Translational Value of the Transdermal Administration of Bergamot Essential Oil and of Its Fractions

The essential oil of bergamot (BEO) has consistently proven antinociceptive and antiallodynic properties. Accordingly, the analgesic efficacy of the decolored essential oil (DEC), with higher levels of limonene, and the deterpenated (DET) fraction, with higher levels of linalool and linalyl acetate, was investigated using a formalin test after inhalation. The present study was aimed at characterizing the effects of BEO, its components with the highest pharmacological activity (represented by linalool, limonene, and linalyl acetate), and its DEC and DET fractions on the formalin test after transdermal administration relevant to clinical translation through topical application. To this aim, the schedule of intervention involved administration immediately after formalin injection or as a 5 min pretreatment followed by washout in ddY-strain mice. This study demonstrates, for the first time, the significant analgesic effect of all three constituents in the first and second phases, accounting for the efficacy of the essential oil in the formalin test. While all fractions revealed equal activity toward the phytocomplex in the early phase, the reduction in time of licking/biting during the late phase was more markedly induced by DEC. Moreover, pretreatment with BEO and its fractions followed by washout did not produce a significant reduction in licking/biting time in both phases of formalin-induced nociceptive response.


Introduction
The essential oil of bergamot (BEO) is obtained by cold pressing of the epicarp and, partly, of the mesocarp of the fresh fruit of bergamot (Citrus bergamia Risso et Poiteau). Of particular interest is its most abundant fraction (93-96%), i.e., the volatile one including oxygenated compounds, such as linalool and linalyl acetate, and monoterpenes and sesquiterpenes, such as limonene [1,2], known to be the most pharmacologically active components.

Animals
The animals used for this research study were male mice ddY (Japan SLC, Hamamatsu, Japan) of 23-26 g of weight. They were housed in individual cages under a 12 h light/12 h dark cycle at room temperature of 22-24 • C with 55% ± 5% relative humidity. The mice were provided with food and water ad libitum. The experiments followed the Ethical Standards for Investigation of Experimental Pain in Animals Guidelines, and they were approved by the Animal Care and Use of Tohoku Medical and Pharmaceutical University Committee from 8 January 2014 for minimizing the suffering of animals and for using the minimum number of animals necessary for reliable results. In agreement with G*power sample size calculation [33], according to previous studies [29,31], the minimum number of animals to achieve reliable statistically significant results was n = 8 per experiment.

Pain Model
The experimental pain model selected was the formalin test, since it consists of an early phase due to the nociceptive stimulation followed by a late phase of pain behavior induced by central sensitization and affected by aging [22]. The test required 1 h of habituation in a transparent cage sized 22.0 × 15.0 × 12.5 cm. After this period of acclimatation, a volume of 20 µL of formalin (2% in saline) was administered through intraplantar (i.pl.) injection to the mice, using a microsyringe with a 26-gauge needle. The time of licking/biting, which was the pain behavioral indicator considered, was monitored with a handheld stopwatch in 5 min intervals. The early phase started immediately after the i.pl. injection of formalin and lasted for 10 min (0-10 min), while the late phase began at the end of the first 10 min and lasted up to 30 min.

Experimental Design
The effects of transdermal administration of the entire phytocomplex BEO and of its components linalool, limonene, and linalyl acetate, as well as of the DEC and DET fractions, on licking/biting due to formalin injection were investigated through a double-intervention protocol. All the fractions were defurocoumarinized to avoid phototoxicity. As the purpose of the present research was to assess the anti-nocifensive efficacy of BEO and of its DEC and DET fractions, as well as of its main volatile components when administered transdermally in a pain model relevant to clinic, two points of administration were established: (a) in the first experimental group, the intervention (BEO/DEC/DET) was applied on the plantar surface immediately after (A) formalin injection (BEO-A/DEC-A/DET-A); (b) in the second experimental setting, the duration of the effect of the intervention after washout was examined by applying the intervention (BEO/DEC/DET) on the plantar surface 5 min before (PRE) formalin administration (BEO-PRE/DEC-PRE/DET-PRE), removing it through absorbent paper just before the i.pl. injection of the algogen compound, followed by the transdermal application of Jojoba oil (JOJ), i.e., the vehicle. The third experimental group received the transdermal administration of a solution of 44% D-limonene (equal to the percentage present in the DEC fraction)/39.8% linalool (equal to the percentage contained in the DET fraction)/44.5% linalyl acetate (equal to the percentage contained in the DET fraction) in JOJ immediately after i.pl. formalin injection. The volume of all the interventions and of the vehicle administered was 10 µL. The experimental schedule is illustrated in Figure 1.

