Protective/restorative role of the adipose tissue-derived mesenchymal stem cells on the radioiodine-induced salivary gland damage in rats

Sixty female Wistar albino rats weighed 230 - 250 g were used in this study. The animals were kept under standard conditions (temperature: 24 - 26°C; relative humidity: 65 - 70%) in 12-h light/12-h dark cycles, with access to water and seeds ad libitum. RAI was administered at a dose of 74 MBq (2 mCi) by oral route to induce salivary gland damage. ADMSC (i.p; 2×10⁶) and amifostine (i.v; 200 mg/m²) were used as cytoprotective/regenerative agents against the effects of radiation. Our main goal was to determine whether ADMSC had a regenerative effect besides its cytoprotective effect.

The study protocol was approved by Dışkapı Yıldırım Beyazıt Education and Research Hospital’s Animal Ethics Board (2013/38), and instructions of local/international animal research were followed. The national/international regulations of radiation safety and waste disposal for contaminated secretion and materials of radio-exposed rats were also followed.

The study groups were designed as follows:

1. Control group: n = 6,

Salivary gland scintigraphy was performed at baseline, and after 1 and 6 months. Three animals in each group were sacrificed at the end of 1^st month, and the rest were sacrificed at the end of 6^th month of the study. The specimens were examined under the light microscope (LM) and transmission electron microscope (TEM). The study and the control groups were compared in terms of RAI-induced tissue damage. Intergroup comparisons were also done among the study groups.

Fried and Moustaid-Moussused’s protocol was used to culture adipose tissue.¹⁷ Adipose tissue explants at a volume of 0.5 cm³ were extracted from the inguinal fat pad of the rats. The tissues were transported from the operating room to the laboratory in a transport buffer (phosphate buffered solution [PBS], 5.5 mM glucose, and 50 g/mL gentamicin), at room temperature. The following procedures were carried out under a laminar flow hood, using sterile equipment. The tissues were transferred to a Petri dish containing 3 mL of PBS, and minced into 10 - 15 mg pieces. The pieces were extensively washed with 10 mL of PBS over a filter containing sterile cotton bandage fabric. Thereafter, they were gently shaken for a short period following transferred to a 25 cm² culture flask, containing 1 mL of PBS. Adipose tissue pieces were cultured within MSC medium (Dulbecco’s Modified Eagle’s Medium - low glucose; Invitrogen; Thermo Fisher Scientific, Waltham, MA, USA), 20% fetal bovine serum (Biological Industries Israel Beit-Haemek Ltd., Kibbutz Beit-Haemek, Israel), 100 U/mL of penicillin, 100 mg/mL of streptomycin (Biological Industries Israel Beit-Haemek Ltd., Kibbutz Beit-Haemek, Israel), and 20 mL/L of L-glutamine. They were incubated at 37°C and in 5% CO₂. Culture medium was changed every three days, and colony-forming unit-fibroblasts were observed at 14 - 16 days with a 70% confluency of the culture plate. The detached cells were subcultured on a standard culture dish, using 0.25% trypsin in 0.02% ethylenediaminetetraacetic acid. After three passages, the cell suspensions were defined with MSC markers including CD11b/c (-), CD45(-), CD90(+), CD44(+), and CD49(+) with the FACS Aria III cell sorter, and stored at -80°C DMEM-LG medium included 10% DMSO and 10% FBS.

To label ADMSC in vitro, 10 μL of BrdU solution (1 mM BrdU in 1 × Dulbecco’s PBS, BD Pharmingen, San Diego, CA, USA) was added for each milliliter of tissue culture medium, and incubated for 2 hours. Cell culture density was 2×10⁶ cells/mL.

To demonstrate the labeled ADMSC, BrdU staining was performed on 6 mm thick slides those were prepared from formalin-fixed, paraffin-embedded tissues, after deparaffinization in xylene to detect BrdU ion. The slides were transferred into 100% and 95% alcohol, then endogenous peroxidase activity was blocked with 3% hydrogen peroxide, and the slides were subsequently rinsed with PBS. Finally, the sections were immunohistochemically stained with a BrdU detection kit (BD Pharmingen) to show the presence of BrdU+ cells.

The excised salivary gland samples were harvested and fixed in 10% formaldehyde solution. After the paraffinization procedure, tissue sections were stained with hematoxylin and eosin, and a blinded pathologist examined all specimens under the light microscope (Olympus BX-50; Olympus Corporation, Tokyo, Japan). Salivary gland damage was categorized into three stages (0: no damage; 1: mild damage; 2: moderate damage; and 3: severe damage) for each of the following findings: acinar epithelial cells (edema, vacuolization, periacinar inflammation, and necrosis), interstitial space (periductal sclerosis, fibrosis, and mucus leakage), ductal system (ductal ectasia), and vascular system (sclerosis). A total score of histological findings reflected the sum of RAI-induced damage in tissues.

