Cellular topography within the highly polarized surface epithelia can be used to identify the location of the stem cells. In some instances, this can be quite precise and allows the characteristics of stem cells to be studied. Our current knowledge of the stem cell population in murine epidermis and small intestinal crypts is reviewed. In the epidermis, the stem cells would appear to make up about 10 % of the basal layer and are distributed towards the centre of the basal layer component of the epidermal proliferative unit. These cells have a long cell cycle and are probably the same cells that retain both tritiated thymidine and radioactively labelled carcinogens for long periods of time. This label retention permits the labelling of the putative stem cell compartment. Over recent years, there has been an accumulation of information indicating various types of heterogeneity within the basal layer, much of which can be interpreted in relation to cellular hierarchies. In the small intestine, cell positions can be fairly precisely identified and the stem cell zone identified. Complex modelling of a wide range of cell kinetic experiments suggests that each crypt contains between 4 and 16 steady state functional stem cells. Radiobiological experiments suggest that up to 32 cells may be capable of clonal regeneration. The repopulation of the clonogenic cell compartment has been determined and the doubling time measured to be 19•7h. Such studies should throw further light on the behaviour of stem cells and identify the timing of periods of increased and decreased cell proliferation (activation and suppression of controls).

The constantly replacing epithelial tissues in the adult body are characterized by having a high degree of structural order. The tissue is polarized as are the individual cells also. Although epithelial tissues differ in structure and function they share a unifying theme in that they all undergo continual lifelong cell turnover which is characterized by specific patterns of proliferation within morphologically defined units. The long life of these units, the patterns of proliferation and the radiobiological response (clonal regeneration) suggest the units contain stem cells. If the question is asked ‘where are the cells that are ultimately responsible for all the cell replacement - the stem cells?’ the superficial answer is generally quite clear. In simple flat stratified epithelia such as the epidermis on the back skin of the mouse they must be somewhere in the basal layer because cell proliferation occurs only in the basal layer. In more complex undulating stratified epithelia where there are rete ridges or pegs, and one example that has been particularly studied in the mouse is the dorsal surface of the tongue and its filiform papilla (Hume & Potten, 1976), the answer is in the basal layer but specifically in those regions of the basal layer that project deepest into the connective tissue. Such an undulating stratified epithelium may be a model for all such epithelia including many regions of the epidermis in man and other regions of oral epithelia such as parts of the gingiva. In simple columnar epithelia such as the mucosa of the small intestine the answer is at the deepest part of the epithelium in the lamina propria which represents the origin of all the rapid cell displacement. In the small intestine this means near the base of the crypt immediately above, or scattered amongst, the Paneth cells. In the colon it may mean literally the cell or few cells right at the base of the crypt. Such precise topographical distribution for the stem cells means that they can be studied in a fairly direct manner. We should like to review our current understanding of some of the characteristics of these special cells.

It has not yet been determined whether the stem cells are merely proliferative cells which occur at special positions or whether they themselves differ intrinsically from other proliferative cells. If their function is merely determined by their position it is unclear what characterizes these positions, micro-environments or niches.

The epidermis

The cells on the basal layer of the epidermis on mouse back skin are arranged in a particular pattern (Fig. 1) which is related to the columnar packing organization of the cells in the suprabasal strata (Fig. 2). These columns of cells are particularly evident from the surface appearance of the epidermis after silver staining of sheets of epidermis or in scanning electron micrographs or even in carefully prepared sections of murine skin. They represent the historical record of the proliferative activity of the group of about 10-11 basal cells immediately.beneath the column and the whole unit, basal cells plus column, represents somewhat autonomous units of proliferation, called epidermal proliferative units (EPUs) (Fig. 2) (Potten, 1974, 1981, 1985). Studies on the effects of ionizing or ultraviolet irradiation of skin and various experiments involving tritiated thymidine labelling together with other morphological and functional cell studies suggest that the basal layer has considerable cellular heterogeneity (Table 1). Each basal cell has a characteristic position within the EPU. The position correlates with differences in proliferative activity and the response of the cells to tumour promoters. The radiobiological data suggest that each EPU probably possesses a single clonogenic cell, i.e. the stem or clonogenic cell compartment comprises about 10 % of the basal cells. Interpretation of cell kinetic experiments suggests that the cell replacement within the basal layer is best explained on the basis of a cell lineage comprising three generations of dividing transit cells. Taking the clonogenic cell estimates, the overall radiobiological observations and all the cell kinetic experiments the current most appropriate model for epidermis is illustrated in Fig. 2. There remains some doubt about the fine details of the structure of the hierarchy, and the position within the hierarchy of individual cells, but the stem cell population can be assumed to make up less than 20 % (probably 10 %) of the basal layer within which about 60 % of the cells are proliferative. The cell cycle of the stem cells is estimated to be about twice that of the transit cells with the overall average cell cycle time being about 125 h. The slow cell cycle of the stem cells may provide a means of identifying these cells (Fig. 1).

Table 1.

Evidence for cellular heterogeneity in the epidermal basal layer (Clausen & Potten, 1988)

Evidence for cellular heterogeneity in the epidermal basal layer (Clausen & Potten, 1988)
Evidence for cellular heterogeneity in the epidermal basal layer (Clausen & Potten, 1988)
Fig. 1.

Autoradiographs of a sheet of epidermis separated from the dermis by exposure to O•5 % acetic acid overnight. The epidermal sheet was prepared from a mouse that received twice daily injections of tritiated thymidine for 3 days starting on the third day after birth. The mouse was killed when it was 8 weeks old. The characteristic pattern of the clustered basal cells can be seen which relate to the EPU (see text). Cells which retain label can be easily seen (arrows). Many are centrally located in the cluster of EPU basal cells, i.e. at the putative stem cell location.

Fig. 1.

Autoradiographs of a sheet of epidermis separated from the dermis by exposure to O•5 % acetic acid overnight. The epidermal sheet was prepared from a mouse that received twice daily injections of tritiated thymidine for 3 days starting on the third day after birth. The mouse was killed when it was 8 weeks old. The characteristic pattern of the clustered basal cells can be seen which relate to the EPU (see text). Cells which retain label can be easily seen (arrows). Many are centrally located in the cluster of EPU basal cells, i.e. at the putative stem cell location.

Fig. 2.

