Restoration of Cyclo-Gly-Pro-induced salivary hyposecretion and submandibular composition by naloxone in mice

Cyclo-Gly-Pro (CGP) attenuates nociception, however its effects on salivary glands remain unclear. In this study, we investigated the acute effects of CGP on salivary flow and composition, and on the submandibular gland composition, compared with morphine. Besides, we characterized the effects of naloxone (a non-selective opioid receptor antagonist) on CGP- and morphine-induced salivary and glandular alterations in mice. After that, in silico analyses were performed to predict the interaction between CGP and opioid receptors. Morphine and CGP significantly reduced salivary flow and total protein concentration of saliva and naloxone restored them to the physiological levels. Morphine and CGP also reduced several infrared vibrational modes (Amide I, 1687-1594cm-1; Amide II, 1594-1494cm-1; CH2/CH3, 1488-1433cm-1; C = O, 1432-1365cm-1; PO2 asymmetric, 1290-1185cm-1; PO2 symmetric, 1135-999cm-1) and naloxone reverted these alterations. The in silico docking analysis demonstrated the interaction of polar contacts between the CGP and opioid receptor Cys219 residue. Altogether, we showed that salivary hypofunction and glandular changes elicited by CGP may occur through opioid receptor suggesting that the blockage of opioid receptors in superior cervical and submandibular ganglions may be a possible strategy to restore salivary secretion while maintaining antinociceptive action due its effects on the central nervous system.

Introduction Saliva exerts multiple functions in the oral cavity such as protection against microorganisms, contribution to the taste and digestion and maintenance of oral health [1][2][3]. Salivary function is controlled by sympathetic and parasympathetic nervous system, which innervate acinar, ductal, myoepithelial and vascular cells in salivary glands [4,5]. The activation of muscarinic receptors in the acinar cells is the most important control of salivary flow rates [6]. Electrical stimulation of sympathetic efferent branch to the salivary glands results in a low flow of saliva which is rich in proteins [7]. Paradoxically, sympathectomy also generates decrease in salivary flow [8]. These findings demonstrate the complexity of the sympathetic regulation on salivary flow and salivary composition [9]. The activation of central pathways develops a great part in salivatory effects of intraperitoneal pilocarpine in rats [10].
Several substances with pharmacological properties can promote changes in salivary function. Morphine is an opioid receptor agonist that plays intense and long-lasting analgesia [11,12]. It was demonstrated that morphine increase lactate levels in serum, however its effects on salivary lactate concentration are unknown [13]. In humans and rats, the morphine administration was correlated to hyposalivation, and associated with changes in the ionic composition [14][15]. Bearing in mind that amylase is the most abundant protein in saliva [16], salivary amylase concentration decreased after morphine treatment [17]. It has been clearly demonstrated that morphine promotes reduction in the sympathetic activity to salivary glands by its action on the superior cervical ganglion and by inhibiting the release of neurotransmitter from postganglionic nerve endings [17]. Additionally, kappa-, delta-, and mu-opioid-receptor agonists are able to inhibit L-, N-and P/Q-types of calcium channels in submandibular ganglion neurons, indicating a reduction in parasympathetic activity to salivary glands [18]. The reduction on the parasympathetic nerve-induced salivary secretion generated by the morphine was partially reversed by naloxone, a non-selective opioid receptor antagonist. However, salivary secretion stimulated by intravenous infusion of acetylcholine was not reduced by morphine [19].
Cyclic dipeptides are among the smallest peptide derivatives frequently found in nature [20]. Cyclo-Gly-Pro (CGP) is an endogenous diketopiperazine derived from N-terminal tripeptide, glycine-proline-glutamate which is naturally cleaved from the insulin-like growth factor 1 (IGF-1) [21,22]. Previous studies have shown that CGP induces neuroprotective effects after ischemic brain injury [22]. CGP 35348 has an adjuvant role to produce a dose-dependent antagonism of antinociception [23]. Recently, our group demonstrated that the antinociceptive effect of CGP seemed to be mediated by the interaction with the opioid system, also reducing the hyper nociception and paw inflammation induced by carrageenan [24]. This might indicate the potential of CGP as a candidate for antinociceptive role with fewer side effects on salivary glands. It is important to emphasize that several effects of CGP in oral territories remain unknown.
Despite the knowledge about the effect of pharmacological agents on salivary glands, and consequently on salivary secretion, the CGP capacity to modulate submandibular and salivary components has never been investigated. Besides, it is important to highlight that the interaction between CGP and opioid receptors has also not been demonstrated. Thus, the aims of the present study were to investigate the CGP acute effects on salivary flow and composition, and on submandibular gland composition compared with morphine. Besides, we characterized the naloxone (a non-selective opioid receptor antagonist) effect on CGP-and morphine-induced salivary and glandular alterations in mice. After that, in silico analyses were performed to predict the 3D-interaction between the CGP and opioid receptors.

