Cycling ⇄ Noncycling Cell Transitions in Tissue Aging, Immunological Surveillance, Transformation, and Tumor Growth

INTERNATIONAL REVIEW OF CYTOu)(iY. VOL. 70

*

Cycling Noncycling Cell Transitions in Tissue Aging, Immunological Surveillance, Transformation, and Tumor Growth SEYMOUR GELFANT Departments of Dermatology and Cell and Molecular Biology, Medical College of Georgia, Augusta, Georgia 1. Introduction . . . . . . . . . . . . 11. Background: Cycling and Noncycling Cells

. . . . . . . . . . . . . . . .

A. Explanation of Cycling and Noncycling Cells . . . . . . B. Tissues and Tumors as Proliferative Ecosystems . . . . . Ill. Procedures for Demonstrating the Existence of Noncycling Go-,GI-, and G2-Blocked Cells in the Same Tissue . . . . . . . . . A. Monitor Cells Entering M and S at Hourly Intervals after Stimulating Quiescent Tissues . . . . . . . . . . . . B. Combined Cytophotometric-Autoradiographic and Unlabeled Mitoses Procedures . . . . . . . . . . . . . . . . IV. Establishment of Normal Tissue Proliferative Ecosystems . . . A. Synopsis Panel I . . . . . . . . . . . . . . . . . B. Commentary Panel I . . . . . . . . . . . . . . . V . Aging and Immunosenescence . . . . . . . . . . . . . A. Synopsis Panel I1 . . . . . . . . . . . . . . . . . B. Commentary Panel 11 . . . . . . . . . . . . . . . VI. Establishment of Primary Tumors . . . . . . . . . . . . A. Synopsis Panels 111 to VI . . . . . . . . . . . . . . B. Commentary Panels 111 to VI . . . . . . . . . . . . VII. Unabated Primary Tumor Growth . . . . . . . . . . . . A. Synopsis Panels VII and VIll . . . . . . . . . . . . B. Commentary Panels VII and VIII . . . . . . . . . . . VIII. Establishment of Secondary Tumors . . . . . . . . . . . A. Synopsis Panel IX . . . . . . . . . . . . . . . . B. Commentary Panel IX . . . . . . . . . . . . . . . 1X. Disseminated Malignant Tumor Growth . . . . . . . . . . A. Synopsis Panel X . . . . . . . . . . . . . . . . B. Commentary Panel X . . . . . . . . . . . . . . . X. S u m m a r y . . . . . . . . . . . . . . . . . . . . . XI. Significance . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . .

I 2 2

4

5 5

7 7 7 10 1I II 12 13 13 15

16 16

17 18 18 19

20 20 20 20 22 22

I. Introduction In a previous report (Gelfant, 1977), we presented a model for cell and tissue proliferation based upon the idea that cycling cells can arrest at three different 1

Copyright @ 1981 by Academic Ress. Inc. All rights of repduction in any form reserved. ISBN 0-12.3644704

2

SEYMOUR GELFANT

points in the cell cycle: in early G, (blocked by a Go barrier); in late G, (by a G, block); and in late G2 (by a G, block). The model describes four major categories of cells: cycling cells, noncycling G,,-blockedcells, noncycling G ,-blocked cells, and noncycling G,-blocked cells. These represent the potential proliferating pool in cells in culture and in tissues and tumors in vivo. The particular proliferative needs of a tissue or tumor are brought about by specific noncyclingecycling cell transitions. The present article extends the details and the significance of this model and uses it to explain and to interrelate the problems of tissue aging, immunological surveillance, transformation, and tumor growth. 11. Background: Cycling and Noncycling Cells

A. EXPLANATION OF CYCLING A N D NONCYCLING CELLS

The scheme presented in Fig. 1 is based upon the idea of three inherent arrest points in the cell cycle: a GI-block located at the G,/S border; a G,-block located I.CYCLING CELLS (lO-30%). or (70-W%) .

Go BLOCKED (80-95%)

GI BLOCKED (2-10%1

G2

BLOCKED (2-10%1

FIG. I . Tissue and tumor proliferative ecosystem (modified after Gelfant, 1977).

CYCLING G? NONCYCLING CELL TRANSITIONS

3

at the GJM transition point; and a Go barrier (dashed line) arrest mechanism located early in the GI period of the cell cycle. In relation to these inherent cell cycle arrest points, there are four categories of cycling and noncycling cells; cells of the same type existing within a tissue or a tumor as a complex heterogeneous proliferative ecosystem. The main import of this concept is that it redefines the histological definition of a tissue, and it emphasizes the inherent cellular heterogeneity of individual tissues and individual tumors (i.e., that tissue and tumor cells of the same type are subdivided into separate and distinct categories in terms of their cell cycle proliferative states). 1. Four Major Categories

In the first category, cycling cells are actively proliferating; they are asychronously moving through the cell cycle, G,+S+G,+M; where GI and G2 are preand post-DNA synthesis gap periods, S is the period of nuclear DNA synthesis in interphase; and M is the period of motosis-which produces two daughter cells. The GI and G, blocks in these cells are depicted as being partially openimplying various physiological states of expansion or retraction. (Also, there is no Go barrier arrest mechanism in cycling cells.) Squares with unshaded circles represent cells in the GI period of interphase with nuclear DNA contents of 2C; the shaded circles indicate synthesis of DNA. During the S period, cells have intermediate nuclear DNA contents between 2C and 4C; cells in G2 have 4C DNA contents. Cells in mitosis are depicted in anaphase configurations. Cycling cells can be identified in S or in M by cytochemical-autoradiographicmicroscopic techniques. See column of cells on right representing cycling cells as they would appear within a tissue, for example, within a single layer of basal epidermis. The potential proliferating pool in a tissue or a tumor is composed of three categories of noncycling cells arrested at different points in relation to the G I and the G2 cell cycle blocks and the Go barrier. Noncycling G,-blocked cells are arrested early in GI by a Go barrier. These cells have 2C nuclear DNA contents, and they are located at a distance in time from the S period. For conceptual uniformity, the GI block is depicted as being closed for Go-blockedcells (also, opening of the Go barrier implies concomitant opening of the GI block). The second noncycling category, GI-blocked cells, arrest late in the GI period and are located at the G,/S border (nuclear DNA contents, 2C). The third noncycling category, G,-blocked cells, arrest late in the G, period and are located at the G2/M border (nuclear DNA contents, 4C). In general, noncycling G,-blocked cells have been demonstrated by a variety of cell kinetic growth fraction techniques in a wide variety of tissues and tumors (Gelfant, 1977). Since noncycling cells are not moving through S or through M, since they cannot be distinguished from cycling cells in the GI or in the G2