Diagrammatic representation of the epidermal proliferative units (EPUs) in mouse epidermis as viewed from the surface (e.g. with SEM) — top of drawing; in section - middle portion; and in an epidermal sheet - lower portion (compare with Fig. 1). Two possible cell lineages that may explain the cell replacement process are shown below. A full lineage with three transit generations was deduced from an analysis of three cell kinetic experiments (Potten et al. 1982). An abbreviated lineage with differentiation from each transit generation was deduced from an analysis of the changing patterns with time of clusters of labelled cells (Loeffler et al. 1987).

Fig. 2.

Diagrammatic representation of the epidermal proliferative units (EPUs) in mouse epidermis as viewed from the surface (e.g. with SEM) — top of drawing; in section - middle portion; and in an epidermal sheet - lower portion (compare with Fig. 1). Two possible cell lineages that may explain the cell replacement process are shown below. A full lineage with three transit generations was deduced from an analysis of three cell kinetic experiments (Potten et al. 1982). An abbreviated lineage with differentiation from each transit generation was deduced from an analysis of the changing patterns with time of clusters of labelled cells (Loeffler et al. 1987).

Studies on label-retaining cells

Slowly cycling cells in mouse epidermis which are positioned towards the centre of the cluster of basal cells associated with each epidermal proliferative unit can be identified by continuous tritiated thymidine labelling studies (Potten, 1974) or by labelling basal cells and waiting for the label to dilute to background levels in the faster cycling cells (Fig. 1) (Bickenbach, 1981; Morris et al. 1985, 1986; Potten etal. 1982). This is most effective when the initial labelling protocol produces high labelling indices, such as after continuous labelling of young adult or neonatal mice. The cells that retain label for longer than the average (label-retaining cells, LRC) can be assumed to have divided less frequently, i.e. be more slowly cycling. Other cell kinetic analyses and modelling studies have suggested that the stem cells are more slowly cycling with a cell cycle time of 200 h (or about 8 days) and studies on the label-retaining cells show that most of these cells are located towards the centre of the EPU (Fig. 1), the site suggested for the stem cells.

The loss of label in the faster cycling cells could be attributed to the cell cycle duration alone or also to the fact that many of these cells may be dividing transit cells that emigrate from the basal layer and are exfoliated from the system. Based on our current cell kinetic model of epidermis, about half of the initially labelled cells would be expected to divide once and then leave the basal layer between 30 and 100 h. A second cohort would be expected to divide twice and leave the basal layer between about 145 and 200 h after the initial labelling. The final, third cohort of transit cells might also have a long cell cycle but nevertheless would be expected to emigrate between about 345 and 400 h after their third post-labelling cell division. However, during this period the stem cells originally labelled would have produced new transit cells with a labelling density half that observed initially. There are several points in time when it can be expected that the stem and early transit cells might have a higher grain density than the majority of cells. However, it is unlikely, based on currently assumed cell cycle differences, that this would ever be more than a two-fold difference in grain density, which is not consistent with the observations illustrated by Fig. 1. In order to obtain the grain differential commonly seen in LRC experiments it would be necessary to assume that the stem cells have a slower cell cycle than has been assumed previously or that other mechanisms operate to conserve radioactivity in the stem cells.

Transit cells, which characteristically have a limited life in the basal layer, are unlikely to be involved in carcinogenesis, which characteristically possesses a long latent period usually measured in months for mouse skin. It is interesting in this context that radioactively labelled carcinogens, tritiated or carbon-14-labelled benzo(a)pyrene (BP) (Morris et al. 1986) and tritiated dimethylbenze(a)anthracene (DMBA) (preliminary unpublished observations), are both retained in a few cells in the basal layer predominantly located towards the centre of the EPU. Other carcinogen label-retaining cells can be observed in the hair follicles as is also observed after the [3H]thymidine ([3H]dThd)-labelling protocols. Double labelling experiments with [3H]dThd and [14C]BP suggest both molecules are detecting the same slowly cycling population of cells.

Other preliminary unpublished data suggest that the cells retaining [3H]dThd label retaining cells behave in culture in a different manner to most pulse labelled cells. Cell suspensions that are prepared from skin which had been pulse labelled with tritiated thymidine produce clusters of cells (colonies) in culture which, if they contain labelled cells, have most of the labelled cells as isolated single labelled cells in a cluster of otherwise unlabelled cells, suggesting that they have been through no cell divisions. In contrast, suspensions prepared from skin which contain LRCs (as a consequence of a multiple labelling protocol one or more months earlier in life) produce colonies many of which have several lightly labelled cells, suggesting that the LRCs divided several times during the formation of the cluster of cells. The observations are consistent with the view that many of the [3H]dThd label-retaining cells possess an in vitro clonogenic capacity further supporting the view that the LRCs may be representative of a population of slow cycling stem cells. The LRCs are also the cells that can be seen to retain carcinogens for long periods of time and hence may be the cells that ultimately produce cancers.

The significance of carcinogen label-retaining cells in two-stage murine cutaneous carcinogenesis

In two-stage carcinogenesis, benign and malignant cutaneous neoplasma are induced on the backs of mice following initiation with a subtumorigenic exposure to a carcinogen and subsequent repetitive exposure to a noncarcinogenic, hyperplasiogenic tumour promoter. Tumour initiation is thought to involve conversion of some epidermal cells into latent neoplastic cells; promotion elicits expression of the neoplastic change (Boutwell, 1974; Scribner & Suss, 1978; Slaga, 1984). Effective initiators of skin carcinogenesis such as polycyclic aromatic hydrocarbons have in common the capacity for covalent binding as activated electrophiles to cellular DNA and the ability to cause an irreversible alteration of the genome (Dippie et al. 1984; Miller & Miller, 1981; Osborn, 1984; Ashurst et al. 1982; Nakayama et al. 1984). The consequences of tumour initiation are essentially irreversible and not expressed in the absence of promotion. The tumour responses are evoked whether promotion is begun one week or one year after the exposure to the carcinogen (Boutwell, 1964; Roe et al. 1972; Stenbach et al. 1981; Van Duuren et al. 1975; Loehrke et al. 1983; Hennings & Boutwell, 1970; Argyris, 1985α) and are surprisingly similar for a tissue such as the epidermis which is characterized by continuous turnover and cyclic growth and regression of the hair follicles (Hennings & Yuspa, 1985; Argyris, 1963; Potten, 1983; Iversen et al. 1968; Leblond, 1964). Conceivably, any epidermal cell could become and remain initiated. Alternatively, a subpopulation having singular growth characteristics could be the target for the carcinogen. The identification of the target cells which maintain the lifelong potential to form tumours is consequently an objective with considerable significance.

