A Method of Staining Nucleoli of Cells in Fresh Benign and Malignant Tissues* Commander W

A Method of Staining Nucleoli of Cells in Fresh Benign and Malignant Tissues* Commander W. W. Ayres, (MC), U.S.N. (From the Naval Medical School, 2Vational Naval Medical Center, Bethesda, Maryland)

(Received for publication July 16, 1947)

INTRODL~CTION lene blue or derivatives of methylene blue have With the increasing interest in the diagnosis of been used to produce the elective staining of nu- malignant cells in various exudates and body fluids, cleoli. No systematic studies have been done on a method demonstrating certain nucleolar changes the homologues of methylene blue as to their abil- in considerable detail may be of value. It is the ity to stain nucleoli. In the development of the purpose of this paper to present such a method and azure C method this was done. Methylene blue, to discuss briefly the nucleolar changes in benign azure A, azure B, azure C, and thionin were studied and malignant tissues and also to point out the for their staining qualities for nucleoli in many potential applicability of this method as an aid in fresh benign and malignant tissues. It was found the diagnosis of cancer. that azure C gave the best results in that the nu- The azure C method is based on the observation cleoli showed a sharper outline and increased meta- that nucleoli of cells of fresh tissues when exposed chromatic effects, and there was less tendency of to certain dyes, stain electively. Naegeli (17, I8) the nucleus to be overstained as compared to the states that the nucleoli of lymphocytes and myelo- other dyes. Azure B gave almost as good results. blasts often stain well with supravital technics as Azure C, or mono-methyl thionin, was first pre- described by Pappenheim, Nakanishi, Sabrazes, pared by Holmes and French (10) in 1926, by the Schilling-Torgau and Cesaris-Demel. These meth- oxidation of methylene blue under acid conditions. ods utilize methylene blue, brilliant cresyl blue or TECHNIC Giemsa stain. Quensel (21, 22), in particular, has Preparation o] the stained slides.--Microslides studied the nucleoli of malignant cells in pleural containing a fine precipitate of azure C are pre- and ascitic fluids using a staining mixture of methy- pared as follows: Solutions of azure C in 0.25 per lene-blue-cadmium and Sudan III-cadmium. He cent and 0.50 per cent concentrations in absolute states that with his method the nucleoli stain bet- methyl alcohol are prepared. A drop of one of ter than with any other method. MacCarty (13, these solutions is placed on a clean microslide and 14), and MacCarty and Haumeder (15), in exten- quickly covered with another slide. The stain dif- sive work on the nucleoli and the nuclear-nucleolar fuses between the slides. Excess stain is removed ratio of benign and malignant cells, recommend from the edges with a cloth. The slides are then Terry's or Unnas' polychrome methylene blue for pulled apart with a rapid sliding motion in the the staining of these bodies in frozen sections. long axis of the slides. The alcohol rapidly dries, MacCarty states that the nucleolus may be better leaving a fine precipitate of dye. After marking demonstrated in fresh tissues than in fixed tissues. one corner of the slides with a wax pencil for iden- Guzman (7, 8), in a study of the nucleoli of lym- tification of the stained surface, they are stored phocytes and monocytes of the peripheral blood, until required. Slides are made with both concen- developed a method utilizing diluted Leishman or trations of the dye. Giemsa stain. Von Haam and Alexander (25, 26), Technic 1o~ tissues--A new cut is made in fresh in their work on the cytology of malignant tumors, tissue and the surface lightly scraped with a scal- developed a method of staining nucleoli of fresh pel. The small amount of tissue thus obtained is unfixed cells in suspension, with high dilutions stirred into a drop of fresh human blood serum (1 : 10,000) of toluidine blue. which has been previously placed on a coverslip. In most of the methods mentioned above, methy- The tip of the scalpel is used for stirring. The * This article has been released for publication by the coverslip is inverted over the stained slide. This Division of Publications of the Bureau of Medicine and results in diffusion of the cell-suspension between Surgery of the United States Navy. The opinions and views set forth are those of the author and not to be con- the two surfaces. Pressure on the coverslip is sidered as reflecting the policies of the Navy Department. avoided. The coverslip is rimmed with vaseline to