4

SEYMOUR GELFANT

periods on the basis of nuclear DNA contents, and since noncycling Go- and G,-blocked cells also cannot be distinguished from one another on the basis of nuclear DNA contents (see columns of cells on the right, depicted as squares with shaded and unshaded circles in the diagrams in Fig. I ) , special cell kinetic growth fraction techniques and specific procedures and experimental designs must be used to identify and distinguish all four categories of cycling and noncycling cells as they exist within the same tissue or tumor (see Section 111). Figure 1 provides an estimate of the relative proportions and fluctuations of the four categories of cycling and noncycling cells. Under normal circumstances only about 10-30% of the cells in a tissue or a tumor are in the cycling state. Most of a tissue or tumor proliferative pool resides in the noncycling state (70-90%). And of the three categories of noncycling cells, most (80-95%)reside in the G,,-blocked state; tissues and tumors also contain small proportions of noncycling GI- and G,-blocked cells (2- 10%). 2 . Subpopulations The concept in Fig. 1 also implies that there are additional subpopulations of cells within the major categories-which are qualitatively different from each other in the sense that they may be selectively and independently activated; or in the sense that they may be in different temporal states. Some examples of specific and selective activation of subpopulations of noncycling G,-blocked cells come from studies of mouse ear epidermis in virro in which there are separate sugar, sodium, and potassium ion-responding subpopulations (Gelfant, 1966); also, noncycling G,-blocked Ehrlich ascites tumor cells can be specifically activated to enter mitosis by antilymphocytic serum (DeCosse and Gelfant, 1968). And both noncycling G,- and G,-blocked mouse liver cells can be specifically activated to enter S or to enter M by injection of lead acetate in vivo (Choie and Richter, 1978). Figure 1 also depicts subpopulations of Go-blocked cells which are in different temporal states of arrest-and when released by different stimuli, they enter S after variable Godelay periods; and also shown are subpopulations of cycling cells representing cells moving through the cell cycle at much slower or faster speeds. All of these subpopulations are depicted as triangles, circles, and hexagonal cells in each of the major categories in Fig. I . With regard to tumors, we speculate that the system of subpopulations of noncycling GI- and G,-blocked tumor cells may have specific metastatic capabilities. A N D TUMORS AS PROLIFERATIVE ECOSYSTEMS B. TISSUES

An ecosystem is defined as a system formed by the interaction of a community of organisms with their environment-which confers adaptive value to the system. By analogy and as speculation, Fig. 1 introduces the concept of tissues and tumors as proliferative ecosystems. It is proposed that tissues and tumors main-

CYCLING a 2 NONCYCLING CELL TRANSITIONS

5

tain an adaptive system of cell proliferation-with the use of the four major categories of cycling cells, noncycling G,,-,G I - , and G,-blocked cells and their subpopulations to service the actual and the potential proliferative needs of the tissue or tumor. Because of their arrest points in the cell cycle (at the GJS and at the G,/M borders), noncycling G I - and G,-blocked cells provide tissues with a fast-acting renewal capacity, for when released by appropriate stimuli, these cells enter the cycling S and M periods without delay-in comparison to the slower acting delayed reentry of released Go-blocked cells. The fact that noncycling cells can arrest at different temporal and biochemical points in the GI and in the G2 gap periods of the cell cycle [recent evidence indicates that neoplastic cells can also arrest in the S period (Darzynkiewicz et al., 1979)], and the fact that most cells reside in the noncycling state offers the tactical advantage of quiescence (at different points in interphase) over the turmoil involved in the continuous synthesis of the genetic and the mitotic machinery necessary for chromosome replication, chromosome movement, and cytoplasmic cleavage (i.e., the cycling state). In terms of tumor survival and resistance to therapy, the quiescent state provides an additional advantage because most chemotherapy acts only on cells in the cycling state (specifically on cells in S and in M). Physiological subpopulations within the major categories of noncycling cells would provide an additional adaptive dimension to the proliferative ecosystem of the tissue or the tumor. Such cells capable of being released to the cycling state only by very specific or unusual stimuli serve as another restrictive system to secure proliferative quiescence. For further support of our concept of tissues and tumors as proliferative ecosystems and for the adaptive significance of noncycling cells as described above, see publications entitled, “Mechanisms Underlying the Differential Sensitivity of Proliferating and Resting Cells to External Factors” (Epifanova, 1977), “The Survival Value of the Dormant State in Neoplastic and Normal Cell Populations” (Clarkson, 1974), and “The Biological Essence of Resting Cells in Cell Populations” (Lerman, 1978).

111. Procedures for Demonstrating the Existence of Noncycling G,,-, G ,-, and G,-Blocked Cells in the Same Tissue A. MONITOR CELLSENTERING M AND S

AT HOURLY INTERVALSAFTER STIMULATING QUIESCENT TISSUES

As depicted in Fig. 2, if one stimulates a quiescent or experimentally suppressed tissue and monitors cells entering M and S in autoradiographs at hourly intervals, one observes a prompt and transient increase in the number of mitoses. representing release of G,-blocked cells into M; there is also a prompt and transient increase in the number of labeled nuclei within the first few hours,

6

SEYMOUR GELFANT

1

" z

STIMULATE

d

QUIESCENT. NONCYCLINO I s i c r ~ m s n l o l l yauppiaissdl CELLS a T I S S U E S l d a y ~ .months. 01 ysorsl I

0

I

2

3

4

10-20

HRS.

FIG.2. Procedure for demonstrating the existence of noncycling Go-, G , - , and G,-blocked cells in the same tissue (after Gelfant. 1977).

representing release of G,-blocked cells into S. Then after a delay of about 10 to 20 hours, one observes a second, much larger increase in the number of labeled nuclei, representing Go-blocked cells entering S after a delay; this is followed by a comparable and subsequent increase in the number of mitoses as depicted in Fig. 2. It should be emphasized that there is very little or no DNA labeling or mitotic activity in quiescent noncycling cells and tissues. Also, cells may remain in the noncycling state for months or years. Examples of quiescent, noncycling (experimentally suppressed) cells and tissues are adult liver, kidney, salivary glands, hormone-depleted or nutritionally starved tissues, in vivo; density or media depleted stationary cell cultures, in vitro. Quiescent tissues and cell cultures can be stimulated by regenerative stimulation such as partial hepatectomy, partial nephrectomy; wounding; hormone resupply; refeeding, in vivo; or by replating or media change of cell cultures, in vitro . In a previous report (Gelfant, 1977), we presented three tables of examples of noncycling Go-, GI-, and G,-blocked cells. Noncycling G,,-blocked cells have been demonstrated in all tissues and tumors and cell culture systems both in vivo and in v i m . Noncycling G,-blocked cells have been found in a wide variety of animal, plant, and tumor tissues both in vivo and in v i m . The number of examples of noncycling G,-blocked cells is small because most workers do not ordinarily monitor DNA synthesis immediately after stimulation; also, the increase in the number of cells entering S from the G,-blocked state is much less

CYCLING # NONCYCLING CELL TRANSITIONS

7

and relatively transient when compared to the subsequent delayed increase in G,-blXked cells entering DNA synthesis (as shown in Fig. 2). Nevertheless, there are reports demonstrating the existence of noncycling G,-blocked cells in tissues such as epidermis, tongue, kidney epithelium, liver, mammary gland, capillary endothelial cells, hemopoietic cells, ascites tumor cells, in vivo; and hemopoietic cells, in vitro.

B. COMBINED CYTOPHOTOMETRIC-AUTORADIOGRAPHIC A N D UNLABELED MITOSES PROCEDURES

The following is an outline of another general procedure for demonstrating and distinguishing all four categories of cycling and noncycling cells within the same tissue in vivo or in v i m (after Gelfant, 1966). 1. To demonstrate both cycling and noncycling cells: Expose cells to [3H]thymidinecontinuously for three to five times longer than the particular cell generation time. a. Labeled nuclei = evidence for cycling cells. b. Unlabeled nuclei = evidence for noncycling cells. 2. To distinguish noncycling cells blocked in G, or in G2: Measure DNA contents of unlabeled nuclei (directly through autoradiographic emulsion with Feulgen cytophotometry). a. Unlabeled 4C nuclei = evidence for G,-blocked cells. b. Unlabeled 2C nuclei = evidence for GI-blocked cells and/or evidence for (&-blocked cells. 3 . To determine further whether unlabeled noncycling cells are G , , Go.or G2 blocked: Stimulate other samples; keep in presence of [3H]thymidine. a. Experimentally release G,-blocked cells: G,-blocked cells promptly enter M and appear as unlabeled mitoses. b. Experimentally release G,- and G,-blocked cells: Unlabeled G ,-blocked cells promptly enter S and appear as labeled nuclei. Unlabeled Go-blocked cells enter S after a delay; and they also appear as labeled nuclei. (All interphase nuclei and all mitoses are now labeled.)

Figure 3 uses this procedure for demonstrating the existence of noncycling G,-blocked cells in Ehrlich ascites tumor and in mouse ear epidermis in vivo.

IV. Establishment of Normal Tissue Proliferative Ecosystems A. SYNOPSIS PANELI (FIG. 4) Panel I (Fig. 4) depicts the origin and the cell cycle point of arrest (in relation to the G, and G, cell cycle blocks and the Gobarrier) of the three major categories

8

SEYMOUR GELFANT

CYCLING

CYCLING 100% origtn

PANEL

NONCYCLING STATES

r

__

9

NONCYCLING CELL TRANSITIONS

70-90% or 10-3OY. three coteqories GI block

t~ssue-C ~ I I U I O ~"oqinq" transitions to noncyclinq stotms embryonic. mmoture. mature

CYCLING STATE 10-30%

01

70.90%

one c a f e g o r y

Balanced Tissue Growth open

=

homeostatic return

QL-y open

¶pecific homeostatic relea;

* homeostottc

return

G

bloik

Flc. 4.

Establishment of a normal tissue proliferative ecosystem.

of noncycling GI-, G,,-,and G,-blocked cells as each category converts from the cycling state to the noncycling state (termed cellular "aging" transitions) in different tissues during different periods of chronological development of the entire organism. Balanced tissue growth is a result of noncyclingGcycling state transitions involving specific homeostatic release and return of each of the three FIG.3. Combined cytophotometric-autoradiographic and unlabeled mitoses procedures for demonstrating the existence of noncycling G,-blocked cells. G,-blocked cells appear as unlabeled interphase nuclei with 4C DNA contents and as unlabeled mitoses in autoradiographs-having been exposed to [3H]thymidinefor long periods of time prior to stimulation. (A) Ehrlich ascites tumor cells (mouse peritoneal cavity) exposed to continuous administration of I3Hjthymidinefor 96 hours (five times longer than EAT cell cycle-generation time). Combined DNA Feulgen stain cytophotometryautoradiography techniques. Unlabeled nucleus (arrow) contains 4C DNA content-thus, demonstrating the existence of noncycling G,-blocked tumor cells. (B) Unlabeled mitosis Ehrlich ascites tumor-representing release of unlabeled noncycling G,-blocked cell shown in (A); released into mitosis by antilymphocytic serum in the presence of and after 48 hours of continuous administration of [3Hjthymidine.Similar results were obtained by injecting other immunosuppressants, hydrocortisone and azathioprine (DeCosse and Gelfant, 1968). (C) Unlabeled mitosis (arrow) mouse ear epidermis in vivo. Demonstrates existence of noncycling G,-blocked epidermal cell, released into mitosis by wounding, in the presence of and after prior continuous administration of [3Hjthymidine for 5 days. Similar results were obtained after 6 months of prior continuous administration of [3H]thymidine(Pederson and Gelfant, 1970).

10

SEYMOUR CELFANT

major categories of noncycling cells. Released noncycling GI- and G,-blocked cells enter S or M promptly-because they had arrested or had been blocked at the Gl/S or at the G,/M transition points (and thus, serve tissues as fast-acting renewal systems). Released Go-blocked cells enter S after a delay in timebecause they arrest in early GI-having been held in the noncycling state by the G,, barrier. Because most noncycling cells come to rest in the Go-blocked state, the overall growth characteristics of a tissue are primarily due to noncycling G,*cycling cell transitions. For a review of the concept of a tissue as a proliferative ecosystem, see Section 11, B.

I B. COMMENTARY PANEL Tissue cellular “aging” transitions to noncycling states: In a previous report (Gelfant and Smith, 1972), we defined tissue cellular aging as, “Aging on a cellular level is described as a progressive conversion of cycling to noncycling cells in tissues capable of proliferation. Embryonic aging transitions: Some tissues complete their cellular aging transitions to the noncycling G,-, Go-, and G,-blocked states during embryogenesis, for example, pancreas, lens, tongue muscle. Immature aging transitions: Other tissues complete their cellular aging transitions to the noncycling states during adolescence, i.e., before completion of maximum growth of the entire organism, for example, liver, kidney, bone. Mature aging transitions to the noncycling states: These take place during animal senescence in tissues such as epidermis and epithelium of the gastrointestinal tract. The following quotation from Pardee (1974) also supports our depiction of cellular aging transitions to the noncycling states: “Most animal cells in vivo exist in a nonproliferating state in which they remain viable and metabolically active. They arose from proliferating cells whose metabolic patterns were switched to quiescence at some time during differentiation. ” Balanced tissue growth: When overall tissue cell birth exceeds cell loss, cycling cells move into the noncycling state. When cell loss due to trauma or to disease exceeds cell birth, noncycling cells move into and remain in the cycling state until repair, size, and balanced tissue growth is achieved; for example, renewal and repair of liver, kidney, epidermis, and other tissues in vivo (Cameron, 197 l). And in restoration of hematopoietic equilibrium after hemorrhage: “Normal hematopoiesis is tightly regulated so that production of new cells exactly balances cell loss due to senescence and other causes. The rate of production can be increased in response to increased cell loss (e.g., hemorrhage), but once the imbalance is corrected, hematopoietic equilibrium is restored at the original level” (Clarkson and Rubinow, 1977). In general, the growth fraction of an unperturbed tissue, i.e., the ratio of cycling to noncycling cells for each tissue depends upon its function and its particular proliferative state. In Panel I, homeostatic release and return from noncycling states are con”

11

CYCLING # NONCYCLING CELL TRANSITIONS

trolled by factors such as hormones (Epifanova, 1971), chalones (Houck and Daugherty, 1974), other endogenous regulatory molecules (Sheppard and Bannai, 1974; Haddox e? al., 1977; Lozzio et al., 1975), pH and metabolic changes within the cell (Gerson, 1978; Rubin and Fodge, 1974), intracellular equilibrium constants (Gilbert, 1977), changes in cell membrane receptor sites (Carney and Cunningham, 1978), other cell surface changes (Allred and Porter, 1977; Pardee et al., 1974), changes in cation levels-and electrical transmembrane potentials (Cone, 1970), osmotic changes (Cone et al., 1968), intracellular water (Beall et al., 1976), intracellular concentration of nutrients (Bhargava, 1977), intra- and extracellular temperature (Zada-Hames and Ashworth, 1978), intracellular waste product accumulation (Hirsch, 1978), supply of oxygen and nutrition (Cameron, 1973, regenerative and repair responses to wounding (Cameron, 1975), tension (Curtis and Seehar, 1978), tissue architectural disturbances (Bertsch et al., 1976), and cell-to-cell contact (Cameron, 1975). Homeostatic release and return from noncycling states are also subject to biological oscillations and to rhythmic-circadian feedback mechanisms (Ehret et al., 1977; Njus et af., 1974; Clevecz, 1978). The three categories of noncycling G I , G o - , and G,-blocked cells (and their subpopulations-which are not depicted in the diagrams in Panel I) can be selectively and independently released in different tissues by specific and by different homeostatic and experimental factors (Gelfant, 1377).

V. Aging and Immunosenescence A. SYNOPSIS PANELI1 (FIG. 5)

To simplify illustrations, all three categories of noncycling G,-, Go-, and G,-blocked cells which were shown in separate drawings in Panel I, have been combined into a single diagram in Panel I1 (Fig. 5). Tissue proliferative aging (in general) and immunosenescence (age-related decline in immune function) are IMPAlUED RELIEASIE O F NONCYCLlNG CIELLS PANEL

Cell

l7

Loss

_-)(_ _ _ _ _ _ _ - - _ _

--__b

Cell B i r t h

U

FIG.5 . Aging and immunosenescencein chronologically aged tissues, including immunocompe-

tent cells.

12

SEYMOUR GELFANT

depicted as being due to impaired or decreased ability of noncycling cells to be released to the proliferative-cycling state. B. COMMENTARY PANELI1 Aging: impaired release of noncycling cells in chronologically aged tissues in (Gelfant and Grove, 1974) or after increasing time spent in the noncycling state in vitro. A good example in vivo is in old rats stimulated by isoproterenol-where noncycling G,,-blocked parotid gland acinar cells show a diminished response and exhibit an increased delay in moving into the cycling state. To quote from Adelman et al. (1972): “Upon administration of identical body weight dosages of isoproterenol to older rats, it is evident that: 1. the time required to initiate DNA synthesis increases progressively and is directly proportional to the chronological age of rats from 2 to at least 24 months; 2. the delayed onset of DNA synthesis may be accompanied by a decreased magnitude of response; and 3. the ability to stimulate cell division may be abolished.” and to quote from Potter (1978): “Essentially 100% of the hepatocytes in a young adult rat. . . are capable of returning from G,, to G , , S, G, and mitosis. In an old rat . . . this is not the case, as it has been estimated.. . that as many as 69% of the hepatocytes cannot return to G , . There is also impaired release of noncycling Go-blockedrat liver hepatocytes, and kidney epithelial cells in response to partial hepatectomy or to partial nephrectomy in old animals (Bucher et al., 1964; Phillips and Leong, 1967). I n vitro, the longer cells remain in the nonproliferating state, the deeper they go into Go and the longer the delay before entering S after stimulation; for example, human fibroblasts kept stationary in the noncycling state for 5, 9, or 18 days show corresponding increasing delays of 8, 14, and 20 hours before entering S after stimulation by medium change (Augenlicht and Baserga, 1974); Chinese hamster lung cells, media depleted for 24 or 48 hours show corresponding delays of 12 or 20 hours after addition of complete medium (Martin and Stein, 1976); and root meristem cells, carbohydrate starved for 48, 72, or 96 hours show increased Go delays of 8, 10, and 14 hours after stimulation by sucrose (Van’t Hof et al., 1973). Immunosenescence: impaired release of noncycling immunocompetent cells with chronological age (Gelfant and Grove, 1974). Immunosenescence involves an age-related decline in both humoral and cell-mediated immune function (Yunis and Lane, 1979) which is primarily due to involution of the thymus and a decrease in the number of functional (responsive) T cells-which are required for both humoral and cell-mediated immune responses (Kay, 1979). There is also a decline in the proliferative capacities of both B and T cell immunocompetent lymphocytes (Kay, 1978). Unstimulated immunocompetent B and T cells reside primarily in the noncycling Go-blocked state (Buell et al., 1971). There is an age-related decline in B cell proliferation in response to antigen stimulation

vivo

”

CYCLING

NONCYCLING CELL TRANSITIONS

13

(Perkins and Makinodan, 1971) and when assessed by spleen colony growth capacity in vivo (Albright and Makinodan, 1976). There is also a decline in B cell proliferative capacity in vitro in response to Pokeweed mitogen (Mathies et al., 1973) and to other B cell mitogens (Gerbase-DeLima et al., 1974). Similarly, there is an age-related decline in T cell proliferative capacity in vitro in response to plant mitogens phytohemagglutinin (Hori er al., 1973) and concanavalin A (Kruisbeek, 1976), and in the mixed lymphocyte culture reaction (Konen et al., 1973). There is also a decline in proliferative capacity of T cells in vivo as demonstrated by the ability of donor bone marrow T cells to regenerate and repopulate the thymus of an irradiated host (Tyan, 1976) and by assays of spleen cell proliferation and lymphoproliferative activity in draining lymph nodes (Perkins and Cacheiro, 1977). The following quotation from Perkins and Cacheiro (1977) supports the view of immunosenescence as depicted in Panel I1 and as described above: “Thus, decreased PHA responsiveness of splenic lymphocytes provided as sensitive an estimate of the age-related decline of immunocompetence in old mice as other classical parameters of cell-mediated immunity (e.g., graft-versus-host reaction or in vivo cellular proliferation of parental spleen cells in lethally irradiated F, recipients). Results could be interpreted to represent a decreased ability of noncycling T-cells to be released to a functional cycling state” (italics mine). So that, in addition to the decreased availability of functional T cells (because of thymic involution) immunosenescence is also due to impaired release of noncycling B and T cells, as depicted in Panel 11.

VI. Establishment of Primary Tumors A. SYNOPSIS PANELS I11 TO VI (FIG.6) A primary tumor may arise from the release and proliferation of preexisting previously transformed dormant tumor cells, or it may arise from normal tissue cells which are directly transformed into tumor cells-and establish a primary tumor. The first method is depicted in Panels 111 and IV; the second method of establishing a primary tumor is depicted in Panels V and VI. Once again, all three categories of noncycling G1-, G o - , and G,-blocked cells are combined into single diagrams. Panel 111 implies that normal tissues contain small numbers of transformed tumor cells which are being restrained in the noncycling G , - and G,-blocked states by an immune inhibition mechanism. The first step in establishing a tumor as depicted in Panel I11 shows release of previously transformed noncycling G,and G,-blocked tumor cells to the cycling state, due to a decline in immunological restraint. Some cycling tumor cells, thus released, return to the Go-blocked state, some also return to the noncycling G, - and G,-blocked states-stablish-

14

SEYMOUR GELFANT

PANEL

m relear of p r m u r l y transformed GI and G, blacked tumor cellr

m Some cycling tumor calla return to the 0, blocked state establishing a prlmary tumor compored of the four mojor cotegories of cycling and noncycllng cells.

PANEL 1p

Primary Tumor

\

tumor homeartatic release homeostatic return H

"Cell Birth"

"COII LOIS"

direct transformation of normal cychng chromosomer reprogmmmed during S or

M

Environmentally transformed "initlated" dau er cells may develop mt0 a tumor directly or they may require additionapcarcinogenic "promotion" Cycling transformed daughter ceIIr establish o primary tumor in a manner rimdar to normal tiraues as shown in Panel I.

PANEL I U

-

tumor-cellular "aging" transitions to noncycling states lransforwmu

Primary Tumor

/

tumor homeostatic releaae

. 1 1 1 1 )

homeoitatiC return

rn "C~II Loss"

"Cell Birth"

FIG. 6 . Establishment of a primary tumor.

ing a primary tumor composed of the four major categories of cycling and noncycling cells as depicted in Panel IV. Panel V implies that primary tumors are environmentally induced by direct transformation of normal cells by environmental factors (chemicals, viruses,

CYCLING

NONCYCLING CELL TRANSITIONS

15

irradiation, etc.). Panel V depicts environmental transformation of normal cycling cells by direct transformation of DNA during DNA synthesis or by reprogramming of DNA by other chromosomal molecules during mitosis-resulting in the production of heritable transformed daughter cells. Cycling transformed daughter cells then establish a primary tumor proliferative ecosystem in a manner similar to normal tissues. This is depicted in Panel VI. Primary tumors (arising by either method) are similar to normal tissues in a number of ways. (1) They exist as independent proliferative ecosystems-their growth patterns being independent of the normal tissues in which they reside. (2) The relative percentages of noncycling Go-blocked (80-95%), GI-blocked (2lo%), and G,-blocked (2-10%) tumor cells in a primary tumor are similar to normal tissues. (3) Most primary tumor cells (70-90%) are in the noncycling state. (4)And in general, most primary tumors maintain a balanced tumor growth by a balance between tumor cell loss and tumor cell birth.

