Salama, Nagla M.

Introduction

Telocytes are a new type of interstitial tissue cells that were discovered in the stroma of a number of organs in 2005 ^1–4. They were formerly known as interstitial Cajal-like cells because of their apparent similarity with gastrointestinal interstitial cells of Cajal, the gut pacemaker cells ^5,6. In 2010, after confirming that these cells have particular features that distinguish them from interstitial cells of Cajal and all other interstitial cells, their name was changed to telocytes (cells bearing long prolongations) as they have one to five very thin, extremely long prolongations named telopodes ⁷.

The existence of telocytes was reported within the interstitium of many cavitary and noncavitary organs including the heart ^8,9, placenta ¹⁰, lung ^11,12, urinary tract ¹³, skeletal muscle ¹⁴, parotid gland ¹⁵, and pancreas ¹⁶. Several roles have been suggested for telocytes in these organs: mechanical support ¹⁷, intercellular signaling ¹⁸, immune surveillance ¹⁹, regulation of stem cell niche ¹³, and tissue regeneration ^14,20.

To characterize telocytes, many different techniques have been used, such as immunohistochemistry ¹, transmission electron microscopy ³, scanning electron microscopy ²¹, electron tomography ¹⁸, and in-vitro isolation in culture ²². The immunohistochemical detection of telocytes is linked to the expression of surface markers like c-kit (CD 117), which is a transmembrane receptor with tyrosine kinase activity and is essential for normal function of the cells ^10,23.

Telocytes have been also discovered in human and rat myometrium ^22,23 and were isolated from rat endometrium ²⁴. The function of uterine telocytes is not well understood; however, experimental studies provide evidence that telocytes may be involved as modulators of spontaneous contraction of the uterus ^25,26. This is possibly carried out under hormonal influence, as uterine telocytes have been shown to express estrogen and progesterone receptors ²⁷.

Published studies have studied uterine telocytes only in adult nonpregnant uteri and, to our knowledge, no data are available about the existence and condition of these cells in the uterus during other reproductive states. Accordingly, this study was conducted to verify the morphology, location, and number of telocytes in premature, pregnant, and postpartum rat uteri in comparison with adult nonpregnant rat uteri.

Materials and methods

This study included 24 apparently healthy female albino rats obtained from the animal house of Kasr Al Aini Faculty of Medicine. They were housed in hygienic cages according to the guidelines for animal research issued by the National Institute of Health and approved by Animal Ethics Committee, Cairo University. They were classified as follows:

• Immature rats: included six female rats about 4 weeks old, weighing 100–120 g.
• Adult nonpregnant rats: included six female rats about 12 weeks old, weighing 170–190 g.
• Pregnant rats: included six pregnant female rats on day 16–18 of gestation. They were about 12 weeks old and weighed 230–270 g. Day 1 of gestation was identified by the appearance of a vaginal plug. The average time of delivery was during the morning of day 23 ²⁸.
• Postpartum rats: included six female rats on day 3 postpartum. They were about 12 weeks old and weighed 180–200 g.

The rats of all groups were sacrificed under anesthesia by intraperitoneal injection of pentobarbitone sodium 60 mg/kg ²⁹. The uteri were dissected and the middle one-third of the right uterine horns (for pregnant rats, the interimplantation sites of the middle one-third) were fixed in 10% buffered formalin solution for 24 h. They were then dehydrated in ascending grades of ethanol and embedded in paraffin. Serial sections of 6–7 µm thickness were cut and subjected to the following stains:

H&E ³⁰ for routine histological examination.

Immunohistochemical detection of telocytes using anti c-kit polyclonal antibody (CD117; Dako, Carpinteria, California, USA) at 1:100 dilution.