352

prevent drying. The nucleoli usually stain within erslip to free the cells from the stroma. Bone mar- a few minutes. The slide is examined under the row obtained by aspiration need not be mixed with oil immersion lens immediately after preparation. serum, whereas bone marrow obtained at autopsy Certain precautions are necessary to obtain good should be so mixed. results. The drop of serum should be very small RESULTS so that with application of the coverslip to the slide, a very thin film of the cell suspension, which The nucleolus stains a brilliant uniformly dark does not quite reach the edges of the coverslip, is azure and is in striking contrast to the relatively obtained. The amount of tissue, likewise, should unstained nucleus. The outline of the nucleolus be very small so that the cells are somewhat flat- is sharply delineated; fine strands of nucleolar ma- tened and no more than one cell layer in thickness. terial may be seen between some nucleoli. Small Clumps of cells are to be avoided, since the cells nucleoli, which are often obliterated by heavily within the interior of the clumps do not stain. stained blocks of basophilic chromatin of the nu- Stained slides prepared with the 0.50 per cent con- cleus in fixed preparations stained with the usual centration of the dye are used for tissue. If the methods, are well seen by this method. Vacuoles nucleoli do not stain, or if they stain poorly, the within the nucleolus are often noted. No staining usual faults are: too much serum, too heavy a sus- distinction between plasmosomes and karyosomes pension of cells for the amount of dye, too light a is noted. The nucleus does not stain, or stains a concentration of dye on the slide or insufficient in- light azure, light green, or light blue, depending terval of time for staining. If the nucleus over- on the type of cell. The cytoplasm usually assumes stains, the usual faults are: too heavy a concentra- about the same color as the nucleolus, but may not tion of dye on the slide for the number of cells, or stain or may assume a light azure to dark blue de- autolyzed tissue. For best results, tissues should pending on the type of cell. The border often is be used immediately after removal from the body, not seen in cells removed from tissue. Staining since the differential staining between nucleolus takes place in I to 15 minutes. and nucleus is not as distinct when old tissues are Although the preparations are not permanent used. The best results are obtained with cells free they may be kept in fair condition for several days in fluid, tissues in which the cells have not adhered if refrigerated. Degenerative changes vary consid- to the stroma, as found in many carcinomas. The erably with the cell type. For example, the mono- so-called "tumor milk," found on the cut surfaces cyte of the peripheral blood degenerates rapidly of carcinomas, is especially suitable for study by and there is a tendency for the nucleus to be over- this technic. Good results are not obtained with stained, obscuring the ? to 5 very small nucleoli dense tissues in which cells are loosened with diffi- contained in this cell. In contrast, there is little culty from the stroma. Trauma often results in tendency for the nucleus to be overstained in malig- distortion of the cells, and necrotic areas within a nant cells and the period of profitable observation tumor are to be avoided. Several parts of a tumor may last for hours. In general, more immature should be sampled to reduce the possibility of miss- cells show less degenerative changes than do more ing the malignant portion of the lesion. mature cells. The cytoplasm undergoes degenera- Staining of blood.--The finger is pricked and a tion more rapidly than does the nucleus, and the small drop of blood is placed in the center of a nucleus more rapidly than the nucleolus. Cells in coverslip which is then inverted over the stained mitosis are easily recognized. The chromosomal mass takes a light blue stain and the nucleolar frag- slide. Slides prepared with the 0.25 per cent solu- ments continue to take the dark azure stain. Indi- tion of dye are ordinarily used, but for leukemias a vidual chromosomes are not stained. Quensel (22), higher concentration of the dye may be required. with his method, states that mitotic figures are not Staining of transudates, exudates and aspiration readily stained. biopsies.--The preparations are made as described It has been observed in actively metabolizing be- above, using the whole fluid or the sediment ob- nign and malignant tissues and particularly in the tained after centrifugation, with or without dilution blood in leukemia, that after once staining, the cells of the material with serum depending on its cellu- may quickly become decolorized. This phenomenon larity. The slides prepared with the 0.50 per cent has also been observed by Simpson (24), using bril- dye concentration are used but it may be necessary liant cresyl blue, and by Quensel (22), using to vary the concentration of the dye. In aspiration methylene blue. The dye is reduced to a leukobase, biopsies, if bits of tissue are obtained, these should probably by enzymatic action of the tissue. The be gently rubbed into the drop of serum on the coy- process is reversible and restaining will occur if one

