Trefoil factor family peptides TFF1 and TFF3 in the nervous tissues of developing mouse embryo

Trefoil factor family peptides (TFF1, TFF2, and TFF3) are predominantly found in mucous epithelia of various organs. However, they have also been reported in the nervous tissue, particularly mouse, rat, porcine, and human brain. Th e aim of this research was to determine the presence of TFF1 and TFF3 in the nervous system of developing mouse embryo. Mouse embryos, at the stages E15 to E17 were isolated, fi xed in 4 paraformaldehyde and embedded in paraffi n blocks. Sagittal 6μm sections were made, processed for immunohistochemistry, and incubated with anti-TFF1 or anti-TFF3 primary polyclonal rabbit antibodies. Labeled streptavidin-biotin method was used for TFF detection. TFF1 and 3 were found in the cytoplasm of ganglion cell somata, while TFF3 staining was also visible in the cytoplasm of neurons in diff erent areas and nuclei of brain and medulla oblongata. Neurons in the gray matter of spinal cord were also TFF1 and TFF3 positive, and signal for both peptides was found in the choroid plexus. TFF peptides might be involved in the complex processes of nervous system development and diff erentiation and


INTRODUCTION
Research on trefoil factor family (TFF) peptides started in 1982, when pancreatic spasmolytic polypeptide, today known as TFF2, and sP2 peptide, today known as TFF1 were discovered [1]. A few years later, intestinal trefoil factor (now TFF3) was described. A three-looped, clover-like structural domain is a common characteristic of these peptides, and is important for their resistance to proteolytic degradation [2,3]. Members of this small family of peptides are predominantly found in mucous epithelia of the gastrointestinal system. However, they are also expressed in many other tissues, for example epithelia of respiratory tract, uterus, conjunctiva, thyroid gland, salivary glands, prostate, mammary gland, gallbladder, and thyroid gland [2][3][4]. Expression of these peptides was investigated in connection with diff erent pathological conditions. It has been found that expression of TFF peptides is deregulated in numerous epithelial tumors [4][5][6][7][8][9] and chronic infl ammatory diseases [10][11][12][13]. Th ese peptides act through various mechanisms and participate in apoptosis, cell migration, and immune response, aff ect mucus viscosity and angiogenesis [4,7,11,[14][15][16]. New properties and potential roles of TFF peptides are continually being discovered, making them a fruitful fi eld for scientifi c research.
TFF1 and TFF3 are both found in the nervous tissue, however, to a much lower extent than in the mucous epithelia. Both peptides are expressed in diff erent parts of adult rat, mouse, and human brain. Signifi cant TFF1 expression was noticed in rat hippocampus, frontal cortex, and cerebellum [17], and weak, but uniform expression was detected in all regions of mouse brain [18]. In mouse and rat, astrocytes have been reported to synthesize TFF1 [17,19,20]. Th ere are very limited data on the neural expression of TFF2, which is the least investigated of the three TFF peptides in the context of neural tissue. Until now, TFF2 was found in the ganglion cell layer of murine retina, which is of neural origin [21], whereas regarding TFF3, mouse brain shows signifi cant expression of this peptide in hippocampus, temporal cortex and cerebellum [18]. TFF3 was also detected in human and rat hypothalamus, particularly in the supraoptic and paraventricular nuclei, co-localized with oxytocin, and accumulated together with oxytocin in the neural lobe of pituitary [22][23][24]. Ultrastructural examination of porcine pituitary gland revealed co-localization of TFF3 Trefoil factor family peptides TFF1 and TFF3 in the nervous tissues of developing mouse embryo  and oxytocin within the same vesicles, and was thus hypothesized that from there, TFF3 is released into the bloodstream to aff ect potential, undiscovered target receptors [25]. Injections of TFF3 monomer into rat amygdala exhibit a dose-dependent eff ect on animal behavior (anxiolytic eff ect at low dose or anxiogenic eff ect at high dose) [24]. Recent research identifi es TFF3 as a neuropeptide that facilitates learning, object recognition and retention of memory [26]. Low TFF3 in cerebrospinal fl uid is a predictive factor for brain atrophy, and its potential role in the pathogenesis of Alzheimer' s disease is suggested [27]. Since TFF1 and TFF3 are found in the adult brain of mice and rats, it is interesting to investigate their expression in mouse embryonic nervous tissue, and also to screen for their presence in the neurons of the peripheral nervous system both in adult animals and embryos. Th e aim of our study was to investigate the presence of TFF1 and TFF3 in nervous tissues of the mouse embryo. Expression of TFF2 was not investigated in this study, because TFF1 and TFF3 were of main interest in recent nervous system research in adult animals. However, this could be a subject of future research on TFF peptides.

