Abstract
Background: The ultimate cause of cancer death is, in most cases, the appearance of metastases. The aim of the present study was to contribute to animal experimental investigations of metastatic tumor development. Materials and Methods: Rat hepatocarcinoma (He/De), mesoblastic nephroma (Ne/De) cells, and in other cases tumor-bearing lymph nodes were transplanted under the renal capsule of F344 rats. Metastatic potential of tumor cells was examined by whole body autoradiography and phosphor image analysis. The organ distribution of cells was also investigated. Results: Transplanted tumor cells resulted in metastases in the parathymic lymph nodes. Implanted India ink also demonstrated connection between the lymphatic vessels of the renal capsule and the parathymic lymph nodes. The metastatic potential was independent of the primary tumor growth rate. Conclusion: The renal capsule-parathymic lymph node complex seems to be suitable for the isolated in vivo examination of metastatic development and for the detailed analysis of secondary tumors.
The ultimate cause of cancer death in most cases is a result of appearance of metastases. The recognition of this fact resulted in several methods for the experimental analysis of metastatic tumor development in the hope that the results would help to understand tumor progression as well as to plan therapeutical and tertiary preventative measures. The chorioallantois membrane system of chicken embryos, the implantation of primary tumors of rodents, the implantation of heterogeneous tumors into immunodeficient rodents and the intravenous (i.v.) injection of tumors are regarded as such models (1-11).
As part of the foundation for developing an in vivo model, we implanted a known number of tumor cells under the renal capsule of rats by means of GelasponR gelatin sponge and followed the kinetics of tumor growth (12). During these experiments, we observed the expansion of mesentheric lymph nodes simultaneously with the initiation of angiogenesis and the growth of parathymic lymph nodes. This led to recognition that the tumor growth was accompanied by infiltration of the parathymic lymph nodes and an increased activity of pyruvate kinase (13, 14). The question arose as to how the tumor cells move from under the renal capsule to the parathymic glands and whether there are methods which could register this transport. Regarding the infiltration of tumor cells, it remained to be determined whether our system would be suitable to answer the questions concerning the formation of metastases.
Materials and Methods
Animals. The experiments were carried out using male and female inbred Fischer 344 rats. Animals were kept in a conventional laboratory environment and fed on a semi-synthetic diet (Charles River Mo, Kft, Godollo, Hungary) and tap water ad libitum. Animals received humane care according to the criteria outlined in the UK “Guide for the Care and Use of Laboratory Animals” (15), authorized by the Ethical Committee for Animal Research, University of Debrecen (permit number: 22/2007).
Experimental tumors. Two types of tumors were used in this study: epithelial liver carcinoma (He/De) and mesenchymal mesoblastic nephroma (Ne/De). Both tumor types were isolated from Fischer 344 rats treated at one-day-old by intraperitoneal (i.p.) injection of 125 μg/animal nitrosodimethylamine (Sigma-Aldrich Kft, Budapest, No 77561) in saline. Tumors were removed 5-7 months after chemical tumorigenesis, minced into smaller pieces and after mincing were frozen in liquid nitrogen (16).
Establishment of tumor cell lines. Tumor pieces, freshly isolated or frozen were further minced into 2×2×2 mm pieces and incubated for 3 h at 37°C in RPMI-1640 medium containing 100 mg collagenase I, 10 mg hyaluronidase and 30 μl DNase I per 100 ml. After digestion, the mixture was filtered through four layers of sterile gauze, washed and resuspended in RPMI-1640 medium supplemented with 10% fetal bovine serum (FBS) and antibiotics. After overnight incubation at 37°C in a 5% carbon dioxide (CO2) atmosphere, nonadherent cells were discarded and adherent cells were subcultured. The primary cell culture was continuously grown and after subculturing for a further 20 days the new cell lines were established, frozen in liquid nitrogen and used for further experiments. Both cell lines (He/De and Ne/De) were used as exponentially growing monolayer cultures (37°C, 5% CO2), maintained by daily passage in RPMI-1640 supplemented with 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin. Cell viability was more than 95%, as assessed by trypan blue exclusion (17).