Materials and methods
This study was carried out in strict accordance with the recommendations in the Guide for the Care and Use of Laboratory Animals of the Brazilian Society of Laboratory Animals Science (SBCAL). Experimental procedures were approved by the Ethical Committee of the Federal University of Alagoas (UFAL) (License 065/2011), according to Ethical Principles adopted by the Brazilian College of Animal Experimentation (COBEA). Animal studies are reported in compliance with the approved guidelines. To minimize the number of animals used and their suffering all effort were taken. Male Swiss mice (Mus musculus, 2 months) weighing 25-36 g were obtained from the breeding colonies of the UFAL and maintained at the Institute of Biological Sciences and Health rodent housing facility. Mice were randomly assigned to standard cages with five animals per cage. Animals were allowed free access to water and standard rodent chow diet and kept at 22 ± 2˚C with a 12 h light/dark cycle, light on at 07:00h. To minimize circadian effects, all experimental procedures were conducted during the light phase. Power analysis was used as a basis to set the number of animals per experiment [25]. The number of samples was insert in each legend.

Experimental procedures
Animals were treated with vehicle (NaCl, 0.9%), morphine or CGP (CGP, � 98% purity) ( Fig  1, Protocol 1). Morphine was applied at dose of 17.5 μmol kg -1 (0.1 ml/10g, i.p.) and CGP 1 μmol kg -1 (0.1 ml/10g, i.p.), considering the similar antinociceptive effects observed in the hot plate test [24]. For randomization, vehicle was injected in the control group, while the other mice received morphine or CGP. Blinding was implemented as follows: the operator was blinded to the group identity, but not to animals of the same group. Thus, for i.p. drug injection, different solutions were prepared: vehicle, morphine and CGP. In order to analyze the opioid receptors involvement in salivary secretion and changes in submandibular composition, similar analysis was performed in another set of animals under naloxone administration (pre-treatment)(15.3 μmol kg -1 , i.p.), an opioid receptor antagonist, 15 minutes before treatment with vehicle, morphine or CGP (Fig 1, Protocol 2).

Saliva and salivary glands collection
One hour after treatment with vehicle, morphine or CGP, the animals were intraperitoneally anesthetized (xylazine 5 mg kg -1 body weight; ketamine 35 mg kg -1 ) and then parasympathetic stimulation was performed for salivary secretion through pilocarpine injection (2 mg kg -1 , i. p.). Total saliva was collected for 10 min from the oral cavity [26]. After that, submandibular and parotid glands were collected and weighted [27]. Salivary secretion was calculated based on volume of saliva acquired in 10 minutes collection divided by the weight of the salivary gland tissues (μl/g tissue) (Fig 1).

Total protein concentration of saliva
Total protein concentration was measured using Bradford Protein assay. Values were expressed in mg/ml using serum albumin as standard protein. [28].

Saliva analysis by dispersive x-ray analysis system (EDX)
Pilocarpine-stimulated saliva from Protocol 1 animals was used to assess the inorganic elements by EDX (Fig 1). In each experiment,15 μL of saliva were used to measure the ionic composition. EDX mode was set up as 180 seconds per sample ion detection using Si (Li) semiconductor detector (Shimadzu, Fukuoka, Japan) at 30 kV in a vacuum chamber. Ions quantification was taken from the excitement of their electrons.

