetic mice and the same increase was measured in diabetic mice treated with PEA for three consecutive days. In pancreas sections of diabetic mice treated with vehicle of 17d group, the islet of Langerhans appeared smaller and less numerous, while in diabetic mice treated with PEA could be observed a mild improvement. A quantitative analysis (the density expressed as number/mm2 and the area expressed as µm2) of islet of Langerhans demonstrated a significantly decreased in diabetic mice treated with vehicle and PEA had a protective partial effect on density and area of islets of Langerhans. In the group 14d pancreas sections of non diabetic mice appeared well organized with large and uniformely distributed islet of Langerhans. In diabetic mice treated with vehicle pancreas appeared with few islets which were small. In PEA treated mice an improvement was present: in fact islets were numerous and they appeared bigger than in diabetic mice. The quantitative analysis confirmed the previously considerations: in fact the density and the area of islets of Langerhans significantly decreased in diabetic mice treated with vehicle compared to non diabetic mice and 7 days PEA treatment highlighted the protective effect on the islets of Langerhans. The findings presented herein strongly suggest that PEA, without any side effects, is a promising compound in the cure of neuropathy. It may prevent mast cell degranulation through the already described ALIA (Autacoid Local Inflammation Antagonism) mechanism, modulates microglia activation in the spinal cord and this effect accounts for the antinociceptive property of PEA. In addition, PEA administered to diabetic and neuropathic mice elicited allodynia and exerts a time-dependent protective effect on islets of Langerhans.