Experimental Design
The effects of transdermal administration of the entire phytocomplex BEO and of its components linalool, limonene, and linalyl acetate, as well as of the DEC and DET fractions, on licking/biting due to formalin injection were investigated through a doubleintervention protocol. All the fractions were defurocoumarinized to avoid phototoxicity. As the purpose of the present research was to assess the anti-nocifensive efficacy of BEO and of its DEC and DET fractions, as well as of its main volatile components when administered transdermally in a pain model relevant to clinic, two points of administration were established: (a) in the first experimental group, the intervention (BEO/DEC/DET) was applied on the plantar surface immediately after (A) formalin injection (BEO-A/DEC-A/DET-A); (b) in the second experimental setting, the duration of the effect of the intervention after washout was examined by applying the intervention (BEO/DEC/DET) on the plantar surface 5 min before (PRE) formalin administration (BEO-PRE/DEC-PRE/DET-PRE), removing it through absorbent paper just before the i.pl. injection of the algogen compound, followed by the transdermal application of Jojoba oil (JOJ), i.e., the vehicle. The third experimental group received the transdermal administration of a solution of 44% D-limonene (equal to the percentage present in the DEC fraction)/39.8% linalool (equal to the percentage contained in the DET fraction)/44.5% linalyl acetate (equal to the percentage contained in the DET fraction) in JOJ immediately after i.pl. formalin injection. The volume of all the interventions and of the vehicle administered was 10 µL. The experimental schedule is illustrated in Figure 1.

Statistical Analysis
Data were reported as the mean ± SEM of the time of licking/biting and evaluated statistically for differences by one-or two-way ANOVA, followed by Bonferroni's test, considering p < 0.05 as statistically significant.

Analgesic Efficacy of the Transdermal Topical Treatment with BEO on Formalin-Induced Licking/Biting Biphasic Nocifensive Behavior
The transdermal administration of 10 µL of the whole phytocomplex BEO immediately after the i.pl. injection of 2% formalin solution (BEO-A) induced a significant reduction in the licking/biting time in both the first and the second phase of the nociceptive response during the formalin test (p < 0.001). On the other hand, the pretreatment, consisting of the topical application of 10 µL of pure BEO on the paw site of the injection 5 min prior to the i.pl administration of 2% formalin, followed by the removal

Statistical Analysis
Data were reported as the mean ± SEM of the time of licking/biting and evaluated statistically for differences by one-or two-way ANOVA, followed by Bonferroni's test, considering p < 0.05 as statistically significant.

Analgesic Efficacy of the Transdermal Topical Treatment with BEO on Formalin-Induced Licking/Biting Biphasic Nocifensive Behavior
The transdermal administration of 10 µL of the whole phytocomplex BEO immediately after the i.pl. injection of 2% formalin solution (BEO-A) induced a significant reduction in the licking/biting time in both the first and the second phase of the nociceptive response during the formalin test (p < 0.001). On the other hand, the pretreatment, consisting of the topical application of 10 µL of pure BEO on the paw site of the injection 5 min prior to the i.pl administration of 2% formalin, followed by the removal of BEO and subsequent application of JOJ (BEO-PRE), did not produce a significant reduction in the licking/biting time in the first and second phase of the nociceptive response induced by formalin. These data are reported in Figure 2. of BEO and subsequent application of JOJ (BEO-PRE), did not produce a significant reduction in the licking/biting time in the first and second phase of the nociceptive response induced by formalin. These data are reported in Figure 2. . This effect did not persist after the 5 min pretreatment followed by washout (BEO-PRE). Overall licking/biting time is expressed as time in s ± SEM during the early phase (EP) (0-10 min) and the late phase (LP) (10-30 min). One-way ANOVA followed by Bonferroni test; n = 8; *** p < 0.001.