Lactate concentration of submandibular glands
Submandibular gland specimens were tested using an enzymatic system for lactate quantitative determination (Labtest, Brazil). Experiment was done according to the manufacturer's Experimental design of the acute treatment with vehicle, morphine or CGP in male Swiss mice in two protocols with or without the previous administration of naloxone. In protocol 1, mice were treated acutely with vehicle (VEH, NaCl, 0.9%), morphine (MOR, 17.5 μmol kg -1 , ip) or CGP (1 μmol kg -1 , ip). After one hour of treatment, a parasympathetic stimulus with pilocarpine (PILO, 2 mg kg -1 , ip) was performed during 10 min for salivary secretion and salivary composition analysis. Immediately after saliva collection, submandibular glands were removed and storage at −80˚C. The CGP and morphine effects on submandibular gland molecular composition were observed by Attenuated total reflectance Fourier transform infrared (ATR-FTIR) spectroscopy and lactate level analysis. In protocol 2, naloxone (15.3 μmol kg -1 , ip) was administered 15 minutes prior to acute treatment with vehicle, morphine and CGP in order to evaluate the opioid receptors involvement in saliva and submandibular gland. instruction. The submandibular tissue was removed, washed using physiological saline (NaCl 0.9%), and immediately frozen at-80˚C. Frozen gland tissue was homogenized in a phosphate buffer (1 : 10 w/v, pH 7.4). To measure the lactate concentration, the homogenate (25 μg) was incubated in a solution with 4-aminoantipyrine (50 mmol/L), peroxidase, L-lactate oxidase (1,000 U/L) e N-ethyl-N-(2-hydroxy-3-sulfopropyl)-3-methylaniline (1.5 mmol/L) and sodium azide (0.09%) at 37˚C for 5 minutes using a spectrophotometer at 340 nm.

Histology
The histological analysis in submandibular glands were performed only in animals described in protocol 1 (Fig 1). Submandibular glands were fixed in 10% buffered formalin. Subsequently, these glands were dehydrated in alcohol (80, 90 and 100%), cleared in xylene and embedded in paraffin. Histological sections with a thickness of 5 μm were acquired using a microtome (Leica RM2125). Then these sections were placed on slides and stained with hematoxylin and eosin. Histological slides were examined and micrograph pictures were obtained using an optical microscope (Olympus BX41).

Acetylcholinesterase activity of submandibular glands
The acetylcholinesterase (AChE) activity quantification in submandibular glands were performed only in animals described in protocol 1 (Fig 1) using a acetylcholine as substrate (Ach, Sigma Chemical, St. Louis, Mo, USA) at room temperature. The submandibular glands were homogenized in saline (1:10, 0.9%), through a tissue shredder (Polytron1). In a 96-well plate, acetylcholine (50μL) was added, and then color reagent was added and taken to incubator for 3 minutes at 37˚C. Subsequently, sample (20μL) was added [the first well received deionized water (20μL) and the second the standard solution (20μL)] and then taken back to the water bath for 2.5 minutes at 37˚C. Finally, a blocked solution (150μL) was added for subsequent reading at the spectrophotometer (at absorbance 410nm). The AChE activity was determined by the sample value product. Enzyme activity values were calculated after normalization by a total protein concentration measured using Bradford assay.

Molecular profile in submandibular glands by ATR-FTIR spectroscopy
Submandibular glands spectra were recorded in 4000-400 cm-1 region using FTIR spectrophotometer Vertex 70 (Bruker Optik) using a micro-attenuated total reflectance (ATR) accessory. The fingerprint region was chosen to be displayed due to the interest region. All spectra were recorded at room temperature (23±1˚C). The crystal material unit in ATR unit was a diamond disc as internal-reflection element. The sample penetration depth ranges between 0.1 and 2 μm and depends on the wavelength and the refractive index of the ATR-crystal material. In the ATR-crystal the infrared beam is reflected at the interface toward the sample. Twenty mg of submandibular were lyophilized using a rotary evaporator (Thermo Savant, San Jose, CA) to obtain sample spectra. The air spectrum was used as a background in ATR-FTIR analysis. Samples spectrum were taken with 4 cm -1 of resolution and 32 scans were performed to each analysis. The ATR-FTIR spectra were also baseline corrected using OPUS software [29]. Table 1 shows the frequencies and assignments of the vibrational modes identified in submandibular glands. Briefly, the vibrational mode between 1687-1594 cm -1 is identified as υ N H (Amide I) bending vibrations [29][30][31]. The δ N H (Amide II) bending vibration is usually represented between 1594-1494cm -1 [29,32]. The vibrational modes between 1488-1433 cm -1 are attributed to CH 2 /CH 3 vibrations. . The band between 1432-1365 cm -1 demonstrates CO groups (ester) stretching vibrations. Besides, spectral area between 1290-1185 cm -1 indicates PO 2 asymmetric [29]. The 1135-999 cm -1 spectral area corresponds PO 2 symmetric [29].