corner of the coverslip is lifted, permitting the en- facts produced by fixation and embedding are trance of air to the cells. Of considerable interest avoided. is the work of Barron (1), who states that methyl- The disadvantages are: (a) impermanence of the ene blue, to which azure C is related, stimulates preparations; (b) degenerative changes occurring oxygen consumption of normal tissues, having aero- in unfixed tissues with the production of artefacts; bic glycolysis, and of tumors. (c) in the study of malignancy, the failure to see A detailed description of specific ceils is not indi- structure of the tumor, its relation to stroma or its cated here but a brief description of a few common invasive properties; (d) sampling of only a small cells found in serous exudates will be given. part of the tissue with the attendant possibility of The red cells do not take the stain. The reticu- missing the lesion. This method, however, comple- lum of the reticulocytes takes a light brown stain. ments other methods and should be used in con- The small lymphocyte (Fig. 7) contains usually 1 junction with other histopathological technics. small round eccentrically placed nucleolus and oc- The nucleoli of malignant cetls.mThe nucleolar casionally 2, but no more than 2, nucleoli. There changes of malignant cells have been studied in is a thin rim of azure cytoplasm around the rela- particular by Pianese (20), QuenseI (21, 22), Mac- tively unstained round nucleus. The monocyte of Catty (13, 14), MacCarty and Haumeder (15), the peripheral blood contains 2 to 5 small nucleoli Fidler (5), Castren (2), Saxen (23), Karp (11), about half the size of those found in the small lym- Zadek (27), and von Haam and Alexander (26). phocyte. The nucleus is approximately oval in shape, Among the nucleolar changes in malignant cells whereas the cytoplasm is relatively abundant and described by these authors are: (a) increase in size takes the azure stain. The mesothelial cell (Fig. 9) of the nucleolus out of proportion to the increase contains usually 1 small round dark azure-stain- in size of the nucleus; (b) irregularity in the shape ing nucleolus in a round to oval nucleus and a of the nucleolus; (c) increase in the number of nu- relatively small amount of azure-stained cytoplasm. cleoli; (d) variation in the number of nucleoli. Macrophages in fluids show an oval to elongated There are two other features which the author has nucleus, abundant cytoplasm that is often vacuo- observed in some malignant cells: (a) the presence lated, and 1 to 2 small oval to round nucleoli. They of fine nucleolar strands between nucleoli, mor- often contain ingested material and cells. Poly- phological evidence of nucleolar division; (b) loss morphonuclear leukocytes may be recognized by of polarity of nucleoli (the nucleoli are arranged at the shape of the nucleus, the fine cytoplasmic gran- right angles to the long axis of an elongated nu- ules and the absence of the nucleoli. Basophils cleus). None of these changes are pathognomonic and mast cells show bright red granules, while in of malignancy. Von Haam and Alexander (26) eosinophils, some of the granules have a light azure state "neither size, shape or number of nucleoli rep- tint. The plasma cell shows an oval, green-tinted, resent any common characteristic for malignant nucleus with usually 1 eccentrically placed small cells." Quensel (22) states, "by my method pecu- round nucleolus, but it may contain up to 3 nu- liar changes in the nucleoli of malignant cells in cleoli. The cytoplasm is abundant, light azure and serous exudates are so regularly found that when finely vacuolated. Normoblasts show a pale blue these changes are marked, they represent an en- stained nucleus. Fat does not stain. Melanin pig- tirely characteristic sign of malignancy." He fur- ment of malignant melanoma stains dark blue. ther states that all malignant cells may not show these nucleolar changes. At present it is the con- DISCUSSION sensus that there is no common pathognomonic Advantages and disadvantages of the method.m morphologic feature of the malignant cell. The advantages of the method are (a) nucleoli may Although the nucleolar changes described above be demonstrated with detail not ordinarily seen in are not pathognomic of malignancy, they are of other types of preparations using fixed and em- great help in arriving at a diagnosis. It has been bedded tissues; (b) rapidity of the method; (c) applicability to exudates, aspiration biopsies and observed that most non-neoplastic tissues and be- other types of fluid or semifluid tissues; (d) com- nign tumors contain from 1 to 3 small round plete cells are seen and the total number and com- nucleoli. However, Von Hamm and Alexander plete outline of the nucleoli may be noted, in dis- found large nucleoli in the corpus luteum cells tinction to sectioned tissues; (e) certain crystals of the ovary, and Ferreira (4) has noticed similar such as Charcot-Leyden crystals, uric acid crystals changes in benign tissues. He observed in the or cholesterol crystals are seen (they are usually cells of the fetus, normal liver, membrana granu- dissolved when other technics are used); (f) arte- losa of the ovary and in exophthalmic goiter, an

increase in the area of the nucleolus and a de- contains one small round nucleolus, as does the crease in the nuclear-nucleolar ratio which approxi- macrophage. In sharp contrast, many malignant mate those of cancer. In the present study, some cells contain large, bizarre-shaped and multiple nu- of the nucleolar changes already mentioned have cleoli. The absence of these changes does not rule been found in rapidly growing benign tissues such out malignancy completely, since the tumor must as the megaloblasts of pernicious anemia, abnor- be on a cavity surface and cells must be sloughing mal lymphocytes in infectious mononucleosis, rap- into the cavity in order that the malignant cells idly proliferating endometrium, rapidly growing may be found in the fluid. This has been pre- fetal tissues and fibroblasts of young granulation viously emphasized by Helwig (9). The nucleolar tissue. On the other hand, small, round, apparent- changes described above should be extensive, and ly benign nucleoli have been found in lymphosar- should be present in many cells, to be o] aid in the coma of the lymphocytic type, in small cell bron- diagnosis o] malignancy. chogenic carcinoma and in neuroblastoma. Lud- At present 200 tissues have been studied at the ford (12) summarizes the matter when he states, Naval Medical School and high diagnostic accuracy "actively growing cells are characterized by their has been achieved. As more data are collected large nucleoli, and by the fragmentation and divi- they will be the subject of a later report. sion of same. In the adult, nucleoli of cells under- The azure C method may be used for a more going active metabolism are relatively large; those accurate description of nucleoli of cells. In general, in the 'resting' condition relatively small." It is nucleoli are rather vaguely described chiefly be- highly probable then that the nucleolar changes cause fixed tissues and fixed smears of tissues are herein described may be found in any rapidly pro- employed. In a fixed preparation there is precipi- liferating tissue, of which malignant tissue is a tation c,~ chromatin on the inner wall of the nuclear notable example. membrane in the lining network and on the outer Application of the azure C method.--The usual surface of the nucleolus. This precipitation of nu- histopathologic methods are adequate in the diag- clear material obscures the outline of the nucleolus nosis of malignancy in tissue and are to be pre- which gives a false picture of size, obliterates fine ferred to scraping of tumors, in which only indi- nucleolar strands and may entirely hide some of the vidual cells or groups of cells are found. Even smaller nucleoli. here, however, more information of the finer cytol- In frozen sections cellular detail is not well seen ogy may be obtained by means of the azure C and it is necessary to wait 24 hours or so for the stain. On the other hand, considerable difficulty paraffin sections. With the azure C method, this has long been experienced in the diagnosis of malig- difficulty may be party obviated since considerable nancy in the individual cells found in various body cellular detail may be seen at the time the frozen fluids and exudates. The sediments of the fluids section is done. The azure C slide may be read are embedded and sectioned with difficulty. The often within 5 minutes, and in conjunction with alternative methods of studying these cells are by the frozen section. the preparation of smears stained with hematoxylin SUMMARY and eosin, or Papanicolaou's technic (19), or the method of Dudgeon and Patrick (3), or by embed- 1. A method is presented by which the nucleoli ding and sectioning the sediment according to the of cells of fresh benign and malignant tissues may method described by Mandelbaum (16) and Ze- be observed in great detail. Cell suspensions in mansky (28). Although a fairly high degree of serum are prepared on a coverslip which is then inverted over a slide containing a fine precipitate of accuracy in diagnosis may be obtained by these azure C dye. The nucleoli stain clearly and elec- methods, difficulty is still experienced in the separa- tively, the nucleus does not stain deeply, permitting tion of macrophages and mesothelial cells from contrast between the nucleus and nucleolus. Ad- some malignant cells. In this regard, Foot (6) vantages as well as disadvantages of the methods states, "mesothelial pleural, pericardial and perito- are described. A brief description of a few com- neal covering cells present the chief obstacle in the mon cells as they appear with this method is given. way of successful diagnosis as they are readily 2. The nucleolar changes in malignant tissues confused with tumor cells on account of their large and fast growing tissues are compared with those size and prominent nucleus." It is in the differen- of benign tissues. tiation of these cells from malignant cells that the 3. The value of the method as an aid in the dif- azure C method is particularly applicable. With ferential diagnosis of benign and malignant cells the azure C method, the mesothelial cell usually in serous effusions is discussed.

