Comparative Study of the Cytoplasmic Organelles of Epithelial Cell Lines Derived from Human Carcinomas and Nonmalignant Tissues
Total Page:16
File Type:pdf, Size:1020Kb
[CANCER RESEARCH 40, 803-81 7, March 19801 0008-5472/80/0040-0000$02.00 Comparative Study of the Cytoplasmic Organelles of Epithelial Cell Lines Derived from Human Carcinomas and Nonmalignant Tissues E. Louise Springer Donner Laboratory, University of California, Berkeley, California 94 720 ABSTRACT tissues, tissue peripheral to carcinomas, and carcinomas has been described in recent publications (44, 55). Briefly, 3 mor The cytoplasmic organeblesof 16 human epithebialcell lines phobogical classes of cells were described, normal, abnormal, have been characterized by electron microscopy. The cell lines and very abnormal. The normal cells were nearby uniform in were derived from normal, nonmalignant tissues of cancerous size, shape, and staining properties; the cells from carcinomas organs and from primary and metastatic carcinomas. Every cell were altered and highly variable in size, shape, and staining section on a grid which contained a cleanly defined nucleus, properties; while those from tissue peripheral to carcinomas nucleolus, and cytoplasm was scored blindly utilizing a check were intermediate for these characteristics. Although the mom list of markers. Mitochondniab pleomorphism was expressed phobogical changes were consistently more aberrant in the slightly by normal, to variable degrees by lines derived from malignant cell lines, they were difficult to quantitate. At the nonmalignant tissues of cancerous organs, and to a much ultrastructunal level, alterations of the nucleus and mitochon greater extent by all lines derived from malignant tissues. dna were noted in all of the tumor-derived lines (55). However, Hypertrophied mitochondmiaand longitudinal cnistabarrange there were subtle or overlapping morphological characteristics ment were found in almost all the malignant lines, but not in among the tumor cell lines that precluded quantitative evalua any lines derived from nonmalignant tissues of cancerous tion. Therefore, a more systematic and quantitative ultrastruc organs or from normal tissues. Although malignant and non tural study of this series of human cell lines was initiated. malignant cell lines could not be distinguished using microfila Thirty-eight ultrastructunal characteristics possibly associated ment bundles (stress fibers) and microtubules as markers, this with cancer (19, 54), as well as characteristics associated with may reflect the effect of insufficient cell spread on the prepar metabolic activity were incorporated into an objective appraisal ative procedures used. All the lines appeared differentiated system for assessing the degree of expression of each char and showed slightly to moderately developed Golgi and smooth actenistic in coded specimens. The nuclear analysis of these and rough endoplasmic reticula. There were no significant human cell lines (56) showed that nuclear bodies and penichno ultrastructural differences in cells at different passage bevelson matin granules were found only in cell lines derived from subconfluent and confluent tumor cells; however, more tight carcinomas, while nuclear envelope dilation was seen in cell junctions were observed in confluent than in subconfbuent lines derived from both carcinomas and tissue peripheral to normal cells. carcinomas, but not in any normal lines. Chromatin margina tion, nuclear indentation, nucleolar fibnibbar centers, and nu INTRODUCTION cleolar manginationwere expressed slightly by the normal lines, to variable degrees by the lines derived from tissue peripheral Studies on the in vitro properties of rodent epithelial cells to carcinomas, and to a much greater extent by all carcinoma have consistently shown subtle morphological changes after derived lines. carcinogen treatment. These light microscopic observations The cytopbasmic portion of the objective appraisal system made on living and fixed-stained cultures showed several fea observations of these human cell lines is presented in this tumes:reduced intercellular cohesiveness (29, 68) and altered paper. colony formation (64, 68); pleomomphismof cell size and shape (29, 40, 41 , 65, 68), as well as nuclear size and shape (41, MATERIALS AND METHODS 52) increased nucleam:cytoplasmic ratio (40, 68); prominent, enlarged, or numerous nucleoli (41 , 52, 68); abnormal mitoses Cell Culture. The cell lines2and their tissue sources are (29, 52); and elongation of cells (68). Cytopbasmic changes listed in Table 1. All cell lines, except HeLu, were initiated at other than basophilia (52) have not been found at the light the Cell Culture Laboratory, University of California, Naval microscope level because standard fixation procedures do not Biological Laboratories, Oakland, Calif. HeLu was