Impact of Oral Intake of Glucosylceramide Extracted from Pineapple on Xerostomia: A Double-Blind Randomized Cross-Over Trial

Background: The aim of this double-blind randomized cross-over trial was to evaluate the effect of oral intake of glucosylceramide extracted from pineapple on oral moisture and xerostomia symptoms. Methods: Sixteen participants who had xerostomia symptoms were randomly allocated into two groups. One group received, as test samples, tablets containing glucosylceramide extracted from pineapple (GCP) followed by placebo tablets. The other group received the test samples in the reverse order. Participants were instructed to take tablets of the first test sample once a day (after breakfast) for two consecutive weeks. Then, after a washout period of four weeks, participants were instructed to take the other test sample for two consecutive weeks. The oral moisture level of the lingual mucosa, xerostomia symptoms, and the number of fungiform papillae was evaluated. Results: The oral moisture significantly increased, and the visual analog scale (VAS) of “How is the dryness of your mouth?” significantly improved after GCP tablets intake and not after placebo tablets intake. The number of fungiform papillae was not significantly different following the intake of GCP tablets or placebo tablets. Conclusion: Results suggested that oral intake of GCP may improve the moisture level and xerostomia symptoms.


Introduction
Xerostomia can be caused by radiation therapy, Sjogren's syndrome, and multidrug therapy in older patients [1,2]. Xerostomia can increase the risk of opportunistic infections, dental caries, periodontal disease, decreased denture retention, and traumatic ulcers, resulting in a negative effect on quality of life [1,2].
For xerostomia resulting from radiotherapy or Sjogren's syndrome, pilocarpine and cevimeline are prescribed to promote saliva secretion [1,2]. However, these drugs can cause adverse side effects such as hyperhidrosis and dyspepsia [1,2]. Therefore, as a symptomatic treatment for xerostomia, oral moisturizers can be prescribed [1][2][3]. However, these products still need further improvement in terms of the prolonged duration of treatment effect and cost effectiveness [2,3].

Participants
The participants were individuals who had subjective symptoms of xerostomia selected among the patients undergoing follow up visits at the Kagoshima University hospital between 7 July 2016 and 31 January 2019. Exclusion criteria were: (a) a history of radiation therapy or chemotherapy; (b) a history of food allergy, including pineapple; (c) a schedule of oral surgical operations, including extraction of teeth, during the research period; and (d) age below 20 years. Each participant received written and oral descriptions of the experimental protocols and provided written informed consent prior to enrollment.
We estimated the required sample size by using G* power 3.1.9.4 [17] assuming, on the basis of a previous study, that a medium effect size of 0.7 points for a mean difference in visual analog scale (VAS) score between with and without xerostomia symptoms could be detected [18]. The results revealed that n = 16 was the minimum sample size (alpha, 0.05; beta, 0.05 (95% power); two sides; dependent samples).

Test Sample
The test samples used in this study were tablets with GCP (GCP tablet) and without GCP (placebo tablet). Composition of the GCP tablet (500 mg) was: carbohydrate (0.2 g), and GCP (1.2 mg), for a total of 0.4 kcal. The composition of the placebo tablet (500 mg) was: carbohydrate (0.2 g), for a total of 0.4 kcal [12,13]. The GCP and placebo tablets were indistinguishable in terms of taste, flavor, and appearance including color, size, and packaging.

Assessment
The oral moisture level, xerostomia symptoms, and number of fungiform papillae were evaluated before (G1 and P1) and after (G2 and P2) taking the GCP and placebo test sample for two weeks, respectively. G1: Moisture level measured before taking GCP tablets, G2: Moisture level Figure 1. G1: Moisture level measured before taking GCP tablets, G2: Moisture level measured after taking GCP tablets, P1: Moisture level measured before taking Placebo tablets, P2: Moisture level measured after taking Placebo tablets. Flow chart for our randomized crossover trial. (G1 and G2) and (P1 and P2) are before and after taking the glucosylceramide extracted from pineapple (GCP) and placebo test sample, respectively.

Participants
The participants were individuals who had subjective symptoms of xerostomia selected among the patients undergoing follow up visits at the Kagoshima University hospital between 7 July 2016 and 31 January 2019. Exclusion criteria were: (a) a history of radiation therapy or chemotherapy; (b) a history of food allergy, including pineapple; (c) a schedule of oral surgical operations, including extraction of teeth, during the research period; and (d) age below 20 years. Each participant received written and oral descriptions of the experimental protocols and provided written informed consent prior to enrollment.
We estimated the required sample size by using G* power 3.1.9.4 [17] assuming, on the basis of a previous study, that a medium effect size of 0.7 points for a mean difference in visual analog scale (VAS) score between with and without xerostomia symptoms could be detected [18]. The results revealed that n = 16 was the minimum sample size (alpha, 0.05; beta, 0.05 (95% power); two sides; dependent samples).