Experimental surgery. The aim of the surgical operations was to place GelasponR discs or lymph nodes under the capsule of the left kidney. To transplant cells, gelatin sponge discs of 4 mm diameter and 1 mm thickness were cut from GelasponR gauze (Germed, Rudolstadt, Germany) and sterilized. To transplant tumor cells, 106 He/De or Ne/De cells in 10 μl saline (0.9% NaCl solution) were placed onto the GelasponR disc.
The implantation of parathymic lymph nodes (PLNs) was carried out in the following way: the left parathymic lymph nodes were excised from 12 tumor-free (control) and from 12 tumor-bearing rats and the surrounding fat was removed.
In the case of India ink, 10 μl Pelikan ink (Gunther Wagner, Pelikan Werke, Hannover, Germany) was placed onto the gelatin disc. Animals were euthanized 3, 6, 24 and 48 h after implantation (2 animals, each) and tissue sections were prepared from the kidney and from the parathymic lymph nodes.
The next series of experiments aimed to define the time of appearance of metastases. In the first series of experiments related to the temporal aspects of tumor metastasis, 106 He/De and Ne/De cells were implanted in six F344 male and six F344 female rats. Animals were euthanized 1, 3 and 6 days (2 each) after implantation, their parathymic lymph nodes were removed and these glands were implanted under the renal capsule of another two (total 6) tumor-free male rats. These experiments were based on the assumption that if the lymph nodes contained metastatic cells, then their implantation would cause tumor growth within two weeks similarly to the direct implantation of He/De or Ne/De cells. Another series of experiments were modified in such a way that after 1, 3 and 6 days of tumor cell implantation, not the parathymic lymph nodes, but the left kidney carrying the implanted tumor cells was removed. Nephrectomy was followed by a further 13, 11 and 8 days of waiting, respectively. After two weeks (1+13, 3+11, 6+8 days), the parathymic lymph nodes were removed and He/De cell-containing lymph nodes were implanted under the renal capsule of tumor-free male rats, and Ne/De cell-containing lymph nodes in female rats. Tumor growth was measured 2 weeks after parathymic lymph node implantation.
In every case, experimental animals were anesthetized by i.p. administration of 3 mg/100g pentobarbital (Nembutal, Ceva-Phylaxia Rt. Budapest, Hungary). The retroperitoneum was opened by abdominal section, the kidney was exposed, and an India ink- or tumor cell-containing disc, or a lymph node was placed under the renal capsule. Stitches were put in the wound and autopsy and autoradiographic experiments were carried out two weeks later unless otherwise noted.
Whole-body autoradiography. On the 14th day after implantation, 12 control and tumor-bearing rats were anesthetized and the radioligand 18FDG (fluoro-deoxyglucose; 15.0 MBq in 1 ml saline) was injected into the left femoral vein of each rat. Animals were euthanized 60 min after the administration of 18FDG with 300 mg/kg pentobarbital. Each animal was then embedded in an ice-cold 1% carboxymethylcellulose solution. After being frozen in liquid nitrogen, 60 μm-thick cryostate sections (Leica CM 3600 cryomacrotome; Nussloch, Germany) were cut in the sagittal plane. Sections were exposed to phosphor imaging plates (GE Healthcare, Piscataway, NJ, USA). For anatomical correspondence, true-color images of the same sections were also obtained by a transparency scanner (Epson Perfection 1640; EPSON Deutschland GmbH, Meerbusch, Germany). Autoradiography and transmission images were overlain to enable merging of the functional and anatomical information. Phosphor image analysis used the average of 16 sections calculated by using the Image Quant 5.0™ (GE Healthcare, Molecular Dynamics) image-analysing software. Results were expressed in intesity/pixel units (18).
Organ distribution. In experiments where the organ distribution was investigated, six male and six female F344 rats after 14 days of subcapsular transplantation of 106 Ne/De or He/De cells, were anaesthetized with pentobarbital and similarly to the previous series of experiments animals were given i.v. 15.0 MBq 18FDG in saline. After 1 h, blood samples were taken from the aorta. The tumor, kidney, thymus, abdominal rectal muscle (musculus rectus abdominalis) and the parathymic lymph nodes were removed. Three tissue samples were taken from each organ and their activites were measured with a gamma counter (Canberra Packard). The weight and the radioactivity of the samples were used to determine the differential absorption ratio (DAR). DAR was calculated as:
Statistical analysis. Statistical analysis was performed with two-way analysis of variance (ANOVA) and Student's t-test. Results are expressed as mean±standard deviation (SD), p<0.01 was considered as significant.