MATERIALS AND METHODS
Th e use of the animals was reviewed and approved by the Croatian board for scientifi c animal experiments and approved by a local ethical committee. Twelve CD1 mouse embryos 15, 16 and 17 days old (Th eiler stage 23, 24 and 25) were used in this study (fi ve 15-day old, four 16-day old and three 17-day old embryos). Embryos of a specifi c age in our collection were obtained in the following manner: after placing the females and males in the same cage, vagina was checked for the vaginal plug every morning. Th e presence of the vaginal plug was considered the beginning of pregnancy (gestational day 0). On a particular day, depending on the age of embryos required, the females were sacrifi ced by cervical dislocation.
Embryos were isolated, fi xed in 4 paraformaldehyde, and then embedded in paraffi n. Prepared paraffi n blocks were cut sagittally into 6 μm thick sections that were transferred onto adhesive slides, and sections from all embryos were carefully selected to show nervous tissues of interest. Slides were dried, deparaffi nized and rehydrated. Epitope retrieval was performed by heating in the microwave for around fi ve minutes, using 0.01 M citric acid (pH 6). Blocking of endogenous peroxidase was done using 0.3 hydrogen peroxide for 15 min. Non-specifi c protein blocking was made with SuperBlock® (Th ermo Scientifi c, Rockford, USA) for 30 min. Affi nity purifi ed primary polyclonal rabbit anti-TFF1 and anti-TFF3 antibodies (proprietary, self-made) were applied to separate slides, which were incubated at 4° C until tomorrow morning.
Th e antibodies were purifi ed by immunoadsorption from the serum, validated by Western blotting and specifi city was confi rmed by immunostaining of the specifi c sections of the alimentary tract. All the antibodies were produced according to the specifi c peptide sequence. Antibody dilutions used were 1:250, 1:500 and 1:1000 for anti-TFF1, and 1:2500, 1:5000 and 1:10000 for anti-TFF3. Phosphate-buff ered saline (PBS) (pH 7.4) was used as a negative control (primary antibody omitted), and for the dilution of antibodies as well. Best results were obtained using 1:500 antibody dilution for TFF1 peptide, and 1:5000 for TFF3 peptide. Sections of adult mouse stomach were used as a positive control for TFF1 staining, and adult mouse intestine for TFF3 staining. Next day, slides were rinsed with PBS+0.05 Tween (Sigma-Aldrich, St. Louis, MO, USA) four times for 5 minutes and after that biotinylated goat anti-rabbit secondary antibody (Dako, Glostrup, Denmark) was applied for 30 minutes at room temperature. After another four rounds of fi ve-minute washes in PBS+0.05 Tween, Streptavidin-HRP (Dako, Glostrup, Denmark) was applied to the slides for 30 minutes at room temperature. Slides were further rinsed four times in PBS+0.05 Tween, after which fi nal exposure to DAB (3,3'-diaminobenzidine) was performed (Sigma-Aldrich, St. Louis, MO, USA). Slides were then washed with PBS+0.05 Tween, counterstained with hematoxylin, dehydrated and mounted with Canada balsam.
Th e following nervous tissues were examined: brain, spinal cord, dorsal root ganglia and cerebral ganglia (if available). Tissues at diff erent developmental stages were analyzed manually using light microscopy at magnifi cations 20×, 40×, 100×, and 400×, and photographed at 400×. Slides were monitored with Olympus® BX50 microscope, photographed using Olympus® C-5050 digital camera and QuickPHOTO PRO imaging software (Promicra s.r.o., Prague, Czech Republic). For image assembly, Paint.NET software was used (version 4.0.5., dotPDN LLC, Rick Brewster and contributors).

RESULTS
Sections revealed diff erent parts of the nervous system of the embryos. Morphologically, no abnormalities were detected, and the nervous tissues were in line with the developmental stages of the embryos. Presence of TFF1 and TFF3 peptides was detected in diff erent parts of embryonic nervous system at all examined stages, from day 15 to day 17. Brain, spinal medulla, spinal ganglia and in some cases cerebral ganglia were available for examination. TFF1 peptide was found in the ganglion cell somata in the dorsal root ganglia and trigeminal ganglion. Th e signal was present in the cytoplasm of neurons, mostly sparing the nucleus ( Figure 1A). Also, a mild to moderate TFF1 signal was detectable in the whole central nervous system, in the white matter, glial cells and neurons of the brain and spinal medulla (Figures 2A and 3A). Th is signal was diff usely present, consistently appearing on all sections. TFF3 peptide was present both in the cytoplasm and the nucleus of ganglion cells of trigeminal ganglion and dorsal root ganglia ( Figure 1B). Positive staining for TFF3 peptide was also detected in the cytoplasm and nucleus of neurons occupying diff erent nuclei of brain and medulla oblongata ( Figure 2B). As for the spinal cord, neurons in the gray matter of spinal cord were also positive in the same manner ( Figure 3B). At day 17, the staining for TFF3 peptide was more cytoplasmic than nuclear. Some gentle staining for TFF3 peptide was visible in the white matter of the embryonic central nervous system. Choroid plexus was both TFF1 peptide and TFF3 peptide positive (not shown in fi gures).
Th ere were no diff erences in the staining pattern at the different stages of embryonic development. All negative controls showed no immunostaining, and all positive controls showed typical staining both for TFF1 and TFF3 peptide (not shown in fi gures).