Although previous studies had demonstrated the fairly uniform initial binding of radioactively labelled carcinogens throughout the epidermal basal layer (Borum, 1960; Bibby & Smith, 1977; Nakai & Shubik, 1964; Solun, 1970; Tarnowski, 1970) very little was known concerning possible relationships between keratinocyte maturity and the persistence of the radioactively labelled carcinogen. We therefore used [3H]‐ and [14C] benzo (a)pyrene to identify in the dorsal epidermis of mice carcinogen label-retaining cells. We compared them in number and distribution to two other distinct epidermal cell populations: (a) the slowly cycling, [3H]thymidine label-retaining cells (Bickenbach, 1981; Mackenzie & Bickenbach, 1985; Morris et al. 1985; Potten, 1986), and (b) those with cellular kinetic features of maturing keratinocytes prior to displacement from the basal layer (Potten, 1986; Iversen et al. 1968; Mackenzie, 1970; Christophers, 1971).

The results (Morris et al. 1986) have demonstrated that a subset of epidermal basal cells retain radioactively labelled carcinogen much longer than most of the basal cells. These carcinogen label-retaining cells resembled the slowly cycling [3H]thymidine LRCs rather than the ‘maturing’ basal cells (demonstrated four days after a single pulse of [3H]thymidine at 8:00 am) in number, position in the tissue architecture, long turnover time, and in mitotic activation following an application of the tumour promoter 12-O-tetradecanoyl phorbol-13-acetate (TPA). Double emulsion autoradiographs prepared one month after continuous labelling with [3H]thymidine for one week and treatment with [14C]B(a)P revealed doubly labelled nuclei and thus provided evidence that the carcinogen label was retained by the slowly cycling population.

These observations (Morris et al. 1986) are consistent with the hypothesis that radioactively labelled carcinogen persists in the slowly cycling cells because they are slow cycling. The observed reduction in the grain density and thus the number of labelled nuclei detected by autoradiography supports this hypothesis. However, the present data cannot exclude the possibilities that (1) all of the adduct might eventually be removed by repair, or (2) that low levels of the adduct might indeed by present but remain undetected by autoradiography. There are alternative possible explanations for the persistence of carcinogen label in a subpopulation of keratinocytes: (1) slowly cycling cells might have an inherently low or a carcinogen-induced reduction in their ability to repair adducts relative to other basal cells (Reddy et al. 1988), (2) the retained adducts might be hidden from the repair enzymes (Bustin et al. 1983; Lajtha, 1979) or, (3) some of the adducts might be irreparable. All of these possibilities deserve further investigation.

The persistence of radioactively labelled carcinogen in the slowly cycling cells is an observation of considerable significance for present and future investigations in skin carcinogenesis especially as it is related to tumour initiation (see below). The carcinogen LRCs may indeed be initiated cells, not necessarily because they retain carcinogen, but because they are stem cells affected before commitment to terminal differentiation (Lajtha, 1979; Buick & Pollak, 1984). Although the long-term presence of the adduct might possibly represent an enduring potential for the formation of initiated cells (Randerath et al. 1983) in the murine skin system, the similar tumour responses at one week and one year after initiation (Boutwell, 1964; RoeβtűZ. 1972; Stenback et al. 1981; Van Duuren et al. 1975; Loehrke et al. 1983; Hennings & Boutwell, 1970; Argyris, 1985α) suggest that the initiated state is established quickly and then maintained.

The nature of the lesion that makes an epidermal cell a ‘latent neoplastic cell’ is not known although alterations affecting both terminal differentiation (Kulesz-Martin et al. 1980; Kawamura et al. 1985) or self-renewal of a stem cell population (Lajtha, 1979; Buick & Pollak, 1984) have been postulated. Elevated expression and point mutation of the Ha-ras proto-oncogene has been observed in murine papillomas after nine weeks of promotion; earlier detection was not possible with the present experimental techniques (Pelling et al. 1987). Other significant genetic lesions cannot be excluded at this time. Nor do we know the identity of the cell that receives and maintains the lesion, although an as yet unidentified epidermal stem cell has long been suggested as a likely candidate. Normal epidermal cells having some of the characteristics of stem cells such as the slowly cycling label-retaining cells and dark basal keratinocytes are consequently under intensive investigation. Admittedly, the possibility that a carcinogen might impose characteristics of slow cycling or delayed displacement upon cells that would not normally express these characteristics cannot be excluded at this time.

Why do initiated mice not develop tumours unless they are promoted? It is possible that the number of divisions required of a slowly cycling population to form a tumour might require a time longer than the lifespan of the initiated animal. Alternatively, there may be something about the process of tumour promotion, e.g. regenerative hyperplastic growth (Argyris, 19856), inflammation, induction of dark cells (Slaga & Klein-Szanto, 1983), activation of stem cells (Lajtha, 1979; Buick & Pollak, 1984), that is needed to establish the condition of initiation or for an initiated cell to express its neoplastic potential. An eventual resolution of these problems will require a better understanding of how proliferation and maturation are regulated in continually renewing tissues such as epidermis.

Small intestine

The crypt of the small intestine in the mouse is probably one of the most extensively studied mammalian epithelial systems from the point of view of its cell kinetics and possibly also from that of its radiation response. As a consequence the cell replacement process is now well understood and these crypts can be regarded as a model system for other sites in the gut. It is an extremely dynamic system with a cell dividing every 5 min in each crypt and the progeny moving rapidly from the crypts onto the villus and exfoliating within their short life span of about 3-4 days (Fig. 3). The cells ultimately responsible for the considerable cell loss are clearly to be found at the origin of all this cell migration, which can be traced to the base of the crypt. Unfortunately the details of the cell replacement at the crypt base are complicated by the presence of 20-30 mature functional Paneth cells. The lineage, or cell migration, ancestors - the stem cells - are to be found either immediately above the Paneth cells in an annulus that contains about 16-18 cells, not all of which are necessarily stem cells (Loeffler et al. 1986; Potten & Loeffler, 1987), or in this annulus and scattered amongst the Paneth cells (Cheng & Leblond, 1974; Bjerknes & Cheng, 1981). The entire structure and its cell proliferation can be summarized as shown in Fig. 3. The proliferative units here are the crypts, which are closed systems, derived from single cells during development, i.e. they are clones (Ponder etal. 1985), which in the adult contain several stem cells each with its own lineage of dividing transit cells. The number of stem cells remains unclear but is probably not less than about 4 per crypt and not more than 16 in steady state, but there may be as many as 32 clonogenic cells - the most recent value determined from several radiation experiments (Potten et al. 1987). It is believed that the stem cells in the crypt have a cell cycle time about twice that of the dividing transit cells (24 h versus 12 h).