Results
Transport of abdominal ink particles to thoracic parathymic lymph nodes. In the initial experiments, subcapsular renal implantation of GelasponR disc containing 10 μl India ink was carried out in 8 rats. After 24 h, ink particles occupied large areas of parathymic lymph nodes infiltrated by neutrophils and other inflammatory cells as well. The most pronounced discoloration of lymph nodes seen under the microscope 24 h after implantation with several recognizable ink granules is shown in Figure 1c, 1d.
The transport of India ink particles from the renal capsule to parathymic lymph nodes could be perceived and examined with unaided eye. In the control experiment, where saline without ink was administered, normal parathymic lymph nodes were isolated (Figure 1e). Twenty four hours after India ink implantation, the parathymic lymph nodes were packed with the ink (Figure 1c, 1d, 1f). Control kidney did not contain ink (Figure 1g), but a significant amount of ink remained under the renal capsule after implantation (Figure 1h).
Metastatic potential of He/De and Ne/De cells followed by whole body autoradiography and phosphor image analysis. Autoradiographic experiments revealed that after 14 days, the majority of the radioactivity had accumulated in the renal pelvis and urinary bladder, where the radiotracer is excreted and accumulated before being removed by the urine. Radioactivity of the tumor, the mesenteric lymph nodes and the parathymic lymph nodes surpassed that of other organs (Figure 2). The same results were obtained when Ne/De cells were implanted.
Even higher differences were obtained by phosphor image analysis. By taking the pixel density of resting striated muscle as one unit, the relative pixel densities were, in decreasing order: 14.23 in He/De tumor, 10.82 in parathymic lymph nodes, 5.36 in kidney, 5.16 in thymus, 2.35 in blood and 1.57 in liver (Figure 3).
The experimental design was the same when 106 Ne/De cells were implanted under the renal capsule of four F344 female rats. After 14 days, the same tissue distribution of radioactivity was measured by phosphor image analysis as that of male rats implanted with hepatocarcinoma (Figure 4). Both the autoradiographic images and the pixel-density values resembled those of He/De experiments.
From the two series of autoradiographic and phosphor image experiments, we concluded that He/De and Ne/De tumors grown under the capsule of kidney represent a significant metastatic burden manifested primarily in parathymic lymph nodes.
Metastatic potential of tumor cells in different tissues. Figure 5 illustrates the distribution of 18FDG radioactivity expressed in DAR values in different organs in the control, He/De and Ne/De tumor-bearing rats. The DAR value of Ne/De tumor was 11-fold and that of the parathymic lymph nodes 9-fold higher than that of the muscle. The same experiment carried out with He/De cells resulted in similar results: He/De tumor had 14-fold and parathymic lymph nodes 9-fold higher DAR values than the muscle taken as a standard unit. These experiments support the notion that parathymic glands are involved in metastatic tumor growth (Figure 5).
Appearance of metastases. A series of experiments were carried out to define the time of appearance of metastases. We found that early implantation of lymph nodes of tumor-bearing animals (after one and three days) did not favor tumor formation as the implanted lymph nodes ‘disappeared’. On the contrary, six days after He/De cell implantation, the isolated parathymic lymph nodes induced tumor growth upon implantation. Two weeks after implantation, the tumor weights were 1.5 and 2.0 g, respectively. For comparison, the weight of the parathymic lymph nodes in control rats was somewhat lower (23.5±2.5 mg) (13, 14).