The immunohistochemical procedure was carried out according to Bancroft and Cook ³¹. After routine deparaffinization and rehydration, antigen retrieval was carried out by boiling the sections in 10 mmol/l citrate buffer (pH 6) in a microwave oven for two cycles for 1 min each, replacing the lost buffer in between the cycles ³². The endogenous peroxidase was blocked by 3% hydrogen peroxide for 5 min. The sections were then incubated overnight with the primary anti-c-kit antibody. The bound primary antibody was immunodetected using the labeled avidin–biotin–peroxidase complex (Histostain SP kit; Zymed Laboratories Inc., San Francisco, California, USA). Diaminobenzidine was used as a chromogen and Mayer’s hematoxylin as a counterstain. To establish immunohistochemical staining specificity, negative control serial sections were processed by replacing the primary antibody with PBS, with all other steps performed in the same manner. Positive tissue control was a specimen of GIT stromal tumor (according to data provided by the antibody manufacturer).

Morphometric and statistical study

Using Leica Qwin 500 C image analyzer computer system (Leica Ltd, Cambridge, UK), the number of c-kit-positive telocytes per high-power field (HPF) was determined in the endometrium and myometrium. For each layer, c-kit-positive telocytes were counted in 10 nonoverlapping randomly chosen fields (×400; measuring frame: 7286.78 μm²). The measurements obtained were analyzed using SPSS software version 9 (SPSS, Chicago, Illinois, USA). Comparison between different groups was made using analysis of variance followed by the post-hoc Tukey test. The results are expressed as means±SD. The differences were considered statistically significant when P values were less than 0.05.

Results

Light microscopic results

The immature group

Examination of H&E-stained sections showed that the uterine wall was composed of the endometrium, myometrium, and perimetrium. The endometrium consisted of a surface columnar epithelium overlying a thick lamina propria (stroma) containing compact stromal cells, many small blood vessels, and a few endometrial glands. The myometrium was composed of a thick inner circular and a thinner outer longitudinal smooth muscle layer with an intervening vascular layer, stratum vasculare (SV). The perimetrium was formed of a very thin, hardly differentiated, connective tissue layer (Fig. 1).

Figure 1

Examination of immunostained sections exhibited the presence of a few scattered c-kit-positive telocytes in the endometrium, inner circular and outer longitudinal smooth muscle layers of the myometrium, and in the SV. They appeared as small fusiform or triangular cells with scanty cytoplasm and large oval or flat nuclei (Fig. 2).

Figure 2

The adult nonpregnant group

Similar to the previous group, the uterine wall consisted of the inner endometrium, middle myometrium, and outer perimetrium. However, the lamina propria of the endometrium contained many endometrial glands. The myometrium was organized into inner circular and outer longitudinal layers of smooth muscles with a prominent layer of large blood vessels, SV, located between the two layers (Fig. 3).

Figure 3

In immunostained sections, many c-kit-positive cells were seen in all layers of the uterine wall (Fig. 4). Two cell types expressed c-kit immunoreactivity: telocytes and mast cells. However, they were clearly distinguished from each other by their morphology. Mast cells were large rounded or oval cells with abundant granular cytoplasm and rounded small nuclei (Figs. 5 and 8), whereas telocytes appeared as small, predominantly spindle or pyriform-shaped cells with scanty cytoplasm and large oval nuclei. One or two long thin processes, with an uneven caliber (telopodes), could be occasionally identified extending from the poles of the cell body. In the endometrium, telocytes were commonly located in close contact with blood vessels or around endometrial glands (Figs. 5 and 6). In the inner circular smooth muscle layer of the myometrium, many c-kit-positive telocytes were arranged in parallel orientation with smooth muscle fibers (Fig. 7), whereas in the outer longitudinal layer, the c-kit-positive telocytes were located mainly on the boundaries of smooth muscle bundles with some telocytes appearing within the bundles between the muscle fibers (Fig. 8).

Figure 4

Figure 5

Figure 8

Figure 6

Figure 7

The pregnant group

The endometrial stroma was relatively thin, loose, and contained numerous congested blood vessels. The myometrium showed hypertrophy and hyperplasia of both inner circular and outer longitudinal smooth muscles, together with widening of the interstitial (interspersing) connective tissue. Prominent SV was seen between the two muscle layers. The myometrium was covered by a thin perimetrium (Fig. 9).