Fig. 3.

Diagrammatic representation of the spatial and proliferative organization of the mucosa in the small intestine of the mouse. The inter-relationship between the villi and longitudinal axis of the crypt is shown. Such sections can be used to analyse the changes that occur in proliferative activity with cell position. The base of the crypt can be identified by the presence of Paneth cells and the central crypt base cell is usually identified as at cell position one. There are usually about 20 cell positions along the side of the crypt (the crypt column). The crypt in total contains about 250 cells and its surface view is represented on the left. The cell lineage that most effectively explains the cell replacement is illustrated on the right. Cells of particular hierarchical status in this lineage can be related to cell positions within the crypt as indicated in the diagram. The cell replacement process in the crypt could be explained on the basis of up to 16 functional stem cells per crypt (Loeffler et al. 1986). It is likely that the crypt contains about 32 cells that possess a clonogenic capacity (Potten et al. 1987).

Fig. 3.

Diagrammatic representation of the spatial and proliferative organization of the mucosa in the small intestine of the mouse. The inter-relationship between the villi and longitudinal axis of the crypt is shown. Such sections can be used to analyse the changes that occur in proliferative activity with cell position. The base of the crypt can be identified by the presence of Paneth cells and the central crypt base cell is usually identified as at cell position one. There are usually about 20 cell positions along the side of the crypt (the crypt column). The crypt in total contains about 250 cells and its surface view is represented on the left. The cell lineage that most effectively explains the cell replacement is illustrated on the right. Cells of particular hierarchical status in this lineage can be related to cell positions within the crypt as indicated in the diagram. The cell replacement process in the crypt could be explained on the basis of up to 16 functional stem cells per crypt (Loeffler et al. 1986). It is likely that the crypt contains about 32 cells that possess a clonogenic capacity (Potten et al. 1987).

Heterogeneity in the stem cell population

There are studies that suggest that some of the crypt stem cells are very sensitive cells, easily killed by radiation (Potten, 1977), and that the cell cycle characteristics of the neighbouring unaffected cells may be very easily disturbed by minor cytotoxic insults including the internal weak beta irradiation resulting from the incorporation of trace levels of tritiated thymidine (Cheng & Leblond, 1974; Potten, 1977, 1986).

The cells killed by small doses of radiation number about 6 per crypt and apparently die through apoptosis and can easily be recognized in H & E sections within 3-6 h. Yet other cells in the crypt base exhibit a high radiation sensitivity when G2 progression (mitotic delay) is studied (Chwalinski & Potten, 1986). Others apparently can enter a reduced cell cycle time (e.g. 16 h instead of 24 h) very rapidly after minor perturbations (Potten, 1986). The cells that die after low doses of radiation exhibit differences in their circadian behaviour to that of the clonogenic population as a whole (Ijiri & Potten, 1988), suggesting they are a distinct subpopulation of cells which because of their position may be part of the stem cell compartment. These cells may have a sensitivity characterized by a Do value of 20-30 cGy while the clonogenic cells have a Do value of 135 cGy. Such observations lead to the conclusion that the stem/clonogenic pool is probably heterogeneous with either distinct subpopulations or a hierarchy (age structure) of cells.

Regeneration of clonogenic cells in the intestine

Doses of radiation above about 8-9 Gy will completely sterilize some crypts and clonal regeneration assays have been developed (Withers & Elkind, 1969, 1970), which have been extensively used and from which the number of clonogenic cells can be estimated (Hendry & Potten, 1974; Potten & Hendry, 1975; Potten et al. 1987). The repopulation of the clonogenic cell population has recently been measured (Fig. 4) and this provides a doubling time of 19•7h for the regenerating clonogenic cells (Potten et al. 1988). In principle the relationship between the clonogenic cell repopulation, the crypt repopulation and the crypt cell production rate would permit the true cell cycle time for the clonogenic cells and their self-maintenance probability {P) to be estimated. At present the unknown element here is what changes occur, over these post-irradiation times, in the number of transit cell generations. In the normal steady state crypt (where P = 0 - 5) the cell migration velocity per column (V) is determined by the number of stem cells (Ns), their cell cycle time (Tc) and the number of transit cell divisions (L) as follows:

Fig. 4.

The results of a recent experiment (Potten et al. 1988) where the number of clonogenic cells per crypt (arithmetic mean) has been calculated using the approach described by Hendry (1979) during the first 7 days after an initial dose of 8•O Gy gamma radiation. In this experiment the control crypts contained 46 ± 7 clonogenic cells. The clonogenic cells repopulate the crypts with a doubling time of 19•7 ± 2•4h. The data are consistent with an initial mitotic delay of about 18 h, i.e. about 2•25 h/Gy. On days 5-7 the number of clonogenic cells overshoots the control value by a factor of 2•2 (average for days 5—7). The regeneration of the clonogenic compartment can be compared with the total cellularity of the crypt and the cell output from the crypt. Such a comparison indicates that some clonogenic cells must differentiate into the transit population during the regeneration phase.

Fig. 4.

The results of a recent experiment (Potten et al. 1988) where the number of clonogenic cells per crypt (arithmetic mean) has been calculated using the approach described by Hendry (1979) during the first 7 days after an initial dose of 8•O Gy gamma radiation. In this experiment the control crypts contained 46 ± 7 clonogenic cells. The clonogenic cells repopulate the crypts with a doubling time of 19•7 ± 2•4h. The data are consistent with an initial mitotic delay of about 18 h, i.e. about 2•25 h/Gy. On days 5-7 the number of clonogenic cells overshoots the control value by a factor of 2•2 (average for days 5—7). The regeneration of the clonogenic compartment can be compared with the total cellularity of the crypt and the cell output from the crypt. Such a comparison indicates that some clonogenic cells must differentiate into the transit population during the regeneration phase.

formula
as was demonstrated from a comprehensive mathematically based model of the crypt that we have developed (Loeffler et al. 1986; Potten & Loeffler, 1987). The model can be summarized as shown in Fig. 5.
Fig. 5.