The same experiment was carried out with six F344 female rats with the exception that Ne/De cells (106) were implanted under the renal capsule. Of the lymph nodes implanted in another six female rats, the one- and three-day-old implants ‘disappeared’, while the six-day-old implants induced tumor gowth. The weights of the tumors under the renal capsule 14 days after the implantation of six-day-old parathymic lymph nodes were 0.5 and 0.9 g. We concluded from these results that 106 cells containing primary tumors initiated metastasis formation between day 3 and 6 after implantation. It appears that the slower growing He/De has a higher metastatic potential than the otherwise faster growing Ne/De tumor (Table I).
The ultimate series of experiments were modified in such a way that after tumor cell implantation, not the parathymic lymph nodes, but the left kidney was removed. As in previous experiments, it was assumed that if the lymph nodes contained metastatic cells, then the implanted lymph nodes would induce tumor growth under the renal capsule. Results corresponded to these expectations and to the observations of previous experiments. Parathymic lymph nodes implanted under the renal capsule after 1 or 3 days of nephrectomy were absorbed. On the contrary, lymph nodes removed 8 days after nephrectomy contained a sufficient number of metastatic cells to induce tumor growth after reimplantation (Table II). Moreover, a significant difference was observed between the metastatic potential of He/De and Ne/De cells. These experiments led to the conclusion that primary tumors containing 106 cells induced metastasis formation in about 3-6 days. Although we observed earlier that the primary tumor growth of He/De was always slower, its metastatic character turned out to be stronger than that of the faster growing Ne/De tumor.
Discussion
Our experiments have shown that tumor cells of epithelial (He/De) or mesodermic (Ne/De) origin implanted under the capsule of the kidney generate metastases in the parathymic lymph nodes. This was confirmed by: i) the implantation of ink particles under the capsule of the kidney which appeared within 24 h in the parathymic lymph nodes; ii) whole-body autoradiography experiments of 18FDG uptake, which showed 18FDG accumulated in growing tumors and parathymic lymph nodes; iii) six days after tumor development (He/De, Ne/De), the removed and implanted parathymic lymph nodes behaved as solid tumors. These observations raise the question as to how the detached tumor cells get from the renal capsule to the parathymic lymph nodes.
The method of implanting tumors under the renal capsule was established by Bodgen et al. (19) for the fast screening of chemotherapeutic agents. Further development of this method allowed us to follow the dynamics of isogenic tumor growth (12). In the course of these experiments, changes in the parathymic lymph nodes and the presence of tumor cells in these glands were observed by histological examinations (13).
Appearance of India ink in parathymic lymph nodes implanted under the renal capsule of rat. Pelikan ink or saline (10 μl) was applied to a gelatin sponge that was placed under the renal capsule of F344 rats. Rats (~150 g) were sacrificed after 24 h of implantation and parathymic lymph nodes (PLNs) were isolated. Sections were cut from PLNs isolated as described in the Materials and Methods, stained with hematoxylin and viewed under a microscope. Sections of control parathymic lymph nodes at low (a) (bar, 350 μm) and high (b) magnification (bar, 50 μm). The appearance of India ink traced 24 h after administration in the cortical region of the left parathymic lymph node at low (c) (bar, 350 μm) and high (d) magnification (bar, 50 μm). Macroscopic visualization of India ink transported from the kidney subcapsule to parathymic lymph nodes (e-h). One day after implantation, the parathymic lymph nodes and the kidneys were removed. e) Control parathymic lymph nodes, f) India ink uptake in parathymic lymph nodes, g) normal kidney, h) India ink under the capsule of kidney (bar, 0.5 cm, e-h).