B. COMMENTARY PANELS 111 TO VI Decline in immunological restraint: (an unconventional aspect of tumor immunological surveillance). Based upon studies with Erhlich ascites tumor cells, we introduced the idea that noncycling tumor cells may be held in restraint by an immune inhibition mechanism (DeCosse and Gelfant, 1968). We showed that three different immunosuppressants, antilymphocytic serum, hydrocortisone, and azathioprine, specifically activated and released noncycling G,-blocked tumor cells into mitosis. There are clinical observations showing that protracted immunosuppressive therapy of organ transplant patients is accompanied by an increased incidence of malignant tumors (Penn and Starzl, 1972). Also, there are reports (Kurland et al., 1978) showing that macrophages can specifically block proliferating hemopoietic tumor cells both in the G, and in the G, periods of the cell cycle-all of which support the contention that noncycling tumor cells may be held in restraint by an immune inhibition mechanism. Therefore, we speculate that a decline in immunological restraint of noncycling-already existing tumor cells (releasing them to the cycling state) would be one method allowing for the establishment of a primary tumor as depicted in Panels 111 and IV. Panel 111, release of previously transformed G,- and G,-blocked tumor cells: The idea that noncycling G , - and G,-blocked tumor cells play a role in establishing a primary tumor comes from a previous report (Gelfant, 1977) in which we suggested that the cancerous property of undetected tumor dormancy resides in a small minority system of tumor cells compatible with the descriptions of noncycling GI- and G,-blocked cells and their subpopulations (depicted and described in Fig. 1). Also, the idea that primary tumors develop from small numbers of previously transformed preexisting tumor cells (or even from a single cell) can be supported by quotations from other reports: “all cancer cells come from prior

16

SEYMOUR GELFANT

cancer cells except the first cancer cell” (Busch, 1978), and “It is proposed that most neoplasms arise from a single cell of origin” from “The Clonal Evolution of Tumor Cell Populations” (Nowell, 1976, 1978). Panel IV-in addition to normal tissue homeostatic release and return factors, solid tumor growth is also controlled by excessive cell loss (Steel, 1967; Refsum and Berdal, 1967), necrosis (Weiss, 1977a), regional differences in vascular growth and control of tumor vascularization (Folkman, 1976; Cavallo ef al., 1972), and in the supply of oxygen and nutrition (Tannock, 1968, 1978), and in unusual architecture and the existence of bizarre cells (Steel, 1973), and by immune elimination of tumor cells (Cooper et al., 1975; Burnet, 1970). Similar to normal tissues, tumor growth is primarily controlled by noncycling G@cycling tumor cell transitions. Panel V, environmentally induced-direct transformation or reprogramming of DNA in a heritable manner can occur in a number of ways. For example, environmental carcinogenic factors, chemicals, viruses, irradiation (Higginson, 1969; Heidelberger, 1975; Ames, 1979), may transform DNA directly (Barrett et al., 1978; Milo and DiPaolo, 1978), by binding to DNA (Neidle, 1980), by electrophilic attack (Barrett, 1979), by energy transfer (Barrett, 1979), by insertion of viruses (Green, 1978), evolution of cancer genes-protovirus hypothesis (Temin, 1971, 1972, 1974), activation of oncogenes (Todaro and Heubner, 1972), activation of transforming genes (Comings, 1973), or these environmental factors may cause the synthesis of unusual gene regulatory molecules during the cell cycle of a normal parent cell (Stein et al., 1978). Chromosomal regulatory molecules, nuclear proteins, and small nuclear RNAs interact with and bind to DNA during replication in the S period or during release and reacquisition of chromosomal molecules-which occurs in the process of chromosome condensation and decondensation during mitosis-and in this way can reprogram or “transform” genetic expression of daughter cells (Goldstein and KO, 1978; Pederson and Bhorjee, 1979; Goldstein, 1978). There is evidence that environmentally induced carcinogenic transformation is a two-stage process: initiation, i.e., the initial mutational event, and promotion-which involves subsequent and additional exposure to environmental carcinogens (Chouroulinkov and Lasne, 1978). The second event may occur soon after the first or not until a long time later. Initiation is irreversible. Promotion is reversible and may be experimentally manipulated-offering therapeutic possibilities (Marx, 1978).

VII. Unabated Primary Tumor Growth A . SYNOPSIS PANELS VII

AND

VIII (FIG.7)

A tumor grows because the number of new tumor cells produced per unit time (cell birth) exceeds the number of tumor cells lost per unit time (cell loss). Two

CYCLING S NONCYCLING CELL TRANSITIONS

D€CL/NE IN /!!UN€

EL /!/NATION

17

OF TUMOR CEL L S

Fic. 7. Unabated primary tumor growth

mechanisms are depicted in Panels VII and VIII (Fig. 7) which would result in an increase in cell birth over cell loss and which, therefore, would result in unabated tumor growth. Panel VII depicts the idea that (under certain conditions) cycling tumor cells lose their ability to return to the noncycling state; i.e., impaired return to the noncycling state. In Panel VIII, the increase in cell birth over cell loss is brought about by a decline in immune elimination of tumor cells. B. COMMENTARY PANELS VII

AND

VIII

Panel VII, impaired return to the noncycling state: The idea that transformed cells arise from noncycling cells and under certain conditions lose their ability to return to the resting state can be supported by the following quotations: “cancer cells appear to arise from quiescent cells that have been switched back to active proliferation (Pardee, 1974); ‘‘the critical difference between transformed and untransformed cells is that the untransformed cells go into a resting state when conditions are not optimal for their growth. Transformed cells continue to grow in such conditions. Transformation is thus an abrogation of the resting state” (Baltimore, 1975); and from Bhargava ( 1977). “Malignant transformation is defined as an inheritable intracellular event, spontaneous or induced, which. . . leads to a loss of the capacity for transition from the dividing to the resting state. Immune elimination of tumor cells (the conventional aspect of immunological surveillance): “In essence, immunological surveillance is the concept that a major function of the immunological mechanisms in mammals is to recognize and eliminate foreign patterns arising in body by somatic mutation or some equivalent process . . . this is important primarily as providing a means by which ”

”

18

SEYMOUR GELFANT

the appearance of malignant disease may be effectively cut short. . . [and] it will be evident that the thymus-dependent system of immunocytes will be almost solely responsible for surveillance” (Burnet, 1970). Although this concept is being challenged by a number of investigators (Moller and Moller, 1976) who believe that the main function of T cell lymphocytes is to defend against microbial infections including viral oncogenesis, there is undisputed evidence of some form of immune elimination of tumor cells within the primary host in both viral and in chemically induced tumors (Bansal et al.. 1978; Old, 1977). The process is most probably mediated via the thymus-dependent immune system of T cell lymphocytes, and it also includes activated immunocompetent macrophages. “A considerable body of evidence shows that cells of the mononuclear phagocyte series can recognize and kill transformed and malignant cells, and a defensive role in host surveillance against neoplastic disease has been ascribed to the macrophage” (Rhodes et al., 1979). Thus, as depicted in Panel VIII, a decline in immune elimination of tumor cells would result in unabated tumor growth. The important point regarding unabated tumor growth whether it occurs by either of the mechanisms shown in Panels VII and VIII is that tumors grow because the overall balance of cell birth exceeds cell loss. Baserga (1976) states: “at a cellular level cancer is characterized by an increase in cell number. . . . Any tissue can increase the number of cells in the population by one of three ways: (a) shortening the cell cycle; (b) increasing the growth fraction by recruiting Go cells into the cell cycle; and (c) decreasing the rate of cell loss [However, tumors may utilize all 3 parameters.]. The most prominent mechanism is recruiting Go cells into the proliferating pool [A statement which agrees with our idea that tissue and tumor growth are primarily due to noncycling G,*cycling state transitions. Put succinctly, the reason a tumor grows is that] the number of new cells produced per unit time [cell birth] always exceeds the number of cells lost per unit time [cell loss].” And according to Steel (1973): “It is characteristic of neoplasia in that the disturbed balance between cell production and cell loss is usually permanent; overgrowth, then is progressive and irreversible. ”

VIII. Establishment of Secondary Tumors A. SYNOPSIS PANELIX (FIG. 8) Panel IX (Fig. 8) suggests that noncycling GI- and G,-blocked primary tumor cells (and their subpopulations) have the specialized capacity to metajtasize, invade secondary sites, and establish secondary tumors. After invading secondary sites, GI- and G,-blocked tumor cells are released to the cycling state and establish secondary tumors as described in Panels I11 and IV.