The cell replacement process in the crypt can be mathematically modelled using a simple matrix to represent the crypts which can be regarded as slit-open and the sheet of cells laid flat (Loeffler et al. 1986; Potten & Loeffler, 1987). The best fit to all the available data is obtained with a cell lineage with up to 16 stem(s) cells and 4 transit generations (1-4). The matrix shows a representative distribution of labelled (S phase) cells (large numbers). The matrix can be reassembled to form a crypt (right-hand diagram). Consideration of the mitotic polarity in mathematical modelling suggests that the shape of the crypt and the cell production can be controlled by two interacting processes (forces): one which tends to reduce the circumference, squeezes the crypt and forces cells to move out of the top; and a counter balancing force, which can be attributed to the mitotic activity which tends to increase the circumference.

Fig. 5.

The cell replacement process in the crypt can be mathematically modelled using a simple matrix to represent the crypts which can be regarded as slit-open and the sheet of cells laid flat (Loeffler et al. 1986; Potten & Loeffler, 1987). The best fit to all the available data is obtained with a cell lineage with up to 16 stem(s) cells and 4 transit generations (1-4). The matrix shows a representative distribution of labelled (S phase) cells (large numbers). The matrix can be reassembled to form a crypt (right-hand diagram). Consideration of the mitotic polarity in mathematical modelling suggests that the shape of the crypt and the cell production can be controlled by two interacting processes (forces): one which tends to reduce the circumference, squeezes the crypt and forces cells to move out of the top; and a counter balancing force, which can be attributed to the mitotic activity which tends to increase the circumference.

Carcinogenic target cells

One challenge for future research is to determine whether target cells for carcinogenesis are solely stem cells or whether cancer may also arise on occasions from dividing transit cells. The arguments in favour of cancer being a disease solely of stem cells can be summarized as follows. The latent period between exposure to known carcinogens and the development of tumours is long, generally months in mice or decades in humans and involves several steps, the first of which is tumour initiation. Thus, initiated cells or their progeny must persist in epithelial tissues at least for these lengths of time. The initiation event is commonly thought of as a mutational event presumably one associated with the genes controlling cell proliferation generally or self-maintenance probability more specifically (expressed by some as a loss of commitment to terminal differentiation). Although such mutational events could in principle occur in any cell within the tissue, the high degree of structural polarity and rapid cell migration would mean that most such mutated cells would be irrevocably shed within a few days of being mutated. It is only the stem cells that are permanent residents and hence persist for long enough to account for the lag period (Potten, 1984). The argument against this would suggest that any cell could be a potential cancer producing cell. Any initiated mid crypt transit cell would have to resist further differentiation and emigration in order to produce a cancer.

This is hard to envisage for a tissue such as the intestine. It would imply that the initiation/mutational event affected not only some aspect of cell proliferation but simultaneously also affected the basement membrane and cell to cell relationships such that the cell no longer moved. Epithelial cells are tightly bound to each other by desomosomes and tight junctions and migrate as clusters or discrete columns of cells (Schmidt et al. 1985). It is hard to see how a cell in the gut could avoid the movement of its adjoining cells - how it could in fact step out of line of the moving column of cells in the epithelial sheet. It would be like trying to stop moving on an escalator packed solid with other moving people. Since, at least in some aspects, a transit cell could be considered to be differentiated, it would also imply that the single mutational event also resulted in the dedifferentiation of the cell.

Stem cell division potential

In epithelial tissues the number of transit cell divisions lies between 2 and 4 as a consequence of which the stem cell population will comprise between 30 % and 10 % of the proliferative compartment of the tissue (see Fig. 6, which places epithelia within the general context of all replacing tissues). The cell cycle duration for the stem cells in the various replacing tissues of the mouse is generally poorly defined but approximate estimates are as follows (Potten, 1986); tongue - 24 h, epidermis - 200h, small intestine - 24h (16-30h), colon 15-22h, bone marrow - 100-120h, testis > 60 h. Consequently, for a laboratory mouse that may live for up to 3 years (which may be up to 6 times its natural life expectancy in the wild) the stem cells will divide between about 100 times (epidermis) and about 1000 times (tongue and intestine); bone marrow - 200 to 300 times and testis up to 400 times. Bone marrow transplantation experiments (from old mice to lethally irradiated juveniles or juvenile W/W mice, which have a bone marrow stem cell deficiency) suggest that the haemopoietic stem cells can be transplanted at least 5 times and that the stem cells then divide as many as 1000 times (Harrison, 1973; Schofield, 1978; Harrison & Astle, 1982), i.e. 5 times the number in the life span of a laboratory mouse. Thus these stem cells possess a division potential far in excess (up to 30 times) of what may be required for the life of an average wild mouse. The potential may be even higher since the act of transplanation may require extra divisions to overcome transplantation damage.

Fig. 6.

The cellular hierarchies in epithelial tissues placed in the perspective of all replacing tissues. The precise number of amplifying transit cell divisions may be somewhat uncertain and variable. There may be up to 13 amplification transit cell divisions in the erythropoietic lineage in the bone marrow. In spermatogenesis there may be up to 12 mitotic cell divisions but there is also appreciable spontaneous cell death at certain points in the lineage, which reduces the effective amplification. The relationship between the number of transit cell divisions and the concentration of stem cells is illustrated.

Fig. 6.

The cellular hierarchies in epithelial tissues placed in the perspective of all replacing tissues. The precise number of amplifying transit cell divisions may be somewhat uncertain and variable. There may be up to 13 amplification transit cell divisions in the erythropoietic lineage in the bone marrow. In spermatogenesis there may be up to 12 mitotic cell divisions but there is also appreciable spontaneous cell death at certain points in the lineage, which reduces the effective amplification. The relationship between the number of transit cell divisions and the concentration of stem cells is illustrated.