Cyclo-Gly-Pro (CGP) and opioid receptor (OR) structure assembly and interaction
The human Opioid Receptor protein FASTA sequence (Homo sapiens, access number in Gene Bank: AAA73958.1) was submitted online in I-TASSER server to predict and generate high-quality 3D predictions of this protein. The best model was verified using RAMPAGE: Assessment of the Ramachandran Plot, and Verify3D web tools to determine the spatial coherence and compatibility of the atomic model (3D) with its own amino acid sequence (1D). The CGP 3D structure was obtained from Pubchem (PubChem CID: 193540). After that, in silico analyses were performed to predict the interaction of both structures. Auto-DOCK Vina [34] was used do predict the molecular docking using the Root-mean-square deviation of atomic positions (RMSd) and free energy calculations. PyMOL Molecular Graphics System, Version 2.0 Schrödinger, LLC, was used to visualize the CGP-OP interactions and export image files.

Statistical analysis
In this study, data and statistical analysis comply with the recommendations on experimental design and analysis in pharmacology [35]. Values are presented as mean ± SEM. The heat map and analyses were performed using GraphPad Prism 5.0 (GraphPad Software, San Diego, CA, USA). Kolmorogov-Smirnov test was used to determine the normality of the sample distributions. Comparisons of results were performed by ANOVA, followed by Turkey's or Dunnett's post-test at P value <0.05.

Effects of opioid receptors (OR) in salivary secretion and salivary composition
Salivary secretion was significantly reduced in morphine compared to vehicle (~15%; p <0.05). Similarly, CGP significantly decreased salivary secretion compared to vehicle (~20%, p <0.05). Differences in salivary secretion were not significant between morphine-and CGPtreated mice (Fig 2A). To confirm the effects of the opioid receptors, naloxone was administered before the treatment with vehicle, morphine or CGP. The pilocarpine-stimulated salivary secretion remained unchanged after naloxone administration in vehicle mice (p >0.05; Fig  2A). On the other hand, naloxone significantly increased the salivary secretion in mice treated with CGP and morphine (~20% and 30%, respectively; p <0.01) (Fig 2A). In order to analyze the effects of CGP and morphine, total protein concentration in saliva was measure using Bradford assay. Total protein concentration in saliva significantly decreased (p<0.01) when CGP (~45%) and morphine (~35%) was administrated and compared to vehicle (Fig 2B). Total protein concentration in saliva was similar (p>0.05) in morphine and CGP-treated mice. The administration of naloxone in vehicle mice kept the total protein concentration stable (p >0.05; Fig 2B). On the other hand, naloxone significantly increased the total protein concentration in saliva of mice treated with CGP and morphine (~40%; p <0.001; Fig 2B).
Dispersive X-ray analysis was performed to analyze the CGP and morphine effects on the ionic composition in saliva. The composition of potassium, chloride, sodium and sulfur ions in saliva remained unchanged after acute treatment with morphine or CGP compared to vehicle (S1 Fig). , an opioid receptor antagonist, was administered 15 minutes before treatment with vehicle, morphine or CGP. One hour after treatment, parasympathetic stimulation was performed for salivary secretion through pilocarpine injection (2 mg kg -1 , i.p.) and total saliva was collected for 10 minutes from the oral cavity. (A) Salivary flow was expressed as salivary secretion per gram of glandular tissue (μl/g tissue), equivalent to the ratio of the volume of salivary secretion and the weight sum of glandular parotid and submandibular. (B) Salivary total protein concentration was expressed as μg/μL. Results are represented as mean ± SEM of 6-9 animals; � P<0.05 vs. vehicle; #P<0.05 vs. morphine; &P<0.05 vs CGP. One-way ANOVA, Turkey as post hoc test. https://doi.org/10.1371/journal.pone.0229761.g002

Maintenance of salivary glands weight under opioid receptors (OR) blockade
In order to investigate the CGP or morphine effects in salivary glands, the parotid and submandibular glands were properly weighed. The results demonstrated that the glands weight remained unchanged after acute treatment with morphine or CGP ( Table 2). As expected, the salivary glands weight also did not change with the naloxone pre-treatment (Table 2).