Test Sample
The test samples used in this study were tablets with GCP (GCP tablet) and without GCP (placebo tablet). Composition of the GCP tablet (500 mg) was: carbohydrate (0.2 g), and GCP (1.2 mg), for a total of 0.4 kcal. The composition of the placebo tablet (500 mg) was: carbohydrate (0.2 g), for a total of 0.4 kcal [12,13]. The GCP and placebo tablets were indistinguishable in terms of taste, flavor, and appearance including color, size, and packaging.

Assessment
The oral moisture level, xerostomia symptoms, and number of fungiform papillae were evaluated before (G1 and P1) and after (G2 and P2) taking the GCP and placebo test sample for two weeks, respectively. G1: Moisture level measured before taking GCP tablets, G2: Moisture level measured after taking GCP tablets, P1: Moisture level measured before taking Placebo tablets, P2: Moisture level measured after taking Placebo tablets. The evaluation methods are explained in detail below.

Oral Moisture Level
Measurements of oral moisture level were performed at the lingual mucosa by using an oral moisture-checking device (Mucus; Life Co., Ltd., Saitama, Japan). The measurements were performed 2 h after a meal, while the subject was in a stress-free environment [18,19]. The oral moisture level was measured three consecutive times, and the median was used as a representative value following a previous study [18].

Number of Fungiform Papillae
To assess the number of the fungiform papillae, a portable microscope with a 30× polarization/ non-polarization lens (Scalar M3, Scalar, Tokyo, Japan) and a wooden tongue depressor (Star tongue depressor, Hoshiseido, Tokyo, Japan) were used. Participants were instructed to place their heads on the headrest of the dental chair, pull their tongue out to the extent that they did not feel pain, and hold their tongue firmly with their upper and lower lips. The assessor lightly placed the wooden tongue depressor on the participant's tongue surface and drew the shape of the anterior margin of tongue on the tongue depressor surface. Based on a point at a distance of 10 mm from the center of the anterior margin of tongue drawn on the wooden tongue depressor, a square hole with sides of 4-mm length was made in the tongue depressor. The assessor placed the tongue depressor on the tongue again by using the drawn shape of the anterior margin of tongue as a reference, and took an image 4-mm length x 4-mm width with the microscope. The captured images were saved as JPEG files on a PC (Let's Note SX2, Panasonic Corporation, Osaka, Japan), and the number of fungiform papillae with red capillary vessels was counted on these images [20]. The measurement was repeated three times, and the mean value was calculated.

Xerostomia Symptoms
Xerostomia symptoms, such as the degree of dryness in the mouth, the amount of saliva, the taste of the meal, and the satisfaction of oral moisture [1,2] were surveyed by using a 100-mm visual analog scale (VAS) by using the following questions: "How is the dryness of your mouth?", "How is the secretion of saliva?", "How was the taste of the meal?", and "Are you satisfied with the moisture of your mouth?". The worst condition was mapped at 0 mm, and the best condition was mapped at 100 mm on the VAS scale.

Statistical Analysis
Possible differences between the two groups at the baseline were assessed by using the t-test, the chi-square test, and the Mann-Whitney U-test. Possible carryover effects were analyzed by using the Wilcoxon signed-rank test for each of the measures taken in G1 and P1 in Sequence 1 and the measures taken in P1 and G1 in Sequence 2. The treatment effect was evaluated by analyzing each of the measures taken before and after taking GCP tablets and placebo tablets (i.e., G1 and G2 values, P1 and P2 values) by using the Wilcoxon signed-rank test. Statistical significance was set at α = 0.05. Data were analyzed by using SPSS v19 (IBM Japan Ltd., Tokyo, Japan).