DISCUSSION
Th e presence and localization of TFF1 and TFF3 peptides in the neural tissue at diff erent stages of development was investigated using immunohistochemistry. Trefoil factors TFF1 and TFF3 were detected in mouse brain and spinal medulla at developmental stages E15-E17. At this point, we investigated the later stages of the embryonic development since they are morphologically and functionally closer to that of adult animals, and, therefore, more comparable to data currently available.
Although consistent with that of adult animals [18,20], detailed localization of TFF1 and TFF3 peptides in a developing brain was not determined in this study. However, we showed for the fi rst time that TFF peptides can be located in the peripheral nervous tissue. Both examined molecules were found in ganglion cell somata: TFF1 peptide mostly in the cytoplasm and TFF3 peptide in the cytoplasm and the nucleus.
In our study, positive staining for TFF1 and TFF3 peptides was detected at similar developmental stages as the expression of TFF genes observed in developing mouse brain by other researchers (E16-E18) [18]. In adult nervous tissue, TFF1 gene was expressed at low level nearly uniformly in all brain regions, while TFF3 gene which was located in cerebral cortex and cerebellum before birth, was restricted to hippocampus, temporal cortex and cerebellum after birth [18]. Expression of TFF1 mRNA in the rat hippocampus reaches maximum around the birth [2].  Th e role of TFF peptides in the nervous system is still not fully understood. Th e presence of TFF3 peptide in porcine [25] and human [22] oxytocinergic neurons suggests their role as a neurotransmitter/modulator in various brain regions innervated by these neurons, like brain stem, spinal cord and pontine tegmentum [2], or as a neurohormone released into bloodstream and aff ecting potential target cells. Recent research has revealed various behavioral eff ects of TFF3 peptide. It aff ects anxiety bidirectionally in dose-dependent manner and improves learning and object recognition memory [24,26]. In our study, TFFs were found in diff erent neuron somata, and (especially TFF1 peptide) diff usely in the white matter of the central nervous system. Th ese results point to a possibility that TFFs may act as neuromodulators and/or neurotransmitters even during embryonic development.
Presence of TFF peptides in developing neural tissues raises questions regarding their possible role in the nervous tissue development. Since TFF peptides were associated with oxytocin secretion in neural tissue and mucin secretion in goblet cells [20], there is a possibility that TFFs may be markers for physiological maturation of specifi c tissues and cells. If this is the case, they might be used as an embryonic mark to pinpoint the start of an active physiological function. Th ere is, however, another potential developmental role for TFF peptides. Research shows that they are also involved in cell proliferation and migration [15]. Th ese processes are indispensable for the development of central nervous system. Since cell proliferation and migration are of particular importance in embryonic development, presence of TFFs in the developing nervous system goes well with similar roles of these peptides already demonstrated in adult tissues. For this reason, our results support the theory that TFFs might have an active role in nervous tissue development. Th ere is also a possibility that TFF peptides are involved in development and plasticity of neural tissue through mechanisms responsible for their protective eff ects in epithelial tissues, for example, in the case of gastrointestinal mucosal protection and restitution [20]. Otto and Patel showed that all three TFFs were up-regulated in response to physical wounding of embryological gut (E17-18], but the expression showed diff erences in the adult [28]. Identifi cation of novel rodent gene encoding for neural regeneration protein, which contains domain with homology to survival-promoting peptide (SPP) and TFF1 peptide, and exerts biological activities as neural migration, proliferation, diff erentiation, enhancement of neurite outgrowth and promotion of neuronal survival [29], supports proposed active role of TFF peptides in neural development and protective function.
Considering various eff ects of TFF peptides in cell migration, proliferation, apoptosis and angiogenesis, and the pattern of their expression in developing nervous system, it can be concluded that TFF peptides might be involved in complex processes of nervous system development and diff erentiation, and brain plasticity. Th ese preliminary results suggest further research in order to elucidate these peptides' features in the nervous tissues and embryonic development. Such research is important, since mechanisms of embryonic development can sometimes be linked to pathological processes leading to different debilitating diseases [30], and embryological research can suggest novel regenerative therapies [31]. Possible next steps would be to investigate TFF protein expression at the earlier stages of embryonic development, and also to investigate if TFF2 is expressed in adult and embryonic nervous system, both of which may be interesting for future research.

DECLARATION OF INTERESTS
Th e authors hereby declare that they have no confl ict of interests