The role of exogenous epidermal growth factor on Ki-67 proliferation marker expression in the submandibular salivary gland of albino rats receiving doxorubicin

Background: This study was conducted to evaluate the role of exogenous epidermal growth factor (EGF) injection on the Ki-67 immuno-expression in submandibular salivary gland tissue of rats receiving doxorubicin (DXR). Methods: A total of 21 two-month-old male albino rats, of 200 g body weight, were divided into three groups: control group; DXR group, the rats received 20 mg/kg body weight DXR as a single intra peritoneal injection; DXR+EGF group, the rats received the same dose of DXR and on the next day they were injected intraperitoneally with 10 µg/kg body weight of EGF daily for one week. Histological sections and immunohistochemical expression of Ki67 sections were examined using a ZEISS Primo Star light microscopy and images taken using Tucsen IS 1000 10.0MP Camera. Results: Ki-67 expression was significantly increased in submandibular salivary glands of rats after DXR injection. However, Ki-67 expression in the glandular tissue was restored to normal levels after EGF injection. Conclusions: EGF preserved glandular architecture after DXR injection and maintained Ki-67 immune-expression within the glandular tissue near to the normal level.


Introduction
Ki-67 is a heavy protein of 395 kDa weight. It is a proliferation marker which is highly expressed in various tumors, and has been used for investigations of many cancer types. Ki-67 is controlled by proteolytic pathways and has similar essential properties with other proteins known to regulate the cell cycle (Hofmann & Bucher, 1995). Furthermore, Ki-67 is an important protein for cell division as antisense nucleotides of Ki-67 will stop the division, and it is a vital factor in the formation of ribosomes (Schluter et al., 1993). This is reinforced by the conclusion that Ki-67 immuno-expression associates with the rate of protein production and function of ribosomes (Plaat et al., 1999).
Proliferation marker evaluation is of high value in pathological diagnosis and prognosis. It has been reported that Ki-67 has a prognostic character for many forms of malignant tumors, such as lymphomas, breast, prostate and colorectal cancers (Tretiakova et al., 2016). Ki-67 is a protein formed during active phases of the cell mitotic cycle, but is not present in resting cells. Therefore, its expression is used as an assessment tool for tissue proliferation (Faur et al., 2015).
The division activity measured by Ki-67 has been reported in previous studies, and has a great prognostic significance in different forms of malignancies (Yerushalmi et al., 2010). Ki-67 is a protein linked with cell production and is noticeable in all active phases of the cell cycle cresting at G2 and persisting at low levels after cell cycle exit. It then becomes undetectable in senescent cells (Sobecki et al., 2017).
Epidermal growth factor (EGF) could motivate the production and differentiation of epidermal cells and assist skin renewal and wound healing. The therapeutic effect of EGF in the treatment of thermal injuries is not only confined to rapid activation of the healing process and a decrease in tissue damage, but also decreases the size of the affected area and reduces hyperergic inflammation. EGF has demonstrated its efficacy in thermic injury by stimulating wound healing and decreasing the possibility of purulent septic complication and tissue damage. This process might also involve modulation of the immune system status (Osikov et al., 2014;Parment et al., 2007). Doxorubicin (DXR) is an important drug for leukemia, Hodgkin's lymphoma and bladder, breast, stomach, lung and ovaries cancer, treatment during chemotherapy (Bielack et al., 1996). Jensen et al. (2002) studied the consequence of chemotherapy on the salivary gland with different solid and hematological tumors. Apart from xerostomia, 50% of the salivary glands of the patients showed ductal dilatation, cyst formation, degenerated acini and inflammatory cell presence. These degenerated salivary glands were markedly detected less than 2 weeks after chemotherapy. DXR motivates reactive oxygen species (ROS) synthesis and depolarizes the membrane potential of the mitochondria. Both excessive ROS synthesis and mitochondrial membrane damage are very important causes of cellular injury (Ghosh et al., 2011).
This study was conducted to evaluate the role of exogenous EGF injection on the Ki-67 immuno-expression in submandibular salivary gland tissue of rats receiving DXR. Rats were used in this study because they have biology similar to humans and therefore can be a model for human carcinogenicity and recovery.