ACKNOWLEDGMENT 14. MACCARTY, W. C. Further Observations on the I wish to thank Dr. J. R. Carter and Dr. S. P. Hicks for Macronucleolus of Cancer. Am. J. Cancer, 31:104- their assistance in this work. 106. 1937. 15. MACCARTY, W. C., and HAUMEDER, EVA. Has the REFERENCES Cancer Cell any Differential Characteristics? Am. 1. BARRON, E. S. G. Catalytic Effect of Methylene Blue J. Cancer. 20:403-407. 1934. on the Oxygen Consumption of Tumors and Normal 16. MANDLEBAUM, F. S. The Diagnosis of Malignant Tissues. J. Exper. Med., 52:447-456. 1930. Tumors by Paraffin Sections of Centrifuged Exu- 2. CASTREN, H. Ueber die Struktur der Zellen der dates. J. Lab. & Clin. Med., 2:580. 1917. Bindegewebsgeschwuelste beim Menschen. Arbeit. 17. NAEGELI, O. Blutkrankheiten und Blutdiagnostik. aus dem pathol. Inst. Univ. Helsingfors., 4:240- 4th Aufl. Berlin: Julius Springer. 1923, pp. 23-26. 318. 1926. 18. NAEGELI, O. Blutkrankheiten und Blutdiagnostik. 3. DUDGEON, L. S., and PATRICK, C. V. A New Method 2nd Aufl. Leipzig: Veit und Comp. 1912, pp. 38-41. for the Rapid Microscopical Diagnosis of Tumors, 19. PAPANICOLAOU,G. N., and TRAUT, H. F. Diagnosis of with an Account of 200 Cases so Examined. Brit. Uterine Cancer by the Vaginal Smear. New York: J. Surg., 15:250-261. 1927. The Commonwealth Fund. 1943, pp. 3-7. 4. FERI~EIRA, A. E. M. Relation of the Macronucleolus 20. PIANESE, G. Quoted by yon Haam and Alexander. and Nucleo-Nucleolar Ratio of Cancer to Histologic 21. QU~NS~L, U. Zur frage der Cytodiagnostik der Grading. J. Lab. & Clin. Med., 26:1612-1628. 1941. Ergusse seroser Hohlen. (Methodologische und 5. F~VLER, H. K. A. Comparative Cytological Study of path. anat. Bemerkungen) Acta med. scandinav., Benign and Malignant Tissues. Am. J. Cancer, 63:427-457. 1928. 25:772-779. 1935. 22. QUE~SEL, U. Zytologische Untersuchungen yon 6. FooT, N. C. The Identification of Tumor Cells in Erguessen der Brust- und Bauchhoehlen, mit Sediments of Serous Effusions. Am. J. Path., 13: besonderer Beruecksichtigung der karzinomatoesen 1-11. 1937. Exudate. Acta reed. scandinav., 63:458-501. 1928. 7. GUZMAN, GONZALES I. Una neuva constante leucoci- 23. SAXEN, ARNO. Pathologisch-anatomische und klinische taria. La imagen nucleolar linfocitica. ReD. reed. Studien ueber die premaeren yon der Nasenkavitaet de biol., 5:259-264. 1925. den angrenzenden Nebenhoehlen ausgehenden 8. GUZ~AN, GONZALES I. El contenido nucleolar de los Papillome und Karzinome. Arbeit aus dem path. monocitos. Arch. latino am. de cardioI, y hemat., Inst. Univ. Helsingfors, 4:1-130. 1926. 4:109-120. 1934. 24. SIKPSON, M. I., and DEMING, J. M. The Identification 9. HELWIG, E. B. Tumor Cells in Body Fluids. J. Mis- of Myeloblasts with Vital Stains. Folia haemat. souri M. A. 39:73-76. 1942. 34:103-124. 1927. 10. HOL~ES, W. C., and FRENCH, R. W. The Oxidation 25. yON HAAS, E., and ALEXANDER, H. G. New Method Products of Methylene Blue. Stain Tech., 1:17-26. of Staining the Nucleus and Nucleolus in Cell 1926. Suspensions and Frozen Sections. J. Tech. Meth., 11. KAy,, H. Die Cytodiagnostik maligner Tumoren aus 15:70-73. 1936. Punktaten und Sekreten. Ztsch. f. Krebsforsch, 26. VON HAAM, E., and ALEXANDER, H. G. Cytological 36:579-605. 1932. Studies of Malignant Tumors. Am. J. Clin. Path., 2. LUD~ORD, R. J. The Morphology and Physiology of 6:394-414. 1936. the Nucleolus. Part. I. The Nucleolus in the Germ- 27. ZADEK, J. Die Zytodiagnostischen Kennzeichen der Cell Cycle of the Mollusc Limned stagnalis J. Roy Krebszenen. Acta reed. scandinav., 80:78-92. 1933. Micr. Soc., pp. 113-150. (June) 1922. 28. ZE~ANSKY, A. P. The Examination of Fluids for Tu- 13. MACCARTY, W.C. The Value of the Macronucleolus mor Cells. An Analysis of 113 Cases Checked in the Cancer Problem. Am. J. Cancer, 26:529-532. Against Subsequent Examination of Tissue. Am. 1936. J. M. Sc., 175:489-504. 1928.