The parathymic lymph nodes of rodents have been neglected contrary to their description in the 1960s (20, 21). In 1967, Blau and Gaugas injected India ink intravenously leading to the observation that parathymic lymph nodes are located inside the thymus capsule in mice, but are sitting on the surface of the thymus capsule in rats, surrounded by a fibrous capsule and brown adipose tissue (22). Interest in parathymic lymph nodes was renewed when Steer and Foot observed in rats that after acute gastroenteritis, the cells of exudative ascites appeared in the parathymic glands (23). This finding directed attention to earlier experiments related to staining and X-ray contrast analysis of materials which clarified that the lymphatic vessels of the peritoneal cavity penetrate the diaphragm, lie on the thoracic aspect of the diaphragm, and consist of three sets. One is the retrosternal vessel running in close proximity to the internal mammary artery (arteria mammaria interna) and reaches the upper mediastinal lymph nodes (24-26). Tilney has mapped the lymphatic system of rats in detail and has described that the lymph to the parathymic lymph nodes comes from the peritoneal cavity, liver, pericardium and from the thymus and pouring its content into the mediastinal lymph trunk (27). Based on these data, it is logical that the metastatic cells of tumors growing under the renal capsule enter first the lymphatic vessels of the diaphragm and then, primarily through the parasternal lymphatic vessels, reach the parathymic lymph nodes. We assume that the capsule of the kidney and the parathymic lymph nodes constitute a complex similar to the Ranke complex after tuberculous infection consisting of peripheral lung lesion with mediastinal lymph nodes. The Ranke complex at the time of its discovery contributed significantly to the understanding of the pathomechanism of the tuberculous infection. Based on this analogy, we intended to develop a relatively isolated system to study the formation of metastases and to widen the range of available methods.
Tissue distribution of 18FDG visualized by whole-body autoradiography in tumor-bearing and tumor-free rats. Tumor cells from He/De cell line were transplanted under the renal capsule of the left kidney and after two weeks of tumor development, 15.0 MBq 18FDG were injected in the left femoral vein. One hour after the injection, the animals were euthanized and whole-body autoradiography was carried out as described in Materials and Methods. a) True-color image of a section obtained with transparency scanner, b) autoradiography of the same section, c) images from a and b superimposed, d) autoradiography of control, tumor-free rat. T, tumor; PLN, parathymic lymph node.
Average distribution of 18FDG in the He/De tumor, major organs and in parathymic lymph nodes. Regions of interest (ROI) were selected from the anatomic image. Pixel densities were normalized to resting muscle which was taken as 1 unit. WT, whole tumor; PLN, parathymic lymph node. Mean value ± SD.
Average distribution of 18FDG in the Ne/De tumor, major organs and in parathymic lymph nodes. Regions of interest (ROI) were selected from the anatomic image. Pixel densities were normalized to resting muscle which was taken as 1 unit. WT, whole tumor; PLN, parathymic lymph nodes. Mean value ± SD.
Tissue distribution of radioactivity expressed in differential absorption ratio (DAR) after intravenous administration of 18FDG in Ne/De or He/De tumor-bearing rats. Metastatic potential of Ne/De and He/De cells, plasma, muscle, tumor, parathymic lymph nodes, thymus, liver and kidney expressed as DAR.
The mass of parathymic lymph node after 14 days of implantation under the kidney capsule.
Metastatic tumor development after implantation of parathymic lymph nodes isolated from nephrectomized tumor-bearing rats.
One can of course argue that the tumor growing under the capsule of the kidney may release metastatic cells to mesenteric lymph nodes as well, and detached tumor cells may reach the parathymic lymph nodes through the blood vessels, as referred to in the experiment of Blau and Gaugas (22). The anastomosis between the renal capsule and parenchymal lymph capillaries is doubted, but has not been excluded, and this would change to some extent the localization of metastases (28). Contrary to these objections, we regard the renal capsule-parathymic lymph node complex as an isolated system which provides an experimental approach to study angiogenesis and the malignant transformation of parathymic lymph nodes. Our system contributes to the understanding of: (a) the metastatic potential of rodent tumors; (b) the connection between the number of primary tumor cells and the temporal aspects of metastates development. By changing the relative mass of parathymic lymph nodes implanted under the kidney capsule, this model offers the opportunity to analyze the nature of primary, secondary and further tumor developments and to pin-point their differences. Finally, our method is suitable for the experimental demonstration of chemical prevention of metastases formation.
Acknowledgements
This work was supported by a grant of the Hungarian National Science and Research Foundation to G.B. (OTKA T42762 grant). The technical contributions by Gyongyi Hadhazi and Tamas Nagy are gratefully acknowledged.
Footnotes
-
Conflict of Interest
It is declared that there is no financial support or relationship which would pose any conflict of interest.
- Received September 19, 2008.
- Revision received January 20, 2009.
- Accepted March 20, 2009.
- Copyright© 2009 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved