DOI 10.1007/s10517-015-3110-7 Bulletin of Experimental Biology and Medicine, Vol. 160, No. 1, November, 2015 MORPHOLOGY AND PATHOMORPHOLOGY
109
Morphological Criteria of Cell Differentiation Stages in Experimental Hepatocarcinoma and Evaluation of Antitumor Drug Efficiency N. P. Bgatova, L. V. Omel’yanchuk*, A. A. Pozhidaeva, V. F. Semeshin*, A. P. Lykov, O. V. Poveshchenko, O. P. Makarova, L. N. Rachkovskaya, Yu. I. Borodin, and V. I. Konenkov Translated from Byulleten’ Eksperimental’noi Biologii i Meditsiny, Vol. 160, No. 7, pp. 126-132, July, 2015 Original article submitted July 17, 2014 Structural polymorphism of 5 cell differentiation stages of hepatocarcinoma-29 from ascitic fluid is detected and the morphological criteria for identification of these stages are defined on the base of optic and electron microscopy findings, cytofluorometry, and DNA cytometry. The percentage of cells at differentiation stages 4 and 5 in the tumor structure increases after hepatocarcinoma cell inoculation into the hip. Injection of a cell cycle-modulating substance to animals with tumor growth shifts the proportion of cells with various differentiation stages. The morphological criteria of 5 stages of hepatocarcinoma-29 cell differentiation can be used for prospective drug testing. Key Words: hepatocarcinoma-29; morphological criteria; differentiation stages Cell composition of a tumor can be heterogeneous [2,10]. The term “cancer stem cells” is now sometimes used when speaking about tumor tissue. These cells are assumed to be capable of long survival under conditions of various therapies and are expected to serve as drug targets [9]. On the other hand, according to the stochastic hypothesis of cancer development, the microenvironment and tumor cells (TC) without characteristics of stem cells, but modulating the stem TC kinetics and the time course of tumor development, play an important role in the tumor progress [8,11]. It is therefore essential to detect the stem TC and to carry out phenotyping of all TC in the population in order to detect the markers to be used for evaluating the effects of therapies. Hepatocarcinoma (HC) is one of the most aggressive human tumors. Despite the progress in its diagnosis and therapy, it still ranks fifth by its incidence and third by mortality in the world, because of its drug Research Institute of Clinical and Experimental Lymphology, Siberian Division of the Russian Academy of Medical Sciences; *Institute of Molecular and Cellular Biology, Siberian Division of the Russian Academy of Sciences, Novosibirsk, Russia. Address for correspondence:
[email protected]. N. P. Bgatova
resistance [13]. Like other solid tumors, HC is characterized by high heterogeneity of TC population [7]. Detection of the genetic markers of tumor variants and search for specific morphologic signs of the tumor, the use of which could become routine in evaluation of the prognosis and treatment efficiency, is an important problem [2]. Morphological criteria characterizing the tumor composition and differentiation of TC are essential for the choice of a therapeutic strategy (including the target therapy) aimed at tumor growth arrest and for understanding the contribution of certain cells to metastatic processes. Experimental models are used for development of methods for target therapy of tumors. The most convenient models for studies of cell heterogeneity are ascitic tumors [6]. Experimental HC-29 tumor is a convenient model for studies of malignant growth; its continuous strain is maintained in ascitic form in CBA/LacYIcgn mice [3]. However, in order to use HC-29 as a model, the structure and functions of the cells constituting this tumor should be studied. We studied structural heterogeneity and detected morphological criteria for HC-29 cell differentiation
0007-4888/15/16010109 © 2015 Springer Science+Business Media New York
110
Bulletin of Experimental Biology and Medicine, Vol. 160, No. 1, November, 2015 MORPHOLOGY AND PATHOMORPHOLOGY
stages in vitro and in vivo: for HC-29 cells from ascitic fluid (AF) and under conditions of TC transplantation into the hip muscles of experimental animals.