Histological changes of submandibular gland under opioid receptors (OR) blockade
To determine whether CGP promoted morphological and structural changes in salivary gland compared to the morphine treatment, we stained the submandibular sections with hematoxylin-eosin. The acinar cells, ductal cells and connective tissue remained unaltered (S2A Fig). However, submandibular glands histological analysis showed of evacuated spaces increased between serous and mucous acini in glandular parenchyma after one-hour treatment with morphine or CGP compared to the vehicle (S2B and S2C Fig).

Effects of opioid receptors (OR) blockade on lactate levels of submandibular glands
Morphine (~15%) and CGP (~15%) increased (p<0.001) lactate levels in submandibular gland compared to vehicle (Fig 3). Lactate levels in submandibular glands were similar (p>0.05) in morphine and CGP-treated mice. The lactate levels in submandibular gland remained unchanged after naloxone administration in vehicle mice (p >0.05; Fig 3). On the other hand, naloxone significantly decreased this parameter in mice treated with CGP and morphine (~20%; p <0.001) (Fig 3).

Effects of opioid receptors (OR) blockade on acetylcholinesterase activity in of submandibular glands
The acetylcholinesterase activity in submandibular gland was unaffected after acute treatment with morphine (0. 16

PLOS ONE
Restoration of cyclo-Gly-Pro-induced salivary hyposecretion

Effects of opioid receptors (OR) blockade on chemical profile in submandibular by ATR-FTIR spectroscopy
The submandibular gland infrared spectrum is a superposition of several compounds and the absorption bands in ATR-FTIR spectrum intensities are directly proportional to the components concentration. The submandibular gland spectra of vehicle-, CGP-or morphine-treated animals in presence or absence of naloxone are represented in Fig 4A. To understand better the response of proteins, phospholipids and lipids in submandibular gland under opioid receptors (OR) blockade, we analyzed the area of molecular components using ATR-FTIR spectroscopy. There are six major vibrational modes highlighted at 1635 cm -1 (1687-1594 cm -1 ), 1550 cm -1 (1594-1494 cm -1 ), 1450 cm -1 (1488-1433 cm -1 ), 1400 cm -1 (1432-1365 cm -1 ), 1232 cm -1 (1290-1185 cm -1 ) and 1031 cm -1 (1135-999 cm -1 ), confirming the presence of proteins, lipids, phospholipids and glycogen, according to the details showed for each peak, corresponding to the specific vibration molecules (Fig 4A). The vibrational modes between 1687-1594 cm -1 and 1594-1494 cm -1 , representing amide I and amide II, respectively, were reduced in morphine-or CGP-treated mice. These changes were reversed by the pre-treatment with naloxone (Fig 4B and 4C). The vibrational modes between 1488-1433 cm -1 and 1432-1365 cm -1 , representing CH 2 /CH 3 and C = O, respectively, were also reduced in morphine-or CGP-treated mice. These changes were also reversed by the pre-treatment with naloxone (Fig 4D and 4E). Besides, two vibrational modes at 1290-1185 and 1135-999 cm -1 were also reduced in morphine-or CGP-treated mice compare to the vehicle. These vibrational modes represent PO 2 asymmetric and PO 2 symmetric, respectively, and the naloxone pre-treatment reversed the changes in both submandibular glands components (Fig 4F  and 4G). A heat map with the mean relative changes clearly demonstrates the expression in these vibrational modes (Fig 4H).

Assembly and interaction of the OR and CGP structure
In silico modeling of OR and molecular docking of OR and CGP were performed by I-TASSER server. Fig 5A shows the full cartoon structure of OP (green) interacting with CGP (red). The extended view of the interaction site from docking analysis demonstrated the polar contacts (yellow dashes) between the CGP (red) and the OR Cys219 residue (Fig 5B). Fig 5C shows the

PLOS ONE
Restoration of cyclo-Gly-Pro-induced salivary hyposecretion full surface of the structure of OP (green) coupled with CGP (red) and Fig 5D presents the OP framework and conformational interaction site.