Tissue samples were put into 2.5% gluteraldehyde for 24 hours for primary fixation, and they were post-fixed in 1% osmium tetroxide after washing with Sorenson’s phosphate buffer solution (pH: 7.4). In the next step, washing with the same buffer and dehydration in increasing concentrations of alcohol series were done. After dehydration, the tissues were washed with propylene oxide, and embedded in epoxy resin embedding media. Semi-thin (2 µm) and ultra-thin (60 nm) sections of the obtained from the tissue blocks were cut with an ultramicrotome (LKB Nova, Sweden). Semi-thin sections were stained with methylene blue, and ultra-thin sections were stained with uranyl acetate and lead citrate. Those sections were examined under a light microscope (Nikon Corporation, Tokyo, Japan), and a transmission electron microscope (Jeol JEM 1200 EX (JEOL Ltd., Tokyo, Japan), respectively. The same microscopes were used to photograph the specimens.

A double-headed large-field gamma camera (ECAM, Siemens, IL, USA), equipped with a low-energy, high-resolution collimator was used. After 0.2 - 0.3 ml bolus intravenous injection of ^99mTc-pertechnetate (74 MBq) through the tail veins of the rats under anesthesia, dynamic images were collected (120 frames at a rate of 1 frame/20 second) for 40 min with a magnification of 1.23 in a 128 x 128-pixel matrix. For image analysis, oval-shaped regions of interest (ROIs) drawn on the right and left salivary gland regions were used. A ROI on the left supraclavicular area was used for background activity. All ROIs had the same size. The image data were processed, and a time-activity curve (TAC) was generated for each subject using a summarization method including background subtraction and three-point smoothing. Functional assessments of TACs were made for following quantitative parameters: The gland-to-background ratio at a maximum count (G/BGmax) defined the degree of gland parenchymal damage when compared with the background activity since this ratio dropped off due to the parenchymal injury and destruction in a direct manner. The time to reach the maximum count ratio over the time to reach the minimum count (Tmax/Tmin) showed whether the gland activity slowed down due to mostly ductal system damage. It is dependent on the processes of transit in tissue and transport out of the gland. The maximum secretion ratio (Smax; maximum count - minimum count/ maximum count × 100) highlighted secretion capacity of the gland as a percentage.

We used SPSS software (version 18.0 for Windows, SPSS Inc., Chicago, USA) for statistical analysis. Kruskal-Wallis test was employed to compare the groups, Mann-Whitney U test was used to analyze the intergroup differences, and Friedman’s test was used to assess the differences in continuous variables.

Table 1 and Figure 1 show the histological findings of the salivary glands in the study groups. Normal salivary gland histology was shown in upper horizontal line from control samples (Figure 1A-C). Homing of ADMSC to RAI-damaged salivary glands and their trans-differentiation into salivary gland cells are seen in Figure 1D-F.

Table 1

Histological findings in the salivary glands of the rats. Mean values of salivary gland damage parameters are given. *Details of groups are given below the table

Figure 1

Intragroup comparison of the parameters in the control group showed that none of parameters had statistically significant difference between 1^st and 6^th months (P > 0.05).

In Group 1, destructive effects of RAI on acinar cells, interstitial space and vascular system over time were demonstrated with presence of necrosis (P = 0.04), periductal fibrosis (P = 0.02), periductal sclerosis (P = 0.02), vascular sclerosis (P = 0.01), and total sum score (P = 0.02). However, the effect of RAI on the ductal system was not significant (P > 0.05). RAI-induced necrosis and increased vacuolization (Figure 1G), periductal fibrosis and inflammation (Figure 1H-I) were shown in Figure 1.

In Group 2, we observed statistically significant differences on acinar epithelial cells with an increase in edema (P = 0.05), vacuolization (P = 0.04) and periductal sclerosis (P = 0.03).

In Group 3, the findings were not related to RAI, and we assumed those findings as insignificant.

In, Amifostine plus RAI (Group 4), Amifostine did not exhibit a sufficient protective effect in intragroup comparison; and yet the damage increased in a statistically significant manner in terms of edema (P = 0.02), ductal ectasia (P = 0.01), periductal fibrosis (P = 0.02) and total sum score (P = 0.02).

Similarly, in the concomitant administration of stem cells plus RAI (Group 5), we determined a statistically significant increase in periductal fibrosis (P = 0.01) and sclerosis (P = 0.01).

ADMSC seemed the most effective in Group 6. There was a statistically significant decrease on 6th month for edema, vacuolisation, periaciner inflammation, periductal mucus leakage (P = 0.02) and ectasia (P = 0.04) compared to the findings obtained on month 1. In addition, the sum of all histologic parameters decreased only in Group 6, with late stem cell administration. This improvement in histologic findings were demonstrated in Figure 1J-L.