Al-Barwari
,
S. E.
(
1978
).
Cell and population kinetics in the irradiated skin
.
Ph.D. Thesis, University of Manchester
.
Allen
,
T. D.
&
Potten
,
C. S.
(
1974
).
Fine structural identification and organisation of the epidermal proliferative units
.
J. Cell Sci
.
15
,
291
319
.
Argyris
,
T. S.
(
1963
).
Hair growth cycles and skin neoplasia
.
Nat. Cancer Inst. Monogr
.
10
,
23
43
.
Argyris
,
T. S.
(
1985a
).
Promotion of epidermal carcinogenesis by repeated damage to mouse skin
.
Am. J. indust. Med
.
8
,
329
337
.
Argyris
,
T. S.
(
1985b
).
Regeneration and the mechanism of epidermal tumour promotion
.
CRC Crit. Rev. Toxicol
.
14
,
211
258
.
Ashurst
,
S. W.
,
Cohen
,
G. M.
,
Nesnow
,
S.
,
DiGiovanni
,
J.
&
Slaga
,
T. J.
(
1982
).
Formation of benzo(a) pyrene/DNA adducts and their relationship to tumour initiation in mouse epidermis
.
Cancer Res
.
43
,
1024
1029
.
Barrandon
,
Y.
&
Green
,
H.
(
1985
).
Cell size as a determinant of the clone-forming ability of human keratinocytes
.
Proc. natn. Acad. Sci. U.S.A
.
82
,
5390
5394
.
Bibby
,
M. C.
&
Smith
,
G. M.
(
1977
).
High-resolution autoradiographic localization of 3methylcholanthrene-3H in the skin of Balb/c mice after topical application
.
Br. J. Derm
.
97
,
429
436
.
Bickenbach
,
J. R.
(
1981
).
Identification and behaviour of label-retaining cells in oral mucosa and skin
.
J. Dent. Res
.
60
,
1611
1620
.
Bjerknes
,
M.
&
Cheng
,
H.
(
1981
).
The stem cell zone of the small intestinal epithelium. I. Evidence from Paneth cells in the adult mouse
.
Am. J. Anat
.
160
,
51
63
.
Borum
,
K.
(
1960
).
Localization of radioactivity in mouse skin following one application of 14C-labelled 9,10-dimethy l,2-benzanthracene
.
Acta Un. int. Cancr
.
16
,
60
62
.
Boutwell
,
R. K.
(
1964
).
Some biological aspects of skin carcinogenesis
.
Prog. exp. Tumor Res
.
4
,
207
250
.
Boutwell
,
R. K.
(
1974
).
The function and mechanism of promoters of carcinogenesis
.
CRC Crit. Rev. Toxicol
.
2
,
419
443
.
Buick
,
R. N.
&
Pollak
,
M. N.
(
1984
).
Perspectives on clonogenic tumour cells, stem cells, and oncogenes
.
Cancer Res
.
44
,
4909
4918
.
Bustin
,
M. B.
,
Kurth
,
P. D.
,
Slor
,
H.
&
Seidman
,
M.
(
1983
).
Immunological studies on the influence of chromatin structure on the binding of a chemical carcinogen to the genome
.
In Application of Biological Markers to Carcinogen Testing
(ed.
H. A.
Milman
&
S.
Sell
), pp.
349
371
.
New York
:
Plenum Publishing Corporation
.
Cheng
,
H.
&
Leblond
,
C. P.
(
1974
).
Origin differentiation and renewal of four main epithelial cell types in the mouse intestine. V. Unitarian theory of the origin of the four epithelial cell types
.
Am. J. Anat
.
141
,
537
563
.
Christophers
,
E.
(
1971
).
Cellular architecture of the stratum corneum
.
J. invest. Derm
.
56
,
165
170
.
Chwalinski
,
S.
&
POTTEN
,
C. S.
(
1986
).
Radiation induced mitotic delay: duration, dose and cell position dependence in the crypts of the small intestine in the mouse
.
Int. J. Rad. Biol
.
49
,
809
819
.
Clausen
,
O. P. F.
,
Kirkhus
,
B.
,
Elgjo
,
K.
,
Pedersen
,
S.
&
Bolund
,
L.
(
1985
).
DNA synthesis rate changes during the S phase in mouse epidermis
.
Cell Tiss. Kinet
.
18
,
445
455
.
Clausen
,
O. P. F.
&
Potten
,
C. S.
(
1988
).
Heterogeneity in epidermal cell proliferation
.
J. Cutaneous Path, (in press)
.
Dipple
,
A.
,
Mosch
,
R. C.
&
Bigger
,
C. A. H.
(
1984
).
Polynuclear aromatic carcinogens
.
In Chemical Carcinogens, vol. 1, ACS Monograph 182
(ed.
C. E.
Sealer
), pp.
41
163
.
Washington, DC
:
American Chemical Society
.
Eisinger
,
M.
,
Lee
,
J. S.
,
HefTon
,
J. M.
,
Darzynkiewicz
,
Z.
,
Chiao
,
J. W.
&
De Harven
,
E.
(
1979
).
Human epidermal cell cultures. Growth and differentiation in the absence of dermal components or medium supplements
.
Proc. natn. Acad. Sci. U.S.A
.
76
,
5340
5344
.
Harrison
,
D. E.
(
1973
).
Normal production of erythrocytes by mouse bone marrow continues for 73 months
.
Proc. natn. Acad. Sci. U.S A
.
70
,
3184
3190
.
Harrison
,
D. E.
&
Astle
,
C. M.
(
1982
).
Loss of stem cell repopulating ability upon transplantation
.
J. exp. Med
.
156
,
1767
1779
.
Hendry
,
J. H.
(
1979
).
A new derivation, from split-dose data, of the complete survival curve for clonogenic normal cells in vivo
.
Rad. Res
.
78
,
404
414
.
Hendry
,
J. H.
&
Potten
,
C. S.
(
1974
).
Cryptogenic cells and proliferative cells in intestinal epithelium
.
Int. J. Rad. Biol
.
25
,
583
588
.
Hennings
,
H.
&
Boutwell
,
R. K.
(
1970
).
Studies on the mechanisms of Skin Tumor Promotion
.
Cancer Res
.
30
,
312
320
.
Hennings
,
H.
&
Yuspa
,
S. H.
(
1985
).
Two-stage promotion in mouse skin: an alternative interpretation
.
J. natn. Cancer. Inst
.
74
,
735
740
.
Hume
,
W. J.
&
Potten
,
C. S.
(
1976
).
The ordered columnar structure of mouse filiform papillae
.
J. Cell Sci
.
22
,
149
160
.
Ijiri
,
K.
&
Potten
,
C. S.
(
1988
).