Discussion
The opioid agonist depressant mechanisms on salivary secretion and salivary glandular tissue effects remain unclear. Additionally, the potential effects of therapeutic agents in secondary target organs as salivary glands still need to be carefully investigated. Given the ongoing attempts to describe opioid side effects on salivary function, it is clearly important to understand and characterize the CGP effects in saliva and salivary glands composition compared with morphine. We showed that either CGP or morphine promoted the reduction in flow rates and in salivary protein concentration, increased lactate levels in glandular tissue and resulted in severe changes in chemical components in submandibular glands. Conversely, the naloxone (a non-selective opioid receptor antagonist) reversed this alteration in saliva and submandibular glands.
Human salivary secretion was decreased from 1 to 4 hours after administration of morphine intravenously [36]. Besides, the treatment with morphine (6 mg/kg) also inhibited the salivary flow in rats [14]. As expect, in this present study the acute treatment with morphine promoted salivary secretion reduction. Furthermore, for the first time, we showed that CGP also reduces salivary secretion in mice.
Previous studies have demonstrated that naloxone partially restores pilocarpine-stimulated and parasympathetic nerve stimulated-salivary secretion under morphine treatment [14,19]. As expected, our data corroborates previous studies demonstrating that naloxone reversed the morphine depressant effects in salivary flow. We further investigated whether the salivary flow changes elicited by CGP were also reversed by naloxone, due to its nonspecific antagonistic action on mu, kappa and delta opioid receptors [37]. It is important to point out that the opioid receptors blockage by naloxone was also able to reverse the inhibitory effect caused by CGP and to restore the salivary flow, indicating the CGP effect in opioid receptors as described to morphine treatment. So, these data suggest similar side effects on salivary glands by CGP or morphine.
Salivary ionic composition remained unchanged after acute treatment with CGP or morphine (5 mg kg -1 ). In previous research, morphine at a dose of 25 mg kg -1 did not alter the sodium and calcium concentrations in parotid saliva of rats; on the other hand, led to an increase in salivary potassium concentration [19]. Interestingly, morphine at a dose of 6 mg kg -1 did not alter the presence of salivary potassium concentration, however it was able to reduce the salivary calcium and increase salivary sodium concentration [14]. It is possible that samples collected using different methods contributed to these different results. Despite the contradictory reports evaluating morphine effects on ions composition under lower and higher doses, our data on CGP showing no changes in salivary ionic composition has never been reported and suggests that CGP may not be involved in the regulation of several channels that regulate ionic composition in salivary glands.
We have also shown that CGP, as well as morphine, decreased salivary protein concentration in mice, which is corroborated elsewhere by the demonstration that an acute treatment with morphine (6 mg kg -1 ) decreased protein concentration in saliva from the rat submandibular gland [14]. Considering that sympathetic activity on salivary glands is the most important control of salivary protein secretion and pointing out that the presence of opioid receptors in submandibular and parotid glands was never demonstrated, the CGP or morphine effects to reduce protein concentration in saliva could be due a direct interference of cAMP in cells that express opioid receptors. Moreover, the tolerance to opioid receptors changes the pathway signal transduction by the cAMP-dependent protein kinase [38,39]. Thus, if opioid receptors are expressed in salivary glands, probably morphine and CGP reduces the cAMP directly in these glands, which is a key mechanism that may be involved in reduced protein secretion in saliva. However, another explanation for morphine and CGP inhibitory effects on salivary protein concentration may be due to the opioid receptors presence that might have inhibited preganglionic or ganglionic sympathetic nerve projecting to salivary gland [40][41][42]. In both hypotheses, it seems clear that our study indicates inhibition of effects promoted by CGP and morphine by interaction with opioid receptors.
The reduction in salivary protein concentration and in Amide I/Amide II of submandibular promoted by opioid agonists is a characteristic of tissues that have low sympathetic activity and/or low glucose utilization, likely because they need to have a low energy status [43]. Besides, we also showed that both treatments are able to reduce glycogen, indicating influence of opioid system in glycogen metabolism on submandibular glands [44]. Specifically, the present study describes evacuated spaces (previously occupied by secretory acini) between acinar and ductal cells in glandular parenchyma of submandibular glands under morphine-and CGP-treatment. The fast (1h) effect of morphine and CGP suggests that it is not solely a consequence of reduction of parasympathetic/sympathetic activity. This is in agreement with morphine-induced apoptosis of murine J774 cells mediated through TGF-beta [45]. Notably, this change is reinforced in morphine-treated mice. Considering the acute morphological changes with morphine and CGP, we can still propose induction of more profound morphological changes during chronic treatment. However, these effects could be promoted by the activation of kappa-, delta-, and/or mu-opioid-receptor by morphine and CGP in submandibular ganglion [18,24,46].