Animals
This study protocol and the animal care and experimental procedures were approved by the Ethics Committee, Faculty of Dentistry, Cairo University, Egypt (#415). All efforts were made to ameliorate any suffering of the animals by adopting the OECD 423 test guidelines, and all applicable international, national, and/or institutional guidelines for the care and use of animals were followed.
In total, 21 male adult albino rats, two months old, pathogen free, with an average weight of 200 gm were used. The animals were obtained from and housed in the Animal house, Faculty of Medicine, Cairo University. Sample size calculation was performed using G*Power version 3.1.9.2 (University Kiel, Germany) (Faul et al., 2009). The effect size was 0.95 using α level of 0.05 and β level of 0.05, i.e., power = 95%; the estimated sample size (n) was a total of 21 samples for three groups.
The animals were housed in a controlled environment (temperature 25±2°C, humidity 70-80% and 12hr dark/light cycle) and had free access to food and water. The animals were fed a natural diet and water ad libitum throughout the whole experiment. The rats were acclimatized to their cages for 1 week.
All 21 rats were given a number (1-21) using a marker pen, then randomized by putting the numbers in an envelope and dividing them into three groups according to the numbers which were taken from the envelope.
The three groups were as follows: • control group, the rats were kept on a normal diet and did not receive DXR or EGF; • DXR group, the rats received 20 mg/kg body weight DXR as a single intra-peritoneal injection (Ayla et al., 2011); • DXR+EGF group, the rats received the same intraperitoneal dose of DXR (20 mg/kg body weight) and on the next day they were injected intraperitoneally with 10µg/kg body weight of EGF daily for one week (Ohlsson et al., 1997).
The rats were injected every morning at 9 am in the animal house laboratory of Faculty of Medicine, Cairo university.

Procedure
This study was performed to detect changes in the salivary glands after doxorubicin injection and the role of EGF, if any, in reversing any negative changes appearing in the glands. Therefore, histological sections as well as Ki67 immuno expression were used to detect these changes.
The rats were sacrificed by euthanization by CO 2 asphyxiation followed by cervical dislocation when the experiment finished after 1 week.
Submandibular salivary glands of both sides were dissected out and preparation of specimens for staining procedure was done as follows.
After the glands were excised, those of the right side were fixed immediately in 10% neutral buffered formalin. Then, the specimens were washed properly under running water, dehydrated through ascending concentrations of alcohol and transferred to xylene to clear the specimens from alcohol. Then, the glands were embedded in paraffin wax and mounted in the center of the paraffin wax blocks. Sections from paraffin embedded tissues blocks were cut into sections 5-µm thick and mounted on glass slides for histological examination using Samples were processed for regular histopathological examination using H&E stain. Other sections were stained with immuno-peroxidase for immunohistochemical detection of Ki-67 in the glandular tissue using staining reaction containing anti-Ki-67 antibodies (Santa Cruz Biotechnology catalogue # sc-23900).
The selected sections were studied by ZEISS Primo Star light microscopy and images taken using Tucsen IS 1000 10.0MP Camera in the Oral Biology Lab, Faculty of Dentistry, Cairo University.
The scoring of the staining reaction of the immunohistochemical parameter (presence of Ki-67) of the different groups was as follows: (-) negative reactivity and (+) positive reactivity.

Image analysis
Image analysis was performed using a computer system software Leica Quin 500 (LEICA Imaging Systems Ltd, Cambridge, England) to assess the area percentage of the immunostaining within the tissues.
The image analyzer was first calibrated automatically to convert the measurement units (pixels) produced by the image analyzer program into actual micrometer units. Measurement of the area of percentage positive cells was done as previously described (Shi et al., 2007). Briefly, the percentage of positive cells was recorded as an area and area percentage within a standard measuring frame of area 114,342.9 mm 2 per 10 fields from different slides. This was done at a magnification of 400X by the light microscope transmitted to the monitor. Areas containing the most uniformly stained tissues were chosen for evaluation. These areas were disguised by blue binary color which could be measured by the computer system. Images were manually corrected for brightness and contrast. Colour thresholding was then performed automatically after which pictures were converted to RGB stack type. Masking of the brown cytokeratin, immuno-stain was performed by blue colour where any brown stain of any intensity was considered positive whereas the background grey stain was considered negative. Area fraction was then calculated automatically representing the area percentage of immune positive cells to the total area of the microscopic field.

Data analysis
Image analysis data representing experimental values of Ki-67 immunostain were given as mean and standard deviation. ANOVA (ONEWAY ANOVA test, n=10, P <0.05) was used to compare the mean area percent of Ki-67 immuno-expression among the specimens of different groups. Tukey's Multiple Comparison Test (Post Hoc Tukey HSD) was performed to calculate a pair-wise comparison between each group. SPSS 25.0 for Windows (SPSS Inc., Chicago, IL, USA) was used for analysis.