Circadian rhythms in the incidence of apoptotic cells and number of clonogenic cells after radiation using normal and reversed light conditions
.
Int. J. Rad. Biol
.
53
,
717
727
.
Iversen
,
O. H.
,
Bjerknes
,
R.
&
Devik
,
F.
(
1968
).
Kinetics of cell renewal, cell migration and cell loss in the mouse hairless dorsal epidermis
.
Cell Tiss. Kinet
.
1
,
351
367
.
Kawamura
,
H.
,
Strickland
,
J. E.
&
Yuspa
,
S. H.
(
1985
).
Association of resistance to terminal differentiation with initiation of carcinogenesis in adult mouse epidermal cells
.
Cancer Res
.
45
,
2748
2757
.
Kimmel
,
M.
,
Darzynkiewicz
,
Z.
&
Staino-Coico
,
L.
(
1986
).
Stathmo-kinetic analysis of human epidermal cells in vitro
.
Cell Tiss. Kinet
.
19
,
289
304
.
Klein-Szanto
,
A. J. P.
&
Slaga
,
T. J.
(
1981
).
Numerical variation of dark cells in normal and chemically induced hyperplastic epidermis with age of animal and efficiency of tumor promoter
.
Cancer Res
.
41
,
4437
4440
.
Kulesz-Martin
,
M. F.
,
Koehler
,
B.
,
Hennings
,
H.
&
Yuspa
,
S. H.
(
1980
).
Quantitative assay for carcinogen altered differentiation in mouse epidermal cells in culture
.
Carcinogenesis
1
,
995
1006
.
Lajtha
,
L. G.
(
1979
).
Stem cell concepts
.
Differentiation
14
,
23
24
.
Lavker
,
R. M.
&
Sun
,
T. T.
(
1982
).
Heterogeneity in epidermal basal keratinocytes: Morphological and functional correlations
.
Science
215
,
1239
1241
.
Leblond
,
C. P.
(
1964
).
Classification of cell populations on the basis of their proliferative behaviour
.
Nat. Cancer Inst. Monogr
.
14
,
119
150
.
Loeffler
,
M.
,
Potten
,
C. S.
&
Wichmann
,
H. E.
(
1987
).
Epidermal cell proliferation. II. A comprehensive mathematical model of cell proliferation and migration in the basal layer predicts some unusual properties of epidermal stem cells
.
Virchows Arch. B. Zellpath
.
53
,
286
300
.
Loeffler
,
M.
,
Stein
,
R.
,
Wichmann
,
H. E.
,
Potten
,
C. S.
,
Kaur
,
P.
&
Chwalinski
,
S.
(
1986
).
Intestinal cell proliferation I. A comprehensive model of steady state proliferation in the crypt
.
Cell Tiss. Kinet
.
19
,
627
646
.
Loehrke
,
H.
,
Schweizer
,
J.
,
Dederer
,
E.
,
Hesse
,
B.
,
Rosenkranz
,
G.
&
Goeritler
,
K.
(
1983
).
On the persistence of tumor initiation in two-stage carcinogenesis on mouse skin
.
Carcinogenesis
4
,
771
775
.
Mackenzie
,
I. C.
(
1970
).
Relationship between mitosis and the structure of the stratum corneum in mouse epidermis
.
Nature, Land
.
226
,
653
655
.
Mackenzie
,
I. C.
&
Bickenbach
,
J.
(
1985
).
Label-retaining keratinocytes and Langerhans cells in mouse epithelia
.
Cell Tiss. Res
.
242
,
551
556
.
Miller
,
E. C.
&
Miller
,
J. A.
(
1981
).
Searches for ultimate chemical carcinogens and their reactions with cellular macromolecules
.
Cancer (Phila)
.
47
,
2327
2345
.
Morris
,
R. J.
,
Fischer
,
S. M.
&
Slaga
,
T. J.
(
1985
).
Evidence that the centrally and peripherally located cells in the murine epidermal proliferative unit are two distinct cell populations
.
J. invest. Derm
.
84
,
277
281
.
Morris
,
R. J.
,
Fischer
,
S. M.
&
Slaga
,
T. J.
(
1986
).
Evidence that a slowly cycling subpopulation of adult murine epidermal cells retains carcinogen
.
Cancer Res
.
46
,
3061
3066
.
Nakai
,
J.
&
Shubik
,
P.
(
1964
).
Autoradiographic localization of tissue-bound tritiated 7,12dimethylbenz[a] anthracene in mouse skin 24 and 48 hours after a single application
.
J. natn. Cancer Inst
.
33
,
887
891
.
Nakayama
,
J.
,
Yuspa
,
S. H.
&
Poirer
,
M. C.
(
1984
).
Benzo(a)pyrene-DNA adduct formation and removal in mouse epidermis in vivo and in vitro-, relationship of DNA binding to initiation of skin carcinogenesis
.
Cancer Res
.
44
,
4048
4095
.
Osborn
,
M. R.
(
1984
).
DNA interactions of reactive intermediates derived from carcinogens
.
In Chemical Carcinogens, vol. 1, ACS Monograph 182
(ed.
C. E.
Searle
), pp.
485
524
.
Washington DC
:
American Chemical Society
.
Parkinson
,
E. K.
,
Grabham
,
F.
&
Emmerson
,
A.
(
1983
).
A subpopulation of cultured human keratinocytes which is resistant to the induction of terminal differentiation-related changes by phorbol 12-myristate 13-acetate: Evidence for an increase in the resistant population following transformation
.
Carcinogenesis
4
,
857
861
.
Parkinson
,
E. K.
,
Pera
,
M. F.
,
Emmerson
,
A.
&
Gorman
,
P. A.
(
1984
).
Differential effects of complete and second-stage tumour promoters in normal but not transformed human and mouse keratinocytes
.
Carcinogenesis
5
,
1071
1077
.
Pelling
,
J. C.
,
Fischer
,
S. M.
,
Neades
,
R.
,
Strawhecker
,
J.
&
Schweickert
,
L.
(
1987
).
Elevated expression and point mutation of the Ha-ras proto-oncogene in mouse skin tumours promoted with benzoyl peroxide and other promoting agents
.
Carcinogenesis
8
,
1481
1484
.
Ponder
,
B. A. J.
,
Schmidt
,
G. H.
,
Wilkinson
,
M. M.
,
Wood
,
M.
,
Monk
,
M.
&
Reid
,
A.
(
1985
).
Derivation of mouse intestinal crypts from single progenitor cells
.
Nature, Land
.
313
,
689
691
.
Potten
,
C. S.
(
1974
).