The CH 2 reduction in lipids [47] after morphine and CGP treatments suggests a decrease in lipid rafts that translocate proteins from intracellular structures to the plasma membrane [48], as well as decreasing energy status. Opioid agonists reduced PO 2 asymmetric of phospholipids, indicating damage in plasma membrane, which is also an apoptotic characteristic [49].
Considering the similar changes in salivary secretion and submandibular composition promoted by both therapeutic agents, we can consider that CGP have similar inhibitory effect on autonomic activity to salivary glands as described to morphine [18]. The CGP antinociceptive effect was antagonized by naloxone, a non-selective opioid receptor antagonist, suggesting that CGP effect may also occur through opioid receptor at the supraspinal level [24,50]. CGP increased neuronal activity in midbrain periaqueductal gray (PAG), a key relay station in the processing of nociceptive information in central nervous system [24,51] and that is interconnected with the hypothalamus [52]. Opioid receptors are spread in the hypothalamus [53] and these hypothalamic neurons may have inhibitory projections to superior salivary nucleus, from where ganglionic fibers of parasympathetic nervous system spread to submandibular glands [54][55][56]. Morphine also activates PAG [57], which suggest similar repercussion of CGP.
The muscarinic receptors sensitivity s in salivary glands was not reduced by morphine [19]. Therefore, we evaluated whether the salivary secretion reduction after treatment with CGP and morphine could be explained by acetylcholine decreased levels in the extracellular fluid, which could be demonstrated by acetylcholinesterase enzyme increased activity. Previous studies have shown an increase in brain acetylcholinesterase expression, 30 minutes after morphine injection (10 mg kg -1 of morphine), indicating a higher enzyme activity and further acetylcholine degradation. [58]. However, another study showed that morphine chronic administration decreases acetylcholinesterase activity in the midbrain [59]. It is noteworthy that the acute effect of morphine and CGP on acetylcholinesterase activity in salivary gland has never been reported. Bearing in mind that the expected reduction of acetylcholine in synaptic cleft due to lower parasympathetic activity [18] is associated with similar acetylcholinesterase activity in submandibular glands after morphine and CGP treatment, it is expected that the acetylcholine presence in synaptic cleft can be further reduced due to the acetylcholine/ acetylcholinesterase ratio. Furthermore, these data emphasize that the reduction of pilocarpine-induced salivary secretion by morphine and CGP is likely to be prejunctional. Considering the present results and previous reports, we suggest a central and autonomic-pathway leading to changes in submandibular gland and hyposalivation by morphine and CGP (Fig 6).
To the best of our knowledge, this is the first report that demonstrates the salivary hyposecretion elicited by CGP, which is mediated by the opioid receptors' system and such effect was reversed by naloxone. The present study also provides new evidence for an inhibitory morphine effect in pilocarpine-salivary secretion mediated by the opioid system in mice. Morphine and CGP also reduced salivary protein concentration and increased the lactate in submandibular gland and naloxone reverted both alterations. Morphine and CGP also reduced several infrared vibrational modes representing Amide I, Amide II, CH 2 /CH 3 , C = O, PO 2 asymmetric and PO 2 symmetric, which was blocked by pre-treatment with naloxone. To confirm the pathway to CGP effects, the in silico docking analysis demonstrate the polar contacts interaction between the CGP and opioid receptor Cys219 residue. Altogether, we showed that salivary hypofunction and glandular changes elicited by CGP may occur through opioid receptor suggesting that opioid receptors blockage in superior cervical ganglion and submandibular ganglion may be a possible strategy to restore salivary secretion while maintaining antinociceptive effects due its effects in central nervous system.  Arrows indicate the salivary and submandibular gland contents regulations ("increase, #decrease) under morphine or CGP treatments. Blue line indicates the inhibitory effect of morphine and CGP to autonomic pathway to submandibular glands. Red and black lines indicate the sympathetic and parasympathetic pathway to submandibular gland. Opioid receptors are expressed in PAG [51], hypothalamus [52,53], submandibular ganglia [18,46] and superior cervical [41,42] ganglia. CGP, cyclo-Gly-Pro.