Histological examination
Histological examination of the submandibular salivary gland of rats of control group (group I) revealed its main structural components was composed of parenchymal tissue supported by connective tissue stroma ( Figure 1A). Histopathological sections of DXR group (group II) showed several pathological changes. The secretory acini appeared with massive cytoplasmic vacuolization, and deformation in the acini and loss of normal cellular orientation were frequently encountered. A clear space separated the parenchymal elements, which might be an index of interstitial edema and extravasated red blood cells (RBCs) in between acini and ducts ( Figure 1B).
Comparing with the histopathological results of DXR group, the DXR+EGF group (group III) sections showed great enhancement in the structural features of the glands. Little evidence of inflammatory condition was present. On the other hand, many blood vessels engorged with RBCs were found in close relation with the striated ducts. Moreover, a rich vascular network was found in association to the excretory ducts ( Figure 1C).

Ki-67 immunoreactivity
The control group sections showed positive cytoplasmic immunoreactivity for Ki-67 protein in the parenchymal tissue of the glands, which appeared more distinctive in the duct system. Scattered nuclear reactivities were identified for the protein antigen. A few localized focal areas in the secretory acini as well as the endothelial cells of blood vessels expressed the proliferation antigen at higher intensity ( Figure 2A). On the other hand, the immunohistochemical findings of the DXR group corroborated with the histological results; Ki-67 protein was localized in the DXR group sections differently to the control group. DXR was found to increase the immuno-expression of Ki-67 protein in the submandibular salivary glands of rats particularly the secretory terminal portions. Wider areas of the acini expressed the antigen and with greater intensity. The expression of the proliferation antigen appeared foamy, probably due to the vacuolar degeneration affecting the glandular tissue. Scattered nuclear and perinuclear reactivities were identified for the protein antigen in this group ( Figure 2B).
The differences between the control and DXRs groups were statistically significant, as there was a significant increase in the mean area percent of Ki-67 immuno-expression found in DXR group in contrast with the control group (p<0.01; Figure 3).    The tissue sections of the DXR+EGF group revealed a remarkable decrease in Ki-67 antigen expression in the acinar cells in comparison to the DXR group. Conversely a slight increase in immunoreactivity to the proliferation antigen appeared in the ductal profiles. Nevertheless, the immunoreactivity seemed to be very similar to that of the control group ( Figure 2C).
Post Hoc Tukey HSD test comparing between the groups showed that the difference between the control group and DXR+EGF group was not significant (p>0.05). However, a significant increase was recorded between the DXR and control groups, and a significant decrease was recorded between DXR+EGF and DXR groups (p<0.01; Table 1 and Table 2).

Discussion
In this study, the systemic injection of DXR in rats resulted in pathological structural alterations within the submandibular salivary gland tissue. Atrophied and degenerated acini with multiple cytoplasmic vacuolization were detected. The chemical composition of DXR leads to creation of free radicals and the stimulation of oxidative stress, which is considered the main factor of cellular damage (Saad et al., 2001). DXR also initiates an inequity between free oxygen radicals and antioxidants. This disturbs the oxidant-antioxidant system, leading to tissue damage that is manifested by lipid peroxidation and protein oxidation within the tissue (Karaman et al., 2006). In our study, the submandibular salivary glands of rats injected with EGF after DXR presented great improvement in the gland architecture. Resolution of vacuolar degeneration was noted apart from minute areas of microvesicles detected in the parenchymal tissue. The acini and ducts cells were more alike to those of the control group. The anti-inflammatory response of EGF was evident in this current work and has been confirmed in previously conducted studies. Berlanga et al. (2002) recorded a protective effect of EGF on the intestine from ischemia/reperfusion injury. They noticed the marked reduction of inflammatory infiltrates (mainly neutrophils) in the intestinal tissue after EGF injection. They also registered decreased level of TNF-α (a major pro-inflammatory cytokine), which may contribute to the cytoprotective effect.
The healing potential of EGF was documented in several studies. It was prevalent in epithelial cell re-epithelialization, proliferation, migration and renewal of gastric glands during renal epithelium regeneration ( Kausch et al. (2003) found that the expression of Ki-67 is increased at G2/M, which is exactly the period during which DXR induces apoptosis. They suggested that DXR could have an inhibitory effect on Ki-67 protein production, which may induce apoptosis. However, cancer cells, in an attempt to survive this effect, increase their mRNA to produce more protein.
Ultimately, the production of protein by the cell and induction of apoptosis by DXR reaches an equilibrium, the result of which has been a lack of change in the protein after DXR treatment.
These results were also constant with findings in human hepatocellular carcinoma in which DXR treatment caused the acceleration of cell cycle transition; at an early time point allowing cell cycle continuance, but finally leading to cell cycle arrest (Choi et al., 2012). Chemotherapy targets rapidly proliferating cells that are closest to blood vessels but poorly penetrated tumor cells located distal to functional blood vessels (hypoxic regions). Hypoxic cells do not respond to treatment because of the cytotoxic effects generated by oxygen-dependent free radicals. Surviving hypoxic cells in intervals between treatments might re-oxygenate and proliferate from enhanced supply of nutrients released from digestion of dead cells close to the blood vessels. Unexpectedly, we found that Ki-67 proliferation marker expression decreased significantly in the DXR+EGF group in comparison to the DXR group (p<0.001). Weak to mild cytoplasmic immunoreactivity for Ki-67 protein was shown in the parenchymal tissue of the glands of the EGF supplemented group in a manner matching its expression in the control group. Although the expression was a bit stronger than the control, the statistical correlation was not significant (P>0.05). The immunohistochemical results might be correlated with the ultrastructure findings, as binucleation was frequently encountered in DXR group displaying high expression of the proliferation marker, while being unidentified with EGF treatment of low antigen expression (Mansy et al., 2020).
Comparable findings were reported by Fatimah et al. (2012), as they found EGF significantly decreased the pluripotent genes expression of cultured human amnion epithelial cells. It is likely that the mitogenic EGF did not favor abnormal proliferation, as had been unexpectedly detected in the current investigation secondary to DXR. In a previous study, a disruption in normal expression of EGF was found correlated with improved proliferation and differentiation of medial cells in developing palate and resulted in cleft palate in rat embryo (Abbott & Bimbaum, 1990).