MATERIALS AND METHODS HC-29 was derived and verified at Institute of Cytology and Genetics [3]. The HC-29 cells were transplanted to CBA mice (n=10) into the abdominal cavity. AF was collected after 10 days. The HC-29 cell suspension from AF was fixed in 10% neutral formalin, processed by the standard methods, and paraffin blocks were prepared, after which the morphology of TC was studied. The sections (5 μ) were stained with hematoxylin and eosin and after Feulgen [12] in order to evaluate cell ploidy. The content of DNA in the nuclei stained with Schiff’s reagent was measured cytophotometrically using digital microphotography [5]. The measurements were carried out under an AxioLab (Carl Zeiss) microscope in transmitting light using A-plan ×40 objective with 0.65 aperture and AxioCam ICm1 camera in a 12-bite dynamic range. The images were processed by ImageJ software with digital processing by original software in the Mathcad medium. The cell cycle and apoptosis level of HC-29 cells was studied by dual-color cytofluorometry by propidium iodide staining of the cells (Becton Dickinson). The cells were cultured in 5 ml RPMI-1640 with 10% fetal calf serum in a Petri dish at a concentration of 106/ml for 24 h. The cells were then washed in buffered saline, concentrated, and the cell cycle and apoptosis were studied. The cell cycle was analyzed by DNA histograms. The percentage of cells with diploid (cells in G0/G1 phases) and hyperploid (cells in the S and G2/M phases) DNA set was evaluated in the TC gate. For ultrasonic analysis, suspension of HC-29 cells from AF was fixed in 1% OsO4 solution in phosphate buffer (pH 7.4), after which the samples were processed according to the standard protocol for electron microscopy [1]. Analysis of HC-29 cell polymorphism in vivo was carried out in male CBA mice (n=10) 20 days after injection (into the right hip muscle) of TC from AF, suspended in 10-fold volume of saline (0.1 ml). The animals were handled in accordance with Regulations for Studies with the Use of Experimental Animals. The possibility of using the TC morphological criteria for testing the cell proliferation-blocking drugs was evaluated using lithium citrate, as according to some data, lithium compounds modified the cell cycle signal pathways and regulation [15]. After induction of the tumor process, mice (n=10) were intraperitoneally injected with lithium citrate (0.92 mg in 0.1 ml per animal) for 5 days. Material for studies was collected
after 20 days. The animals were sacrificed by cervical dislocation. Specimens of tumor tissue for morphological studies were processed by the standard protocol for electron microscopy [1]. Semithin sections (1 μ) were stained with toluidine blue and examined under a Leica DME microscope. Ultrathin sections (35-45 nm) were contrasted with uranyl acetate and lead citrate and examined in a JEM 1010 electron microscope. Morphometric studies on microphotographs were carried out using ImageJ software. Volume densities of the TC nuclei and cytoplasms were evaluated in an open test system at a 0.7 μ step at electronogram magnification 12,000, and the nucleus/cytoplasm ratio was calculated. Volume density of granular endoplasmic reticulum (GER) cisterns in TC cytoplasm was evaluated in a closed test system of 154 points at a 100 nm step at magnification 30,000. The digital data were processed by routine statistical methods with Student’s t test.