Analgesic Effect of BEO, DEC and DET Fractions on Formalin Test after Transdermal Administration
Similarly to what was observed for the entire phytocomplex, the topical application of 10 µL of DEC fraction solution immediately following the i.pl. injection of formalin (DEC-A) produced a significant reduction in the licking/biting time in both the early and the late phase (EP, LP) of the formalin test (p < 0.001). In this case, the topical pretreatment 5 min before formalin administration and followed by washout and vehicle application did not evoke a significant decrease in the time of licking/biting in any of the two phases of the formalin test. The results of DEC transdermal administration are displayed in Figure 3. on licking/biting behavior in the formalin test. The essential oil of bergamot (BEO) reduced licking/biting behavior when administered immediately after formalin injection in the early and in the late phase of formalin test (BEO-A). This effect did not persist after the 5 min pretreatment followed by washout (BEO-PRE). Overall licking/biting time is expressed as time in s ± SEM during the early phase (EP) (0-10 min) and the late phase (LP) (10-30 min). One-way ANOVA followed by Bonferroni test; n = 8; *** p < 0.001.

Analgesic Effect of BEO, DEC and DET Fractions on Formalin Test after Transdermal Administration
Similarly to what was observed for the entire phytocomplex, the topical application of 10 µL of DEC fraction solution immediately following the i.pl. injection of formalin (DEC-A) produced a significant reduction in the licking/biting time in both the early and the late phase (EP, LP) of the formalin test (p < 0.001). In this case, the topical pretreatment 5 min before formalin administration and followed by washout and vehicle application did not evoke a significant decrease in the time of licking/biting in any of the two phases of the formalin test. The results of DEC transdermal administration are displayed in Figure 3.  The DET fraction of BEO, enriched in linalyl acetate, administered after formalin injection (DET-A) significantly reduced the nocifensive response of licking/biting in the first (p < 0.01) and in the second phase (p < 0.05) of the formalin test. It is interesting to notice that this proved to be the least pharmacologically active fraction in terms of reversal of licking/biting behavior in formalin test. Moreover, as described for the whole The decolored fraction of the essential oil of bergamot (DEC), enriched in D-limonene, produced a reduction in the licking/biting behavior when administered immediately after formalin injection in the early and in the late phase (EP, LP) of the formalin test (DEC-A). The latter analgesic effect did not last after the 5 min pretreatment followed by washout (DEC-PRE). Overall licking/biting time is expressed as time in s ± SEM during the EP (0-10 min) and the LP (10-30 min). One-way ANOVA followed by Bonferroni test; n = 8; *** p < 0.001. The DET fraction of BEO, enriched in linalyl acetate, administered after formalin injection (DET-A) significantly reduced the nocifensive response of licking/biting in the first (p < 0.01) and in the second phase (p < 0.05) of the formalin test. It is interesting to notice that this proved to be the least pharmacologically active fraction in terms of reversal of licking/biting behavior in formalin test. Moreover, as described for the whole phytocomplex and for the DEC fraction, the effect of 5 min pretreatment with DET waned after its removal and replacement with JOJ vehicle. The results are reported in Figure 4. Figure 3. Analgesic effect of the transdermal administration of the decolored fraction of the essential oil of bergamot (DEC), enriched in D-limonene, on licking/biting behavior in the formalin test. The decolored fraction of the essential oil of bergamot (DEC), enriched in D-limonene, produced a reduction in the licking/biting behavior when administered immediately after formalin injection in the early and in the late phase (EP, LP) of the formalin test (DEC-A). The latter analgesic effect did not last after the 5 min pretreatment followed by washout (DEC-PRE). Overall licking/biting time is expressed as time in s ± SEM during the EP (0-10 min) and the LP (10-30 min). One-way ANOVA followed by Bonferroni test; n = 8; *** p < 0.001.
The DET fraction of BEO, enriched in linalyl acetate, administered after formalin injection (DET-A) significantly reduced the nocifensive response of licking/biting in the first (p < 0.01) and in the second phase (p < 0.05) of the formalin test. It is interesting to notice that this proved to be the least pharmacologically active fraction in terms of reversal of licking/biting behavior in formalin test. Moreover, as described for the whole phytocomplex and for the DEC fraction, the effect of 5 min pretreatment with DET waned after its removal and replacement with JOJ vehicle. The results are reported in Figure 4.

Effects of the Transdermal Administration of D-limonene, Linalool, and Linalyl Acetate on Formalin-Induced Licking/Biting Behavior
To unveil the contribution of the main individual constituents of the volatile fraction of BEO to the phytocomplex analgesic pharmacological activity when transdermally administered, the effects of D-limonene, linalool, and linalyl acetate were analyzed on the first and on the second phase of the formalin test and throughout the time course. The effects of D-limonene were tested by applying 10 µL of a solution of D-limonene at 44% (equal to the percentage of D-limonene in the DEC fraction) diluted in JOJ on the paw site of the injection, immediately after i.pl. injection of formalin. D-Limonene induced a significant reduction in the licking/biting time in the first (p < 0.001) and the second (p < 0.05) phase of the nociceptive response induced by the i.pl. formalin ( Figure 5). To investigate the analgesic efficacy of the topical application, a volume of 10 µL of a 39.8% linalool solution (corresponding to the percentage of linalool present in the DET fraction), immediately after formalin, produced a significant decrease in the licking/biting time in the first (p < 0.001) and the second phase (p < 0.01) ( Figure 5). Similarly, the transdermal administration of 10 µL of a solution of linalyl acetate at 44.5% (equal to the percentage of linalyl acetate present in the DET fraction) in JOJ immediately following the i.pl. injection of formalin produced a statistically significant reduction (p < 0.001) in the licking/biting time in both the early and the late phase of the nocifensive behavior induced by the i.pl. injection of formalin ( Figure 5). According to the gathered results, linalyl acetate was the most active, linalool solution (corresponding to the percentage of linalool present in the DET fraction), immediately after formalin, produced a significant decrease in the licking/biting time in the first (p < 0.001) and the second phase (p < 0.01) ( Figure 5). Similarly, the transdermal administration of 10 µL of a solution of linalyl acetate at 44.5% (equal to the percentage of linalyl acetate present in the DET fraction) in JOJ immediately following the i.pl. injection of formalin produced a statistically significant reduction (p < 0.001) in the licking/biting time in both the early and the late phase of the nocifensive behavior induced by the i.pl. injection of formalin ( Figure 5). According to the gathered results, linalyl acetate was the most active, followed by linalool and limonene in the late phase, while all these constituents were almost equally effective in the early phase. Figure 5. Analgesic effect of D-limonene, linalool and linalyl acetate on licking/biting behavior in the formalin test. The main pharmacologically active components of the volatile fraction of the essential oil of bergamot (D-limonene, linalool, and linalyl acetate), transdermally administered immediately after formalin injection, significantly decreased the licking/biting time in the early and in the late phase (EP, LP) of the formalin test. All these constituents were almost equally effective in the EP, while linalyl acetate was the most active followed by linalool and limonene in the LP. Overall licking/biting time is expressed as time in s ± SEM during the EP (0-10 min) and the LP (10-30 min). One-way ANOVA followed by Bonferroni test; n = 8; * p < 0.05, ** p < 0.01, *** p < 0.001.
The time course showed a marked and significant reduction (p < 0.001) in the licking/biting behavior observed between 0 and 20 min post injection by limonene ( Figure  6). Moreover, linalool particularly exerted its analgesic efficacy at all time intervals between 0 and 25 min post injection ( Figure 6). Linalyl acetate produced a noteworthy Figure 5. Analgesic effect of D-limonene, linalool and linalyl acetate on licking/biting behavior in the formalin test. The main pharmacologically active components of the volatile fraction of the essential oil of bergamot (D-limonene, linalool, and linalyl acetate), transdermally administered immediately after formalin injection, significantly decreased the licking/biting time in the early and in the late phase (EP, LP) of the formalin test. All these constituents were almost equally effective in the EP, while linalyl acetate was the most active followed by linalool and limonene in the LP. Overall licking/biting time is expressed as time in s ± SEM during the EP (0-10 min) and the LP (10-30 min). One-way ANOVA followed by Bonferroni test; n = 8; * p < 0.05, ** p < 0.01, *** p < 0.001.
The time course showed a marked and significant reduction (p < 0.001) in the licking/biting behavior observed between 0 and 20 min post injection by limonene ( Figure 6). Moreover, linalool particularly exerted its analgesic efficacy at all time intervals between 0 and 25 min post injection ( Figure 6). Linalyl acetate produced a noteworthy significant decrease in the licking/biting time up to 25 min after formalin administration (p < 0.001) ( Figure 6).

Discussion
The present study demonstrates, for the first time, the analgesic efficacy of the transdermal administration of DEC and DET phytocomplexes and of the main components of BEO endowed with pharmacologic activity, i.e., D-limonene, linalool, and linalyl acetate, on the biphasic behavior of licking/biting induced by 2% formalin. While all the fractions were equally active to the whole phytocomplex in the early phase, the DEC fraction induced a more marked decrease in the nocifensive behavior represented by the time of licking/biting during the late phase. This is in agreement with findings

Discussion
The present study demonstrates, for the first time, the analgesic efficacy of the transdermal administration of DEC and DET phytocomplexes and of the main components of BEO endowed with pharmacologic activity, i.e., D-limonene, linalool, and linalyl acetate, on the biphasic behavior of licking/biting induced by 2% formalin. While all the fractions were equally active to the whole phytocomplex in the early phase, the DEC fraction induced a more marked decrease in the nocifensive behavior represented by the time of licking/biting during the late phase. This is in agreement with findings previously observed for inhalatory administration, under which the highest volume (800 µL) of the DEC fraction was as effective as comparable volumes of the whole BEO [31], while even the highest volume of the DET fraction resulted less effective [31]. Moreover, all the constituents, i.e., D-limonene, linalool, and linalyl acetate, were almost equally effective in the early phase, while linalyl acetate was the most active followed by linalool and limonene in the late phase. This is inversely correlated with the effects of the latter constituents on spontaneous motor activity [31], linked to the capability of BEO to increase the immobility time in the open field and the forced swimming behavioral tasks [34], representing the attempt to escape anxiety [35]. Interestingly, all three main components exerted a significant analgesic effect in both the first and the second phase, accounting for the efficacy of the whole essential oil in this experimental pain model, selected precisely because it is characterized by a late phase of pain behavior induced by central sensitization and affected by aging [22]. In fact, the demonstration of the analgesic properties of the different fractions and of BEO components after transdermal administration is necessary for clinical use since BEO in the pharmaceutical form of a base cream incorporating a nanotechnological delivery system based on solid lipid nanoparticles [36] is under investigation in an active clinical trial for proof of concept of the management of severe dementia with agitation through analgesia (NCT04321889) [37]; this is fundamental since clinical studies for the treatment of several forms of pain often exclude cognitively impaired patients [38,39], not testing the most novel therapies [40] and not informing about the best analgesic practice in this fragile population subjected to physiological modifications and variability in drug response [41]. The DEC fraction was enriched with 44% limonene, confirming the previous observation supporting the role of D-limonene in analgesia. In particular, D-limonene was proven to reduce vimentin levels similarly to the autophagy inhibitor chloroquine [42], and this is important since alterations of the autophagic machinery have been widely reported to occur during pain conditions relevant to the clinic [43] and during dementia, supporting a role for drug repositioning [44]. In fact, agitation is often triggered by underdiagnosed and unrelieved pain [45], being treated with antipsychotics that increase death risk in the aged patients due to accidents of a cardiac and cerebrovascular nature [46], and this issue has worsened during the pandemic [47,48]. Another important feature of the pharmacological activity of BEO for clinical use is its lack of sedative effects (that could increase cognitive deterioration), supported by the involvement of serotonergic neurotransmission [49]. This neuromodulation is a fundamental target during aging and neurodegeneration [50], as demonstrated by the action of rapid-acting antidepressants [51], which could represent a revolutionary therapy for the neuropsychiatric symptoms [52]. Therefore, the deep characterization of the analgesic properties of BEO, as well as of its fractions and components, via the transdermal route is fundamental since its engineered nanotechnological delivery system, NanoBEO, is administered via the same route and is the soundest candidate essential oil for investigation in the clinical management of pain and agitation. From a pharmacokinetic perspective, the effect of 5 min pretreatment of the whole phytocomplex, as well as of the DEC and DET fraction, waned after removal and replacement with JOJ vehicle. In fact, the latter did not persist after the 5 min pretreatment followed by washout, supporting the need for the topical exposure to last over 5 min. Future studies will investigate the pharmacokinetics and herb-drug interactions [53] of NanoBEO, fundamental in case of polypharmacy [54], noteworthy for the elderly [55]. In fact, according to our findings, the pretreatment consisting of the transdermal administration of BEO or of its enriched fractions on the paw site of the injection 5 min before formalin injection, followed by washout, did not produce a statistically significant decrease in the formalin-induced licking/biting time both in the early and in the late phase. Therefore, the investigation of the time of exposure to the phytocomplex topically applied required for a long-lasting