CYCLING

19

NONCYCLING CELL TRANSITIONS

M€TAS;TAS/S AND f?€POWLAT/ON PANEL ZK

:@ .non

0%

GI and G, Mocked primary tumor cells metastasize and invade secondary sites establish secondary tumors (see Panel m)

8% brn& D

FIG.8.

Establishment of secondary tumors (independent proliferative ecosystems).

B. COMMENTARY PANELIX The ideas depicted in Panel IX that noncycling GI- and G,-blocked tumor cells metastasize and invade secondary sites can be supported by the following quotations: “neoplasms are also heterogeneous with regard to invasion and metastasis, i.e., that they contain a variety of subpopulationsof cells with differing metastastic potentials ” (Fidler, 1978); “cancer metastases originate from subpopulations of cells with high metastatic potential that preexist within the parent tumor” (Kripke er al., 1978); and “only certain. . . malignant cells possess characteristics that enable them to travel through the body and establish new tumors” (Nicholson, 1979). In general, “It is suggested that the necrotic regions of tumors and products derived from them, facilitate the detachment of tumor cells. . . thereby potentially promoting metastasis and invasion” (Weiss, 1977a,b). Newly invaded malignant (noncycling G,- and G,-blocked) primary tumor cells may remain dormant, retaining the capacity to be released at a later date [see ‘‘On the Latency of Tumor Cells” (Stein-Werblowsky, 1978)l; or they may immediately begin to proliferate and repopulate. As indicated in Panel IX,some repopulating-cycling tumor cells return to the Go-blockedstate, some also return to the noncycling GIand G,-blocked states-thus establishing secondary tumors composed of the four major categories of cycling and noncycling tumor cells-in a manner similar to the situation depicted in Panel 111. Secondary tumors as independent proliferative ecosystems: The proliferative states and tumor cell composition of secondary tumors, i.e., the relative proportions of cycling and noncycling tumor cells and the particular compositions of noncycling subpopulations may differ from the primary tumor and also from secondary tumors at other metastatic sites (Steel, 1977; Bellamy and Hinsull, 1978), thus, making the situation more difficult from the point of view of generalized chemotherapy (Simpson-Hemen er al., 1974; Slack and Bross, 1975). Also from Nicholson (1979), “The heterogeneity of tumor populations may be responsible for some clinical experimental failures of chemotherapeutic

20

SEYMOUR GELFANT

drugs if resistant subpopulations survive the drug treatment. A drug may be chosen because it inhibits the growth of a primary tumor or of its tissue-culture counterpart and yet may fail to halt the growth of some tumor-cell subpopulations that are present in micrometastases. ”

IX. Disseminated Malignant Tumor Growth A . SYNOPSIS PANELX (FIG.9)

Panel X (Fig. 9) depicts disseminated malignant growth resulting from a combination of metastasis and repopulation (Fig. 8) and unabated tumor growth (Fig. 7). Secondary disseminated tumors originate not only from the primary tumor, but there is also “Metastasis of Metastases” (Hoover and Ketcham, 1975). B. COMMENTARY PANELX There is evidence that disseminated malignant tumor growth in man arises from secondary “generalizing sites” such as lung, liver, or bone, or from release of previous primary tumor metastases which had remained dormant after lodging in these secondary sites (Bross and Blumenson, 1976; Viadana et al., 1978; Hoover and Ketcham, 1975). X. Summary

The ideas in this article are presented in the form of diagrams in 10 separate panels which depict and describe cycling*noncycling cell proliferative transitions as they apply to the problems of aging, immunological surveillance, transformation, and tumor growth. Almost all of the ideas presented in this article are based on and are supported by published statements and data. In

PANEL

X rnetastasi~ and repopulatm

unoboled secondary tumor growth

n

FIG.9.

Disseminated malignant tumor growth.

CYCLING ZS NONCYCLING CELL TRANSITIONS

21

addition to illustrative and diagrammatic depictions, each panel is accompanied by a written synopsis and by a point by point commentary. Panel I (Fig. 4) describes the cycling to noncycling cell transitions which lead to the establishment of normal tissues composed of four major categories of cycling, noncycling G , -,G o - ,and G,-blocked cells-which have arrested at different temporal and biochemical points in the G, and the G, periods of the cell cycle. Balanced tissue growth is a result of noncyclingecycling state transitions involving specific homeostatic release and return of each of the three major categories of noncycling cells. Tissues maintain an adaptive system of cell proliferation with the use of these various categories of cycling and noncycling cells (and physiological subpopulations) to service the actual and the potential proliferative needs of the tissue (and the organism). In this regard, tissues are viewed as proliferative ecosystems. In Panel I1 (Fig. 3,tissue proliferative aging (in general) and immunosenescence (age-related decline in immune function) are depicted as being due to impaired release of noncycling cells to the proliferative cycling state. Subsequent panels deal with tumors. Panels I11 to VI (Fig. 6) take up two ways in which primary tumors may arise. A primary tumor may arise from the immunological release and proliferation of preexisting, previously transformed dormant noncycling tumor cells which had been held in restraint by immunological suppression. Or primary tumors may be induced by direct transformation of DNA and reprogramming of chromosomes of normal cycling cells by environmental factors such as chemicals, viruses, and irradiation. Similar to normal tissues, primary tumors are composed of the four major categories of cycling, noncycling (3,-, Go-, and G,-blocked tumor cells (and subpopulations); and tumors also exist as independent proliferative ecosystems. Panels VII and VIII (Fig. 7) describe unabated tumor growth as being due to an imbalance of the overall ratio of cell birth to cell loss. Two mechanisms are presented. One mechanism depicts the idea that under certain conditions, cycling tumor cells lose their ability to return to the noncycling state. In the second mechanism, the increase in cell birth over cell loss is brought about by a decline in immune elimination of tumor cells. Panel IX (Fig. 8) suggests that noncycling G,-and G,-blocked primary tumor cells and their subpopulations have the specialized capacity to metastasize, invade secondary sites, and establish secondary tumors which then also behave as independent proliferative ecosystems. Panel X (Fig. 9) depicts disseminated malignant tumor growth as resulting from a combination of establishing secondary tumors by metastasis and repopulation and by unabated secondary tumor growth. In addition to the 10 panels, quotations, and references which deal with the main thesis, the article contains (as background material) three figures including

22

SEYMOUR GELFANT

photomicrographs explaining and providing evidence for our overall concept of cycling and noncycling cells.