Results
In total, 16 participants were randomized in the two groups and assessed from July 2016 to January 2019. Follow-up evaluations were completed in March 2019. No side effects caused by the intake of GCP tablets or placebo tablets were observed in any of the 16 participants. Table 1 outlines the baseline characteristics of the study participants in each group. There were no significant differences in age, number of remaining teeth, use of removable prostheses, medications, and oral moisture level at the baseline between the two groups, whereas the gender distribution was significantly different between the two groups.  Table 2 shows the results of the moisture level of the lingual mucosa. The upper part of the table shows the median, minimum, maximum, 25th and 75th percentile values, and 95% confidence interval (CI) of the moisture level at each evaluation point. The Wilcoxon signed-rank test revealed that there were no statistically significant differences in oral moisture levels measured in G1 and P1 in Sequence 1 and in P1 and G1 in Sequence 2, suggesting no carryover effects. The lower part of the table shows the median, minimum, maximum, 25th, and 75th percentile, and 95% CI of the moisture level before and after taking GCP and before and after taking placebo tablets. The Wilcoxon signed-rank test revealed that the moisture level measured after taking GCP tablets (in G2) was significantly higher than the one measured before (in G1), whereas no significant differences were observed after taking placebo tablets and (P2 versus P1).  Table 3 shows the results of the number of fungiform papillae. The upper part of the table shows the median, minimum, maximum, 25th and 75th percentile values, and 95% CI of the number of fungiform papillae at each evaluation point. The Wilcoxon signed-rank test revealed that there were no statistically significant difference in the number of fungiform papillae measured in G1 and P1 in Sequence 1 and in P1 and G1 in Sequence 2, suggesting no carryover effects. The lower part of the table shows the median, minimum, maximum, 25th and 75th percentile values, and 95% CI of the number of fungiform papillae measured before and after taking GCP tablets and placebo tablets. The Wilcoxon signed-rank test revealed that there were no statistically significant differences in the number of fungiform papillae measured before and after taking GCP tablets and placebo tablets.  Table 4 shows the results of the VAS values for a question "How is the dryness of your mouth?". The upper part of the table shows the median, minimum, maximum, 25th and 75th percentile values, and 95% CI of the VAS values at each evaluation point. The Wilcoxon signed-rank test revealed that there were no statistically significant difference in the VAS values measured in G1 and P1 in Sequence 1 and in P1 and G1 in Sequence 2, suggesting no carryover effects. The lower part of the table shows the median, minimum, maximum, 25th and 75th percentile values, and 95% CI of the VAS values measured before and after taking GCP tablets and placebo tablets. The Wilcoxon signed-rank test revealed that the VAS values measured after taking GCP tablets (in G2) were significantly higher than those measured before (in G1), whereas no significant differences in VAS values were observed after taking placebo tablets (P2 versus P1). Similarly as in Table 4, Table 5, Table 6, and Table 7 show the results of the VAS values for the following questions: "How is the secretion of saliva?" (Table 5), "How was the taste of the meal?" (Table 6), and "Are you satisfied with the moisture of your mouth?" (Table 7). Similarly as in Table 4, here, the Wilcoxon signed-rank test revealed that there were no statistically significant differences in any of the VAS values measured in G1 and P1 in Sequence 1 and in P1 and G1 in Sequence 2, suggesting no carryover effects. In addition, the Wilcoxon signed-rank test revealed that there were no statistically significant differences among any of the VAS values measured before and after taking GCP tablets and placebo tablets.