RESULTS Light microscopy showed that HC-29 cells from AF differed by cell and nucleus size and cytoplasm contents (Fig. 1, a). Analysis of DNA content in interphase and metaphase HC-29 cells showed two peaks in metaphases (Fig. 1, b). One peak corresponded to metaphases with DNA content of 8.0-8.5 pg, the other to 15.0-16.5 pg. As is known, the metaphases contain a tetraploid chromosome set and hence, the diploid set for G1 cells is 4.0-4.2 pg DNA. This value corresponded to peak 1 of interphase cells (4.0-4.5 pg). At least a part of interphase peak at 7.5-8.5 pg was presented by G2 cells with diploid DNA of 4.0-4.2 pg. The metaphase peak of 15-16 pg corresponded to G1 diploid cells with 7.5-8.2 pg DNA, i.e. G1 peak of these cells overlapped with G2 peak of cells with 4.0-4.5 pg DNA. The interphase peak at 15.5 pg corresponded to G2 cells (Fig. 1, b). Hence, the second fraction of proliferating cells contained 2-fold more DNA than the first fraction, i.e. was autoploid relative to the first fraction. In addition to these peaks, the interphase curve had one more peak of cells with 12 pg DNA. As there were no metaphase cells with DNA content equal to this value or 2-fold lower, we hypothesized the existence of the third HC-29 cell population incapable of forming mitotic figures. The predominance of diploid cells in HC-29 population and their distribution by the cell cycle phases was verified by dual-color flow cytofluorometry. The percentage of cells with diploid chromosome set (G0/ G1 cells) was 75.2%, of cells with hyperploid DNA set (S, G2/M cells) – 12%. Apoptotic cells with fragmented DNA formed a characteristic hypodiploid peak and constituted 8%.
N. P. Bgatova, L. V. Omel’yanchuk, et al.
111
Fig. 1. Heterogeneity of HC-29 cell population. a) HC-29 cells of different size. Hematoxylin and eosin staining, ×400; b) content of DNA in interphase and metaphase cells of transplanted HC-29, evaluated by digital cytometry; c) morphological types of HC-29 cells at various differentiation stages. Toluidine blue staining, ×900; d) ultrastructural organization of HC-29 cells of various differentiation stages, ×4000.
Analysis of the nucleus/cytoplasm proportion and ultrastructural organization of TC showed the heterogeneity of HC-29 population by these parameters. Five types of tumor cells were distinguished by their nucleus/cytoplasm ratio and content of intracellular organelles (Fig. 1, c, d). Changes in the ultrastructural organization of TC with increase of the cytoplasm volume percentage characterized these cells as cells at various stages of differentiation, the content of GER cisterns in the cytoplasm serving as one of the criteria of cell differentiation stage [4]. Evaluation of the levels of AF cells at various stages of differentiation showed that cells at differentiation stage 1 constituted 6%, 20% cells were at stage 2, 35% at stage 3, 29% at stage 4, and 10% of TC population at stage 5. The cytoplasm of stage 1 cells (nucleus/cytoplasm ratio 0.797±0.008) contained solitary mitochondria and poorly discernible GER membranes; free polysomal ribosomes predominated (Fig. 1, d; Fig. 2, a). The cytoplasm of differentiation stage 2
cells (nucleus/cytoplasm ratio 0.681±0.005) contained, in addition to the above organelles, solitary lipid droplets, while the volume density of the GER cisterns was 2.5 times higher (Fig. 1, d; Fig. 2, b, f). Stages 3 and 4 cells (nucleus/cytoplasm ratio 0.596±0.005 and 0.492±0.004, respectively) were characterized by a higher percentage of the cytoplasm, accumulation of mitochondria, GER membranes, membrane-bound and free polysomal complexes, lysosomes, and lipid incorporations (Fig. 1, d; Fig. 2, c, d). The volume density of the GER cisterns in cells at differentiation stages 3 and 4 was 3- and 4-fold higher, respectively, than in stage 1 cells (Fig. 2, f). Cells at differentiation stage 5 (nucleus/cytoplasm ratio 0.384±0.006) were characterized by a greater volume of the cytoplasm with just few GER membranes and mitochondria, with lipid droplets, and high content of free polysomal complexes (Fig. 1, d; Fig. 2, e, f). Twenty days after HC-29 cell injection into the hip muscle the tumor consisted of large cells forming
112
Bulletin of Experimental Biology and Medicine, Vol. 160, No. 1, November, 2015 MORPHOLOGY AND PATHOMORPHOLOGY
Fig. 2. Levels of GER cisterns in the cytoplasm of HC-29 cells at various stages of differentiation. a) Differentiation stage 1; b) stage 2; c) stage 3; d) stage 4; e) stage 5. ×10,000 (a-e). f) Volume density of GER cisterns in HC-29 cells. *p