XI. Significance From a proliferative function point of view, this article redefines the histological definition of a tissue as a structure being composed of a homogeneous group of similar cells all engaged in the same functional activity. Instead, we define a tissue as a heterogeneous system of cells of the same type-cells which do not exist in the same state of proliferative function, i .e., in the same cell cycling or noncycling states. In this context we describe and develop the concept of a tissue as a proliferative ecosystem-a system which confers adaptive survival value to the tissue and which also allows for a variety of tissue proliferative responses: fast acting, slow acting, and specialized tissue renewal responses. This article describes and views both primary and secondary tumors also existing as independent proliferative ecosystems (independent from one another-and from the tissues in which they reside), thus making generalized tumor chemotherapy more difficult. Using our scheme of cyclingenoncycling cell transitions, we specifically explain and specifically interrelate the problems of tissue aging, immunological surveillance, transformation, and tumor growth. In effect, this article provides a new framework to view the proliferative aspects of these problems.

ACKNOWLEDGMENTS

To Juanita Jones, secretary, who worked with me on every aspect of this manuscript as a research associate. I thank James F. Danielli, F.R.S., for his encouragement and for his editorial support. This work was supported by NIH Grant AM 19735.

REFERENCES Adelman, R. C., Stein, G., Roth, G. S., and Englander, D. (1972). Mech. Aging Dev. 1, 49-59. Albright. J. W., and Makinodan, T. (1976). J . Exp. Med. 133, 1204-1213. Allred, L. E. and Porter. K . R. (1977). In "Growth Kinetics and Biochemical Regulation of Normal and Malignant Cells" (B. Drewinko and R. M. Humphrey, eds.), pp. 295-314. Williams & Wilkins, Baltimore. Maryland. Ames. B. N . (1979). Science 204, 587-593. Augenlicht, L. H.,and Baserga. R. (1974). Exp. Cell Res. 89, 255-262. Baltimore D. (1975). Cold Spring Harbor Symp. Quanr. Biol. 39, 1187-1200.

CYCLING + NONCYCLING CELL TRANSITIONS

23

Bansal, S. C., Mark, R., Bansal, B. R., and Rhoads, J. E., Jr. (1978). J. Nutl. Cancer Insr. 60, 667-675. Barren, J. C.. Tsutsui, T.,and Ts’o, P. 0. P. (1978). Nature (London) 274, 229-232. Barren, T. W. (1979). Cancer Biochem. Biophys. 3 , 189-192. Baserga, R. (1976). “Multiplication and Division in Mammalian Cells.” Dekker, New York. Beall, P. T., Hazlewood. C. F., and Rao, P. N. (1976). Science 192, 904-907. Bellamy, D., and Hinsull. S. M. (1978). Br. J. Cancer 37, 81-85. Bertsch, S . , Csontos, K., Schweizer, J., and Marks, F. (1976). Cell Tissue Kinet. 9, 445-457. Bhargava, P. M. (1977). J. Theor. Biol. 68, 101-137. Bross, I. D. J., and Blumenson, L. E. (1976). In “Fundamental Aspects of Metastasis” (L. Weiss, ed.), pp. 359-375. North-Holland Publ., Amsterdam. Bucher, N. L. R., Swaffield, M. N., and DiTroia, J. F. (1964). Cancer Res. 24, 509-512. Buell, D. N., Sox, H. C., and Fahey, J. L. (1971). In “Developmental Aspects of the Cell Cycle” (I. L. Cameron, G. M. Padilla, and A. M. Zimmerman, eds.), pp. 279-296. Academic Press, New York. Burnet, M. (1970). In “Immunological Surveillance” (M. Burnet, ed.), pp. 161-185. Pergamon, Oxford. Busch, H. (1978). Z. Krebsforsch. 92, 123-135. Cameron, I. L. (1971). In “Cellular and Molecular Renewal in the Mammalian Body” (I. L. Cameron and J. D. Thrasher, eds.), pp. 45-85. Academic Press, New York. Cameron, I. L. (1975). In “Cell Cycle in Malignancy and Immunity” (J. C. Hampton, ed.), pp. 76-103. Conf. 731005, National Technical Information Service, U.S. Dept. of Commerce, Springfield, Virginia. Carney, D. H., and Cunningham, D. D. (1978). Cell 15, 1341-1349. Cavallo, T., Sade, R., Folkman, J., and Cotran, R. S. (1972). 1.Cell Biol. 54, 408-420. Choie, D. D., and Richter, G. W. (1978). Cell Tissue Kinet. 11, 235-239. Chouroulinkov, I . , and Lasne, C. (1978). Bull. Cuncer 65, 255-264. Clarkson. B. D. (1974). In “Control of Proliferation in Animal Cells, Cold Spring Harbor Conferences on Cell Proliferation” (B. D. Clarkson and R. Baserga, eds.), Vol. I , pp, 945-972. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Clarkson, B., and Rubinow, S. I. (1977). In “Growth Kinetics and Biochemical Regulation of Normal and Malignant Cells’’ (B. Drewinko and R. M. Humphrey, eds.), pp. 591-628. Williams & Wilkins, Baltimore, Maryland. Clevecz, R. R. (1978). I n “Cell Reproduction: In Honor of Daniel Mazia, ICN-UCLA Symposia on Molecular and Cellular Biology” (E. R. Dirksen, D. Prescott, and C. F. Fox,eds.), Vol. XII, pp. 139-146. Academic Press, New York. Comings, D. E. (1973). Proc. Nutl. Acud. Sci. U.S.A. 70, 3324-3328. Cone, C. D., Jr. (1970). National Aeronautics and Space Administration Scientific Publication L-7164, pp. 1-49. Cone, C. D.. Murphy, T. E., and Tongier, M.(1968). J . Vu. Acud. Sci. May. Cooper, E. H., Bedford, A. J., and Kenny, T. E. (1975). Adv. Cancer Res. 21, 59-1 19. Curtis, A. S. G., and Seehar, G. M. (1978). Narure (London) 274, 52-53. Darzynkiewicz, Z . , Traganos, F., Andreeff, M., Sharpless, T., and Melamed, M. R. (1979). J . Histochem. Cytochem. 27, 478-485. DeCosse, J. J., and Gelfant, S. (1968). Science 162, 698-699. Ehret, C. F., Meinert, J. C., Groh, K. R., Dobra, K. W., and Antipa, G. A. (1977). In “Growth Kinetics and Biochemical Regulation of Normal and Malignant Cells” (B. Drewinko and R. M. Humphrey, eds.), pp. 49-76. Williams & Wilkins, Baltimore, Maryland. Epifanova, 0. I. (1971). In “The Cell Cycle and Cancer”(R. Baserga, ed.), pp. 145-190. Dekker, New York.

24

SEYMOUR GELFANT

Epifanova, 0. 1. (1977). Int. Rev. Cytol. Suppl. 5, 303-335. Fidler, I. J . (1978). Cancer Res. 38, 2651-2660. Folkman, J. (1976). Sci. Am. 234, 58-73. Gelfant, S. (1966). I n “Methods in Cell Physiology” (D. M. Prescott, ed.), Vol. 11, pp. 359-395. Academic Press, New York. Gelfant, S. (1977). Cancer Res. 37, 3845-3862. Gelfant, S . , and Grove, G. L. (1974). I n “Theoretical Aspects of Aging” (M. Rockstein, ed.), pp. 105-1 17. Academic Press, New York. Gelfant, S . , and Smith, J. G., Jr. (1972). Science 178, 357-361. Gerbase-DeLima, M.. Wilkinson, J., Smith, G. S . , and Walford, R. L. (1974). J. Gerontol. 28, 26 1 -268.