FIo. 1.---Cell of malignant melanoma. The dark irregu- FIo. 3.--Cells of adenocarcinoma of rectum. larly shaped central masses represent nucleoli. FIe. 4.--Cell of transitional cell carcinoma of bladder. Fzo. 2.--Cell of malignant melanoma. All nucleoli are not in focus. All photographs at a magni- tude of 900 diameters.

FIGs. 1-4

Downloaded from cancerres.aacrjournals.org on September 30, 2021. © 1948 American Association for Cancer Research. DESCRIPTION OF FIGURES 5 TO 10 FI~. 5.--Normal cell from bronchiolar epithelium. Note the small round nucleoli. Fie. 6.--Malignant cell from metastatic carcinoma to bone. Aspiration biopsy of lesion in scapula. Note the bizarre shape of the nucleoli and the fine nucleolar strands. FIG. 7.--Normal mature small lymphocyte. Note the single small round eccentrically placed nucleolus. One cell is not in focus. FIG. 8.--Megaloblast from a patient with pernicious anemia. Note the nucleolar strand, and the bizarre shape of the nucleolus present 'in a benign cell. FIG. 9.--Mesotheliai cells from visceral pleura of lung. FIG. 10.--Malignant cell of adenocarcinoma of rectum, metastatic to peritoneum. In ascitic fluid. All photographs at a magnification of 900 diameters.

358

il ...... #"t "~

i 9 o

7 8

Flos. 5-10

W. W. Ayres

Cancer Res 1948;8:352-359.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/8/8/352.citation

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/8/8/352.citation. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.