Discussion
In this randomized cross-over test on patients with xerostomia symptoms, the effects of oral intake of GCP on oral moisture and xerostomia symptoms were compared with that of the ingestion of placebo tablets. As a result, the oral moisture significantly increased, and the visual analog scale (VAS) of "How is the dryness of your mouth?" significantly improved after GCP tablets intake and not after placebo tablets intake. The cross-over design used in this study has the advantages that the variation in data is smaller than that of a group comparison test, and the number of cases for obtaining results can be reduced. Based on previous studies [12,13], we set an eight-week study period (two-week intake and four-week washout) to ensure sufficient safety and an adequate washout period. It is widely known that carryover effects may occur in cross-over design studies. In these kinds of studies, it is important to assess whether the intervention effect is carried over, as it may occur even if a given washout period is set. Orally ingested glucosylceramide is absorbed in the intestine and taken up by epithelial cells [10]. Considering that the turnover rate of cells in the epithelial tissue of the tongue is about seven days [21], a safe washout period of four weeks was set in this study. In fact, results showed that there were no statistically significant differences in any of the measures here considered between the G1 and P1 values in Sequence 1 and the P1 and G1 values in Sequence 2, suggesting that the washout period here used was sufficient to avoid carryover effects.
This study has some limitations. Specifically, these include the relatively low number of subjects in each group (n = 8) and the relatively high oral moisture levels at the baseline. In this study, a high number of participants (35 individuals) were excluded, as they were not eligible based on the exclusion criteria. Specifically, these participants were postoperative patients with oral cancer who had a history of radiation therapy in the head and neck, which is a risk factor that causes salivary gland damage and xerostomia [1,2].
It is desirable that the examination of dry mouth be performed simply and quickly chair-side. Moisture-checking devices are available for such moisture measurements. Moisture-checking devices measure the moisture of the submucosal layer indirectly [18,19]. Oral moisture, which was recorded as the moisture value, was expressed as a percentage: B/(A + B) × 100(%), where A is the weight of the dried protein membrane, and B is the weight of the water [18]. It has significant advantages for the examination of oral moisture levels, because the measurement time is only 2 s, and is not affected by the patient's chewing ability [18]. It is reported that the degree of moisture in the lingual mucosa is correlated with the salivary flow rates (unstimulated whole salivary flow and stimulated whole salivary flow) [19].
The participants in this study had an average age of about 70 years. Their main underlying disease was hypertension, and they were taking antihypertensive drugs. This is a risk factor for xerostomia [1,2]. However, there were no significant differences in the number of underlying disease and medications at the baseline between the two groups. Also, there was no significant differences in the moisture level at the baseline between the two groups. However, it is important to note that the median moisture level of these groups was 31.55, which is higher compared to the standard average value of xerostomia patients (27.9 or less) [19]. A study measured the moisture level of the lingual mucosa in patients with xerostomia symptoms by using an oral moisture-checking device, and showed that the moisture level was 27.0, i.e., lower than the values observed in this study [18]. However, those results were related to patients with jaw defects who received radiation therapy, and as such, they cannot be compared directly with those obtained in this study.
Moreover, according to the results of this study in which xerostomia symptoms were assessed by using a 100-mm VAS (i.e., the degree of dryness in the mouth, the amount of saliva, the taste of meal, and overall satisfaction of oral moisture), it also became clear that many participants were not satisfied with their oral environment. Some reported that xerostomia symptoms were correlated with the amount of stimulated and unstimulated saliva at rest, whereas others did not find any correlation; therefore, no consistent patterns were observed [18]. In this study, we set the participants' xerostomia symptoms as inclusion criterion, because this criterion was easier to assess compared to the measurement of salivary secretion. The results of this study, in which a significant difference in gender distribution was found between the two groups at the baseline, might be attributed to the reasons described above. At the same time, these results are consistent with previous studies that have reported that women are more likely to experience xerostomia compared to men [1,2]. In order to clarify this point, gender stratification may be necessary when assigning subjects in future studies.
It is reported that an oral intake of glucosylceramide derived from konjac improves transepidermal water loss in health human subjects [9]. However, to date, no clinical studies have been conducted to examine the effects of oral ingestion of glucosylceramide on xerostomia. The GCP used in this study was a substance extracted and refined from pineapple fruit, i.e., a fruit that can be eaten daily. Regarding this, clearance has already been obtained for GCP as a food material [12,13]. In this study, tablets containing 1.2 mg of GCP, which have been tested for humans and have been confirmed for safety, were used [12,15]. Within the scope of our investigation, there have been no reports of health hazards due to GCP-containing foods, and no study using GCP has reported side effects. However, in general, there is a possible risk of allergic reactions to pineapple, and for this reason, patients with a history of allergies to fruits, including pineapple, were excluded. Overall, there were no reports of adverse events such as discomfort, feeling of strangeness, or other physical or mental abnormalities with the use of GCP tablets in this study.
The results of a randomized, double-blind, placebo-controlled, parallel-group comparison study of a 12-week oral intake of 1.2 mg of GCP in healthy Japanese adults who were concerned about skin dullness and dryness indicated decreases in transepidermal water loss, changes in color difference (increase in L* value, decrease in a* value in Lab color space), and increases in the number of textures on skin [15]. The reasons for the observed results were related to the skin structures becoming complete [15] due to increased ceramide synthesis [11] and the formation of cornified envelopes. In a similar way, it could be assumed that similar improvement effects could be produced on the lingual mucosa by the oral intake of GCP, as the lingual mucosa consists of the same stratum corneum squamous epithelium as the skin. The results of this study showed that GCP administration significantly increased the oral moisture level of the lingual mucosa and the VAS value related to xerostomia, but there was no significant change in the number of fungiform papillae after treatment. This may be due to the relatively high oral moisture level of the lingual mucosa at the baseline, as well as to differences in the GCP dosing period, but further investigations are necessary in this regard.
Xerostomia is caused by many factors, including in the elderly [1,2]. It has been shown that salivary gland function declines with age, even in the case of no systemic disease or medication [22]. Recently, in addition to the application of moisturizers [1][2][3], which is a well-established symptomatic treatment for the prevention of various symptoms [1,2] caused by xerostomia in the elderly, salivary gland massage and facial muscle and tongue exercises have also been reported as possible strategies to improve saliva secretion [23]. Each of these methods, as well as the GCP administration addressed in this study, have been studied in isolation in terms of their effects on xerostomia. In the future, it will be important to assess if, and to what extent, the beneficial effects of treatment for xerostomia may be enhanced by a combined application of more than one method. In addition, it is necessary to examine the effects of oral ingestion of glucosylceramide on the quality of life in patients with dry mouth symptoms.

Conclusions
Within the limitations of this study, the results suggested that the oral intake of glucosylceramide extracted from pineapple may improve the moisture level of the lingual mucosa and xerostomia symptoms.