Conclusions
We conclude that EGF has a cytoprotective and reparative effect against DXR induced changes on salivary gland tissue in rats. DXR injection significantly increased Ki-67 immunoexpression in the glandular tissue. However, exogenous EGF preserved the immunohistochemical expression of Ki-67 in the glands or restored it to approximately to the normal level. The research by Mansy et al. is an interesting study which examined the effect of Epidermal growth factor on salivary gland of albino rats receiving doxorubicin detecting ki67 expression as a proliferation marker. The title of the study is suffiecient and reflects the purpose of the study. The methods sections was more than sufficient and I think there is more than enough details that it can be shortened and only referencing is enough. Histological and immunohistochemical results figures were clear. The manuscript is structurally organized and well written but needed minor modifications in some areas as in the discussion which needed some correlation between other studies and the current study.

Data availability
Conclusions were drawn adequately and the overall manuscript is well constructed and meets the criteria for indexing.

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Yes
Yes reviewer has provided some suggestions below for the modification of the 'discussion' section. Similar suggestion would apply for the 'Methods' section, as the section is unnecessarily too long and in some cases lack proper scientific language use. The authors should consider rewriting the 'methods' section in succinct form as well. As the authors have demonstrated changes in the rat tissue after the injection of EGF, which is encouraging, but what does it means, that is, what is the significance of doing this study? The authors should consider adding few sentences about the significance of this study in the manuscript.
The authors have performed this study on rat's epithelium, but drawn comparisons with human epithelial cells, which are considered two different species. Have the authors performed investigation and observed similar findings in human tissue? If not, then the authors should carefully conclude the study with their findings and make relevance with the human. Furthermore, the authors have performed the drug experiment on rats for 7 days, the reviewer is concerned about the length of the experiment. Can the authors provide reference of such study length is used in any other previous studies? If so, please add those references in the manuscript. Although this manuscript might have met the criteria for publication on this journal should the following several concerns be addressed properly.
The title of the manuscript looks too long, authors should consider meaningful shortening of title.
In the first paragraph of the 'Methods' section, authors mentioned that they have followed institutional guidelines for animal study. Can the authors mention in detail what guideline their institution follows, is it Declaration of Helsinki guidelines ? If so, please mention that.

9.
In the fourth paragraph of the 'methods' section, did the authors kept all 21 rats in one case of separate cages ? Please, clarify. 10.
The authors should consider rewriting the 'methods' section in succinct but mention all necessary details, as this section is unnecessarily too long and lacks proper scientific language use.
Reviewer Expertise: Head and Neck Cancer, Molecular Biology, Immunotherapy, Genome editing, Cancer genomics, Translational research.
I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above.
The benefits of publishing with F1000Research: Your article is published within days, with no editorial bias • You can publish traditional articles, null/negative results, case reports, data notes and more • The peer review process is transparent and collaborative • Your article is indexed in PubMed after passing peer review • Dedicated customer support at every stage • For pre-submission enquiries, contact research@f1000.com