generously preserve cytoplasmic structures and many of the onganebbes provided by Dr. Daniel Rifkin, Rockefeller University, New York, are below the resolution limits of the microscope. Comparative N. Y. The normal epithebiabcell lines were derived from fetal studies of morphology at the light microscope bevelhave been intestine and an adult bladder from a patient with prostate difficult to quantitate. carcinoma. Two normal fibroblastic cell lines, from fetal lung Similar morphological differences at the light microscopic and foreskin were included also. Three lines derived from level have also been noted in human epitheliabcells in culture. kidney tissue peripheral to a carcinoma of the same organ were The morphology of human epithebialcells cultured from normal evaluated separately since hyperplasias are often found in the @ I This work was supported by Contract CP-7051 0 from the National Cancer 2 Nomenclature used conforms with that of Federoff (1 4): a cell line' ‘arises @ Institute. from a primary culture at the first subculture, while an established cell line is Received March 15, 1979; accepted November 26, 1979. one which has demonstrated the potential to be subcultured indefinitely. MARCH 1980 803 Downloaded from cancerres.aacrjournals.org on September 28, 2021. © 1980 American Association for Cancer Research. E. L. Springer Table 1 History of patients and biopsy specimens (yr)RaceDiagnosisCarcinomas785TM58ColonDesignationSexAge carcinoma675TMNo carcinoma761 informationColon (kidney)769TF59NTransitionalTM65C9Transitional cell carcinoma (urethra)578TF74CCarcinosarcoma cell carcinoma breast766TM64CPancreatic of carcinoma metastatic to lymph node746TM74CStomach muscle700TM61CCarcinoma carcinoma metastatic to (probably colon, intestine, or pan hip696TM430Adenocarcinoma creas) metastatic to (probably colon, intestine, or sacrumPeripheral pancreas) metastatic to tissue of carcinoma tousorgans699KM60CNormal kidney tissue from patient with renal carcinoma761 cell KM65CNormal kidney tissue from patient with transi tional cell carcinoma kidney (same patient 761T)71 as 5KM76Normal kidney tissue from patient with renal carcinomaNonmalignant tissues6771nt3—4 disorder68Olnt3—4 mo. fetusFamilial history of immune syndrome741nt3—4 mo. fetusFamilial history of Wiskott-Aldrich abnormalities767BLM45CNormal mo. fetusTherapeutic abortion, no known bladder mucosa from patient with carcinomaa prostate c, Caucasian; N, negro; 0, oriental. uninvolved tissues of carcinomatous organs (13, 66). The so that the electron microscope operator was not aware of the malignant lines were derived from primary carcinomas of the cell source. To ensure that a cell was scored only once, the rectum, colon, breast, and transitional cells of the urethra and serial sections were surveyed to locate and identify each cell kidney, as well as metastatic lesions of pancreas, stomach, profile in the randomly chosen sample; then every cell in the and 2 lines derived from metastatic tissue of unknown origin. sample was evaluated. Scoring was accomplished by identify The growth medium used was Dubbecco's modification of ing the presence, absence, or degree of expression of various Eagle's medium (Grand Island Biological Co., Grand Island, cytoplasmic ultrastructunal characteristics for every cell section N. Y.; No. 196G) containing glucose (4.5 g/Iiten), supple showing a nucleus, nucleolus, and cytoplasm. Although most mented with 10% fetal bovine serum and insulin (10 @zg/ml; of the scoring was done directly at the electron microscope Calbiochem, San Diego, Calif.). Unless otherwise indicated, console, some photographs were taken of each specimen. cells were harvested at confluence. All of the lines, both tumor Similar results were obtained when the photomicnographs were and those derived from nonmalignant tissue, grew slowly in scored blindly. culture (44). There was no difference in growth rate as a Statistical Analysis. Tests for statistical significance were function of nuclear (56) and cytoplasmic ultrastructure. made using the Fisher exact probability test (43). A difference Electron Microscopy. The cell lines cultured in Falcon flasks was considered significant if p was 0.025 or less. were fixed in situ with 2.5% glutarabdehyde in 0.1 M sodium cacodylate buffer (pH 7.3) at room temperature for 3 hr, rinsed RESULTS with buffer, and fixed with 2% osmium tetroxide in the same buffer for 2 hr at room temperature. Both fixatives were nou Plasma Membrane Specializations. Three features of the tinely monitored for any indication of deterioration which could plasma membrane were evaluated: microvilbi, desmosomes, adversely affect cell preservation, i.e. , pH of glutanaldehyde and tight junctions. The microvibli which occur on the apical and color changes of the osmium tetnoxide. En b!oc staining and lateral surfaces of the cell as finger-like