At 1^st month, we found a statistically significantly difference among the groups for periductal fibrosis, sclerosis and the total sum score were (P < 0.05). The differences among the groups were statistically significant for the changes in edema, vacuolisation, necrosis, ectasia, sclerosis, periductal fibrosis, periductal sclerosis, and the total sum score (P < 0.05) at 6^th month. We supposed that interstitial space damage and total sum score were good indicators of RAI-induced damage. Total sum scores indicated that histologic improvements were statistically significant in all preservative treatment groups (Groups 4, 5, and 6 vs. Group 1). Total sum score was the smallest in Groups 4 and 5 at 1^st month, and in Group 6 at 6^th month. We concluded that, although concomitant administrations could be effective in the early phase of RAI-induced damage, late stem cell administration (Group 6) was better for recovery of the functional damage over time when compared to the other administration methods.

TEM demonstrated cellular damage and fibrosis (Figure 2C-D). Intercellular edema and intracellular vacuolization showed cellular damage.

Figure 2

TEM examinations were normal in the control group at 1^st and 6^th months. We observed large intracellular vacuoles and intense intercellular edema in Group 1. RAI-induced damage was more distinct in samples obtained at 6^th month, since the samples exhibited smaller intracellular vacuoles and minimal intercellular edema (Figure 2). The findings of ADMSC-administered group (Group 2) were identical to the findings of the control group. Minimal vacuolization and fibrosis were seen in Group 3, which was administered amifostine alone, but this was reversible, since the 6^th month evaluation was normal. We observed vacuolization, edema, and fibrosis due to RAI damage at 1^st month in the treatment groups (Groups 4, 5, and 6). The cellular damage and fibrosis improved in both the stem cell plus RAI-administered groups (Groups 5 and 6) at 6^th month (Figure 2). However, the RAI damage was not partially reversible with amifostine administration in-group 4. The TEM findings demonstrated the regenerative/restorative effect of ADMSC on RAI-induced salivary gland dysfunction.

The baseline findings were similar for all parameters in all study groups (P > 0.05). RAI groups (Groups 1, 4, 5 and 6) and others (Groups 2 and 3) showed statistically significant differences for Smax (P = 0.01) and G/BGmax (P = 0.01), but not for Tmax/Tmin (P > 0.05) at 1^st month. On the other hand, 6^th month scans revealed statistically significantly differences between RAI and non-RAI groups for Smax (P = 0.01), G/BGmax (P = 0.01) and Tmax/Tmin (P = 0.03). Combined treatment groups showed significant differences for 1^st and 6^th month findings for Smax and G/BGmax values (P = 0.04), but not for Tmax/Tmin (p > 0.05). This might be due to preservation of ductal secretion. RAI-dependent impairment in function at 6^th month was the most prominent in Group 1. The measurements for Smax and G/BGmax seemed to be better in concomitant protective administrations with RAI (Groups 4 and 5), but the difference was not statistically significant. Tmax/Tmin ratio was similar among treatment groups with RAI (Groups 4 - 6). These findings suggested that protective agents could partially overcome functional damage related to RAI (Table 2).

Table 2

The mean values (the counts from left and right salivary glands) of the Smax, Tmax/Tmin and G/BGmax, and their statistical significance. Baseline and follow-up salivary gland scintigraphies were performed on the subjects who had single treatment (Group 1; radioiodine, Group 2; adipose tissue-derived mesenchymal stem cells, Group 3; amifostine) and combined treatments (Group 4; radioiodine plus amifostine, Group 5; radioiodine plus adipose tissue-derived mesenchymal stem cells concomitant, Group 6; radioiodine plus adipose tissue-derived mesenchymal stem cells after 48 h)

The functional status on a salivary gland scintigraphy can be evaluated easily by a TAC generated from dynamic scintigraphic data. We compared pre- and post-treatment TACs generated from the sequential salivary gland scintigraphies of rats at baseline and at 6^th month to observe time-related healing or worsening in the study groups, and to determine protective/regenerative effects of amifostine and ADMSC. Figure 3 depicts the functional change in activity of rat salivary glands over time in RAI and combined treatment groups. A review of the time-sequential findings in Group 1 showed a significant deterioration in function over time with decreased gland uptake and secretion capability (Figure 3A-B). As expected, TACs of two consecutive scans were not different from the baseline in the groups not treated with RAI (Groups 2 and 3). In the analysis of TAC patterns for RAI plus concomitant amifostine and ADMSC groups (Groups 4 and 5) as well as the late ADMSC-administered group (Group 6) mild to moderate deteriorations were seen in function over time in terms of slope of TAC and secretion capability of the gland (Figure 3C-H).

Figure 3