Figure 9

In immunostained sections, the endometrium showed many c-kit-positive telocytes located mainly around the endometrial glands and adopted an orientation that often contoured the shape of the adjacent gland (Figs. 10 and 11).

Figure 10

Figure 11

In the myometrium, apparently fewer c-kit-positive telocytes were detected. However, they were more frequent in the outer longitudinal smooth muscle layer than in the inner circular one. They were parallel with circular smooth muscle while occupying peripheral position on the borders of longitudinal muscle bundles with only a few cells located inside the bundles among the muscle fibers (Figs. 10, 12, and 13).

Figure 12

Figure 13

The postpartum group

The myometrium showed marked reduction in the size of smooth muscles, especially the outer longitudinal ones. Mononuclear cell infiltration was detected in both the endometrial stroma and the myometrium. Large congested blood vessels were seen within all uterine wall layers (Fig. 14).

Figure 14

In immunostained sections, many c-kit-positive telocytes were seen in the endometrium, mainly around the endometrial glands (Figs. 15 and 16). In the myometrium, there was an apparent increase in the number of c-kit-positive telocytes, especially in the outer longitudinal layer. They were parallel to the circular smooth muscle and located on the borders and inside the bundles of longitudinal muscles (Figs. 15, 17, and 18).

Figure 15

Figure 16

Figure 17

Figure 18

Morphometric and statistical results

The mean counts of c-kit-positive telocytes/HPF in the endometrium and myometrium of all the studied groups are presented in Table 1 and Histogram 19.

Table 1

Histogram 19

Compared with immature uteri, adult nonpregnant uteri showed a significant increase in telocyte number in both the endometrium and myometrium. The pregnant uteri showed a further significant increase in the endometrial telocytes but a significant decrease in myometrial telocytes. The postpartum uteri showed the highest significant count of myometrial telocytes as compared with all other groups but a nonsignificant difference in endometrial telocyte count as compared with the adult nonpregnant group.

Discussion

The discovery of telocytes in the uterus is fundamental to a totally new approach toward the mechanisms controlling myometrial contractility ³³.

The present study demonstrated the presence of telocytes in the endometrium and myometrium of rat uterus in different reproductive states. The lowest count of both endometrial and myometrial telocytes was recorded in the immature uteri, whereas the highest count of endometrial telocytes was detected in the pregnant uteri and the highest count of myometrial telocytes was present in the postpartum uteri.

Identification of telocytes in this study was carried out immunohistochemically using anti-c-kit antibodies. However, two types of cells expressed c-kit immunoreactivity, which were clearly distinguished by their different morphologies. The first cell type was mast cells, which appeared as large round-to-oval cells with abundant cytoplasm and small round nuclei. The presence of c-kit-positive mast cells in the uterine wall was reported in mice as well ²⁵. These cells have been suggested to play a functional role since an increase in uterine smooth muscle contractility could be obtained by a mast cell-dependent pathway through their degranulation ³⁴. The second cell type was telocytes, which appeared as small pyriform or spindle-shaped cells with one or two thin long processes. These processes have been named telopodes and were reported to reach exceptional lengths up to hundreds of micrometers ⁷. However, these telopodes were not constantly seen projecting from all immunostained telocytes, as they are often very thin processes, thinner than 0.2 μm under the resolving power of light microscopy. In addition, they were commonly convoluted in a 3D manner, which impedes their observation in a 2D very thin section ³³. Furthermore, in contrast to CD 34, c-kit was demonstrated to stain mainly the cell body of telocytes with less preferential affinity to stain the cell processes ³⁵.