Gerson, D. F. (1978). I n “Cell Cycle Regulation” (J. R. Jeter, Jr., 1. L. Cameron, G. M. Padilla, and A. M. Zimmerman, eds.), pp. 105-131. Academic Press, New York. Gilbert, D. A. (1977). BioSystems 9, 215-228. Goldstein, L. (1978). I n “Cell Reproduction: In Honor of Daniel Mazia, ICN-UCLA Symposia on Molecular and Cellular Biology” (E. R. Dirksen, D. Prescott, and C. F. Fox, eds.), Vol.’XII, pp. 155-171. Academic Press, New York. Goldstein, L., and KO, C. (1978). Chromosoma 68, 319-325. Green, M. (1978). Perspect. Biol. Med. 21, 373-397. Haddox, M. K.. Furcht, L. T., Gentry, S. R., Moser, M. E., Stephenson, J., and Goldberg, N. D. (1977). In “Growth Kinetics and Biochemical Regulation of Normal and Malignant Cells” (B. Drewinko and R. M. Humphrey, eds.), pp. 255-269. Williams & Wilkins, Baltimore. Maryland. Heidelberger, C. (1975). Annu. Rev. Biochem. 44, 79-1 13. Higginson, J . (1969). I n “Canadian Cancer Conference, Proceedings of the Eighth Canadian Cancer Research Conference, Honey Harbour, Ontario” (J. F. Morgan, ed.), pp. 40-75. Pergamon, Oxford. Hirsch, H. R. (1978). Mech. Ageing Dev. 8, 51-62. Hoover, H. C., Jr., and Ketcham, A. S. (1975). Am. J . Surg. 130, 405-411. Hori, Y ., Perkins, E. H., and Halsall, M. K. (1973). Proc. SOC.Exp. Biol. Med. 144, 48-53. Houck, J. C., and Daugherty, W. F., Jr. (1974). “Chalones: A Tissue-Specific Approach to Mitotic Control.” MEDCOM Press, New York. Kay, M. M. B. (1978). I n “The Handbook of Cancer Immunology” (H. Waters, ed.), Vol. I , pp. 63-93. Garland STPM Press, New York. Kay, M. M. B. (1979). J. Invest. Dermatol. 73, 29-38. Konen. T. G . . Smith, G. S., and Walford, R. L. (1973). J . Immunol. 110, 1216-1221. Kripke, M. L., Gruys, E., and Fidler, I. J. (1978). Cancer Res. 38, 2962-2967. Kruisbeek, A. M. (1976). Mech. Ageing Dev. 5 , 125-138. Kurland, J. I., Traganos, F., Darzynkiewicz, Z., and Moore, M. A. S. (1978). Cell. Immunol. 36, 318-330.

Lerman, M. I. (1978). J. Theor. B i d . 73, 615-629. Lozzio, B. B., Lozzio, C. B., Bamberger, E. G., and Lair, S. V. (1975). Int. Rev. Cytol. 42, 1-47. Martin, R. G., and Stein, S. (1976). Proc. Natl. Acad. Sci. U.S.A. 73, 1655-1659. Marx, J. L. (1978). Science 201, 515-518. Mathies, M., Lipps, L., Smith, G. S . , and Walford, R. L. (1973). J. Gerontol. 28, 425-430. Milo, G. E., Jr., and DiPaolo, J. A. (1978). Nature (London) 275, 130-132. Moller G., and Moller, E. (1976). Transplant. Rev. 28, 3-97. Neidle, S. (1980). Nature (London) 283, 135. Nicholson, G. L. (1979). Sci. Am. 240, 66-76. Njus, D., Sulzman, F. M., and Hastings, J. W. (1974). Nature (London) 248, 116-120.

CYCLING

NONCYCLING CELL TRANSITIONS

25

Nowell, P. C. (1976). Science 194, 23-28. Nowell. P. C. (1978). Virchows Arch. B. Cell Pathol. 29, 145-150. Old, L. J. (1977). Sci. Am. 236, 62-79. Pardee, A. B. (1974). Proc. Natl. Acad. Sri. U .S . A . 71, 1286-1290. Pardee, A. B., DeAsua, L. J., and Rozengurt, E. (1974). In “Control of Proliferation in Animal Cells, Cold Spring Harbor Conferences on Cell Proliferation” (B. D. Clarkson and R. Baserga, eds.), Vol. I , pp. 547-561. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Pederson, T., and Bhorjee, J. S. (1979). J . Mol. Biol. 128, 451-480. Pederson, T . , and Gelfant, S. (1970). Exp. Cell Res. 59, 32-36. Penn, I., and Starzl, T. E. (1972). Transplantation 14, 407-417. Perkins, E. H., and Cacheiro, L. H. (1977). Mech. Ageing Dev. 6 , 15-24. Perkins, E. H., and Makinodan, T. (1971). Proc. Rocky MI. Symp. Aging. 1st Colo. Stare Univ. F t . Collins pp. 80-103. Phillips, T. L., and Leong, G. F. (1967). Cancer Res. 27, 286-292. Potter, V. R. (1978). Br. J. Cancer 38, 1-23. Refsum, S. B.. d. y.. and Berdal, P. (1967). Eur. J . Cancer 3, 235-236. Rhodes, J., Bishop, M., and Benfield, J. (1979). Science 203, 179-182. Rubin, H.. and Fodge, D. (1974). In “Control of Proliferation in Animal Cells, Cold Spring Harbor Conferences on Cell Proliferation” (B. D. Clarkson and R. Baserga, eds.), Vol. 1, pp. 801-816. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Sheppard, J. R.,and Bannai, S. (1974). In “Control of Proliferation in Animal Cells, Cold Spring Harbor Conferences on Cell Proliferation” (B. D. Clarkson and R. Baserga, eds.), Vol. I , pp. 571-579. Cold Spring Harbor Lab., Cold Spring Harbor, New York. Simpson-Herren, L., Sanford, A. H., and Holmquist, J. P. (1974). Cell Tissue Kinet. 7, 349-361. Slack, N. H., and Bross, 1. D. J. (1975). Br. J . Cancer 32, 78-86. Steel, G. G. (1967). Eur. J . Cancer 3, 381-387. Steel, G. G. (1973). In “Cancer Medicine” (J. F. Holland and E. Frei, eds.), pp. 125-140. Lea & Febiger, Philadelphia, Pennsylvania. Steel, G. G. (1977). “Growth Kinetics of Tumours.” Oxford Univ. Press (Clarendon), London and New York. Stein, G. S., Stein, J. L., and Thornson, J. A. (1978). Cancer Res. 38, 1181-1201. Stein-Werblowsky, R. (1978). Br. J. Exp. Pathol. 59, 386-389. Tannock, I. F. (1968). Br. J . Cancer 22, 258-273. Tannock, I. (1978). Cancer Trear. Rep. 62, 1117-1133. Temin, H. M.(1971). In “The Biology of Oncogenic Viruses, Proceedings of the Second Lepetit Colloquium, Paris” (L. G . Silvestri, ed.). pp. 176-187. North-Holland Publ., Amsterdam. Temin, H. M.(1972). Proc. Natl. Acad. Sci. U.S . A . 69, 1016-1020. Temin, H. M. (1974). Cancer Res. 34, 2835-2841. Todaro, G. J., and Heubner, R. J. (1972). Proc. Natl. Acad. Sci. (1. S. A . 69, 1009-1015. Tyan, M. L. (1976). Proc. SOC.Exp. Biol. Med. 152, 33-35. Van’t Hof, J . , Hoppin, D. P., and Yagi, S. (1973). Am. J . Bot. 60, 889-895. Viadana, E., Bross, 1. D. J., and Pickren, J . W. (1978). Oncology 35, 114-126. Weiss, L. (1977a). Int. J. Cancer 20, 87-92. Weiss, L. (1977b). GANN Monogr. Cancer Res. 20, 25-35. Yunis, E. J . , and Lane, M. A. (1979). J . Invest. Dermatol. 73, 24-28. Zada-Hames, 1. M.,and Ashworth, J. M. ( 1978). J . Cell Sci. 32, 1-20.