The epidermal proliferative unit: the possible role of the central basal cell
.
Cell Tiss. Kinet
.
7
,
77
88
.
Potten
,
C. S.
(
1977
).
Extreme sensitivity of some intestinal crypt cells to X and γ irradiation
.
Nature, Land
.
269
,
518
521
.
Potten
,
C. S.
(
1981
).
Cell replacement in epidermis (keratopoiesis) via discrete units of proliferation
.
Int. Rev. Cytol
.
69
,
271
318
.
Potten
,
C. S.
(
1983
).
Stem cells in epidermis from the back of the mouse
.
In Stem Cells: Their Identification and Characterization
(ed.
C. S.
Potten
), pp.
200
232
.
New York
:
Churchill-Livingstone
.
Potten
,
C. S.
(
1984
).
Clonogenic, stem and carcinogen-target cells in small intestine
.
Scand. J. Gastroenterol
.
19
,
suppl. 109
,
3
14
.
Potten
,
C. S.
(
1985
).
Radiation and Skin
.
London
:
Taylor-Francis
.
Potten
,
C. S.
(
1986
).
Cell cycles in cell hierarchies
.
Int. J. Rad. Biol
.
49
,
257
278
.
Potten
,
C. S.
&
Hendry
,
J. H.
(
1973
).
Clonogenic cells and stem cells in epidermis
.
Int. J. Rad. Biol
.
24
,
537
540
.
Potten
,
C. S.
&
Hendry
,
J. H.
(
1975
).
Differential regeneration of intestinal proliferative cells and cryptogenic cells after irradiation
.
Int. J. Rad. Biol
.
27
,
413
424
.
Potten
,
C. S.
,
Hendry
,
J. H.
&
Moore
,
J. V.
(
1987
).
New estimates of the number of clonogenic cells in crypts of murine small intestine
.
Virchows. Arch. B. Zellpath
.
53
,
227
234
.
Potten
,
C. S.
&
Loeffler
,
M.
(
1987
).
A comprehensive model of the crypts of the small intestine of the mouse provides insight into the mechanisms of cell migration and the proliferative hierarchy. J
.
theor. Biol
.
127
,
381
391
.
Potten
,
C. S.
,
Taylor
,
Y.
&
Hendry
,
J. H.
(
1988
).
The doubling time of regenerating clonogenic cells in the crypts of the small intestine
.
Int. J. Rad. Biol, (in press)
.
Potten
,
C. S.
,
Wichmann
,
H. E.
,
Loeffler
,
M.
,
Dobek
,
K.
&
Major
,
D.
(
1982
).
Evidence for discrete cell kinetic sub-populations in mouse epidermis based on mathematical analysis
.
Cell Tiss. Kinet
.
15
,
305
329
.
Randerath
,
E.
,
Agrawal
,
H. P.
,
Reddy
,
M. V.
&
Randerath
,
K.
(
1983
).
Highly persistent polycyclic aromatic hydrocarbon-DNA adducts in mouse skin: Detection by 332-postlabeling analysis
.
Cancer Lett
.
20
,
104
114
.
Reddy
,
M. V.
,
Gupta
,
R. C.
,
Randerath
,
E.
&
Randerath
,
K.
(
1983
).
32P-postlabeling test for covalent DNA binding of chemicals in vivo: Application to a variety of aromatic carcinogens and methylating agents
.
Carcinogenesis
5
,
231
243
.
Roe
,
F. J. C.
,
Carter
,
R. N.
,
Mitchley
,
B. C. V.
,
Peto
,
R.
&
Hecker
,
E.
(
1972
).
On the persistence of tumor initiation and the acceleration of tumor progression in mouse skin carcinogenesis
.
Int. J. Cancer
9
,
624
673
.
Schmidt
,
G. H.
,
Wilkinson
,
M. M.
&
Ponder
,
B. A. J.
(
1985
).
Cell migration pathway in the intestinal epithelieum: An in situ marker system using aggregation chimeras
.
Cell
40
,
425
429
.
Schofield
,
R.
(
1978
).
The relationship between the spleen colony-forming cell and the haemopoietic stem cell
.
Blood Cells
4
,
7
25
.
Scribner
,
J. D.
&
Suss
,
R.
(
1978
).
Tumor initiation and promotion
.
Int. Rev. exp. Path
.
18
,
137
198
.
Slaga
,
T. J.
(
1984
).
Mechanisms involved in two-stage carcinogenesis in mouse skin
.
In Mechanisms of Tumor Promotion
, vol.
2
(ed.
T. J.
Slaga
), pp.
1
16
.
Boca Raton, FL
:
CRC Press Inc
.
Slaga
,
T. J.
&
Klein-Szanto
,
A. J. P.
(
1983
).
Initiation-promotion versus complete carcinogenesis in mice: Importance of dark basal keratinocytes (stem cells)
.
Cancer Invest
.
1
(
5
),
425
436
.
Solun
,
S.
(
1970
).
High-resolution autoradiographic localization of 3,4-benzo(a)pyurene-3H in mouse skin
.
Cancer Res
.
30
,
1123
1128
.
Staiano-Coico
,
L.
,
Higgins
,
P. J.
,
Darzynkiewicz
,
Z.
,
Kimmel
,
M.
,
Gottlieb
,
A. B.
,
PaganCherry
,
L
,
Madden
,
M. R.
,
Finkelstein
,
J. L.
&
Hefton
,
J. M.
(
1986
).
Human keratinocyte culture. Identification and staging of epidermal cell subpopulations
.
J. clin. Invest
.
7
,
396
404
.
Stenback
,
F.
,
Peto
,
R.
&
Shubik
,
P.
(
1981
).
Initiation and promotion at different stages and doses in 2200 mice. I. Methods and the apparent persistence of initiated cells
.
Br.J. Cancer
44
,
1
14
.
Tarnowski
,
W. M.
(
1970
).
Autoradiographic localization of tritiated 7,12-dimethylbenz[a]anthracene in mast cells of hairless mouse skin
.
Cancer Res
.
30
,
1163
1167
.
Van Durren
,
B. L.
,
Sivak
,
A.
,
Katz
,
C.
,
Deidman
,
I.
&
Melchionne
,
S.
(
1975
).
The effect of aging and interval delay between primary and secondary treatment in two-stage carcinogenesis on mouse skin
.
Cancer Res
.
35
,
502
505
.
Withers
,
H. R.
&
Elkind
,
M. M.
(
1969
).
Radiosensitivity and fractionation response of crypt cells of mouse jejunum
.
Rad. Res
.
38
,
598
613
.
Withers
,
H. R.
&
Elkind
,
M. M.
(
1970
).
Microcolony survival assay for cells of mouse intestinal mucosa exposed to radiation
.
Int. J. Rad. Biol
.
17
,
261
268
.