C-kit-positive telocytes were detected in the endometrium of all studied groups. Their location appeared as a strategic one, mainly around the glands, adopting an orientation that frequently contoured the shape of the adjacent gland. This strategic location could raise the suggestion that telocytes may support the structure of the gland and the nearby stroma by forming a scaffold around them. Coculture studies have shown that uterine stromal cells are able to enhance epithelial cell growth ³⁶. Therefore, it could be postulated that telocytes may play a role in the maintenance of glandular epithelial cells. In addition, Hatta et al.²⁴ assumed that endometrial telocytes, through their connection to stromal cells, may play an active role in endometrial maintenance by communicating across long distances. They based this assumption on the observation that, in the culture of rat endometrium, telopodes were seen connecting distant colonies of stromal cells, as well as on the finding that telocytes usually form and release shed vesicles, thus sending macromolecular signals to target cells ¹⁸. Recently, Cretoiu et al.³³ speculated that telocytes can perform intercellular communication through two mechanisms: a paracrine secretion of small signaling molecules and/or shedding microvesicles that transport important macromolecules among neighboring cells. Hence, Hatta et al.²⁴ hypothesized that telocytes are frequently present in tissues that are low in cell density, with significant space between neighboring cells. This hypothesis could be supported by the present finding that in the pregnant group, the number of endometrial telocytes was significantly higher than in other groups as the endometrium, unlike the myometrium, becomes loose with less cellularity during pregnancy ³⁷, a condition that necessitates the presence of more telocytes to facilitate cell-to-cell contact over long distances. Moreover, telocytes were previously demonstrated to express connexin 43, a gap junction protein, which probably has a vital role in decidua maturation as its decrease was associated with recurrent pregnancy loss ³⁸.

An opposite condition was present in the endometrium of the immature uterus where the stroma was compact, densely cellular, and contained few glands, a state that requires fewer telocytes for glandular support and stromal cell communication. This might explain why the least count of endometrial telocytes was detected in the immature group.

The current work also demonstrated the presence of c-kit-positive telocytes in the myometrium of all studied groups. This finding goes in hand with that of Popescu et al.³⁹, who disclosed that telocytes constituted about 7% of the total cell number in nonpregnant myometrial cell culture and about 3% of the entire cell population in the myometrium of adult nonpregnant humans.

Myometrial telocytes were suggested to play a c-kit-dependent central role in the generation and coordination of myometrial contractile activity ²⁵. In-vitro studies revealed that c-kit inhibition, by imatinib, led to a reduction in both the amplitude and frequency of myometrial contraction in a dose-dependent manner ^23,25,26. Preliminary evidence also implies that telocytes could possess electrical properties. They produce electric slow waves that trigger and coordinate smooth muscle contraction ²². Morphologically, telocytes have numerous gap junctions and calcium handling units that are the phenotypic features of a pacemaker cell ⁴⁰. In addition, Popescu et al.³⁹ speculated that telocytes are actively involved in the regulation of smooth muscle contraction as they may be implicated in neuromuscular transmission integrating the signals between nerves and myocytes and/or they can act as mechanotransducers; they can sense and translate the stretch information to the surrounding smooth muscle cells.

Supporting the role of telocytes in initiating myometrial contraction, was the study of Shmygol et al.⁴¹ who detected, through calcium imaging of live myometrial tissue slices, that contractile signaling starts on the borders of smooth muscle bundles where telocytes are mainly located.

The present morphometric study revealed that, in comparison with adult nonpregnant uteri, the number of myometrial telocytes was significantly lower in immature and pregnant uteri and significantly higher in postpartum uteri. An explanation to this finding may be that the immature uterus has no or weak contractile activity, which then progresses with puberty for expulsion of menstrual debris, sperm transport, and parturition ⁴².

It can also be suggested that the number of myometrial telocytes is reduced during pregnancy to prevent premature uterine contractility and preterm delivery, whereas in postparturition the number of myometrial telocytes increased to enhance myometrial contraction during involution of the uterus.

To our knowledge, this is the first report determining the existence and frequency of telocytes in premature, pregnant, and postpartum uteri. Further studies are needed to characterize uterine telocytes at the time of parturition. Gaining better understanding of uterine telocytes could help in developing therapeutic strategies for the treatment of problematic conditions such as dysmenorrhea, recurrent pregnancy loss, and premature birth.

Table

Acknowledgements

Conflicts of interest

There is no conflict of interest to declare.

References