A quantitative analysis of the human bone marrow granulocytic cell lineage using the SAMBA 200 cell image processor. I. The normal maturation sequence

A quantitative image analysis of the human normal bone marrow granulocytic line was performed using the SAMBA 200 image analyzer. The steps of image acquisition, preprocessing, segmentation and parametrization are described. Forty-one parameters were computed on 941 cell images belonging to the various maturation stages. The automated classification of these cells based upon a stepwise linear discriminant analysis resulted in 77% correctly classified cells; the five most discriminating parameters were the nuclear area, the nuclear convexity degree, the average cytoplasmic hue, the regularity of the nuclear boundary and the average cytoplasmic luminance. The evolution of the parameters correlates well with the cytologic evolution and the biochemical and functional events during the maturation process. It can be inferred from our results that the maturation sequence can be subdivided into two phases according to the evolution of the cell profiles. The first phase, from myeloblast to myelocyte, is discontinuous and appears as the critical point with regard to the expression of genes. The second phase, from myelocyte to polymorphonuclear cell, is a continuous sequence of transformations leading to the functional granulocyte.     

Terminal differentiation of PWM-stimulated human B lymphocytes.

Patterns of surface and cytoplasmic immunoglobulins were simultaneously studied on human B blast cells induced by pokeweed stimulation of peripheral blood lymphocytes. A double-staining immunofluorescent technique was used. After 4 and 7 days of culture, a gradual loss of surface IgD was observed on blast cells whereas numbers of plasmablasts with cytoplasmic immunoglobulin showed a marked increase. After 7 days, 92% of surface IgA positive blasts had passed terminal differentiation to cytoplasmic IgA-producing plasma-blasts. At the same time, 73% of surface IgM positive blasts were found to contain cytoplasmic IgM, and 30% of surface IgG positive cells had cytoplasmic IgG. Only a small fraction of blast cells with surface IgD was able to mature to IgD producing plasmablasts. In general, the class of surface and cytoplasmic immunoglobulin coincided in single blast cells, with the exception of surface IgD which was present on 10% of the cytoplasmic IgM-containing blasts.     

A quantitative analysis of the human bone marrow erythroblastic cell lineage using the SAMBA 200 cell image processor. I. The normal maturation sequence

A quantitative image analysis of the normal maturation sequence for the human bone marrow erythroblastic lineage was performed using the SAMBA 200 cell image processor. The different image analysis steps (image acquisition, preprocessing, segmentation, parametrization and data analysis) are briefly described. Thirty-three parameters related to geometry, color, texture and densitometry were computed on 638 cell images belonging to the five erythroblastic maturation stages. The automated classification of these cells, based upon a stepwise linear discriminant analysis, resulted in 80% correctly classified cells. Acceptance of confusions between successive maturation stages enhanced the rate of correctly classified cells to 100%. Among the ten most discriminating parameters, the nuclear area showed the highest correlation with the changes throughout the maturation process. The projection of the maturation sequence onto the factorial plane resulting from the canonical analysis emphasizes the existence of three phases of the maturation process, a finding that correlates well with the cytologic evolution and the biochemical and functional events during the maturation. The trajectory of cells within this factorial plane is thus regarded as a differentiation path from which a measure of the maturation could be derived.     

The cytoplasmic expression of CD3 antigens in normal and malignant cells of the T lymphoid lineage

Anti-CD3 (T3) Ab reacting with different proportions of thymocytes (anti-CD3a: UCHT1, anti-CD3b: T10B9, and anti-CD3c: OKT3) were tested for cytoplasmic (cCD3) and membrane (mCD3) expression in the bone marrow, thymus, and blood in man and selected primates. The expression of cCD3a and cCD3c in the perinuclear and Golgi area of large, BrdU-incorporating, strongly TdT+ thymic blasts probably represents one of the earliest signs of T cell commitment, because these blast cells are CD1-, CD4-, CD8-, and mCD3-. The cCD3+, TdT+ cells are normally restricted to the thymus and are absent among the TdT+ cells of bone marrow. The anti-CD3b Ab used, T10B9, co-caps and co-modulates with the other anti-CD3 Ab and is a T cell-specific reagent at a membrane level but does not bind to perinuclear cCD3. Instead, this reagent cross-reacts with a filamentous cytoplasmic network in non-T cells in man and in primates S. oedipus and M. rhesus despite their T cell negativity for mCD3. The characteristics of all T-ALL cases studied: cCD3+, CD7+ along with nuclear TdT+ suggest lineage fidelity to early thymic blasts. As a marked contrast, cCD3 is absent in common ALL and in AML, including cases that concomitantly express CD7 and myeloid antigens. Thus, the cCD3, TdT combination provides a very sensitive assay for residual T-ALL blasts outside the normal thymus.     

Morphologic mast cell cycles

During ultrastructural studies of the early kinetics of anaphylactic degranulation and early and late recovery intervals from these release reactions induced in isolated, partially purified, cultured human lung mast cells stimulated to release histamine and granules by exposure to anti-IgE, we noted the ability of mast cells to undergo morphologic cycles. Following release of most cytoplasmic granules and shedding of large amounts of membranes and cellular processes, small, immature mast cells, nearly devoid of granules, were seen. These lymphocyte-like cells had condensed nuclear chromatin and large nucleoli. Blast transformation of these small cells and expansion of synthetic machinery-laden cytoplasm resulted in large, immature mast cells with small numbers of small, immature progranules and large numbers of lipid bodies in their cytoplasm. This cycle of mast cell morphologic change has considerable similarities to, as well as some differences from, lymphocyte morphologic cycles.     

Image morphometry of acute leukemias. Comparison between lymphoid and myeloid subtypes

OBJECTIVE: To determine if image morphometry has any role in distinguishing blasts of acute lymphoblastic leukemia (ALL-L2) from those of acute myeloid leukemia (AML-M1) and (AML-M2). STUDY DESIGN: Ten cases each of ALL-L2, AML-M1 and AML-M2 diagnosed according to the French-American-British criteria were studied. In all cases May-Grünwald-Giemsa-stained bone marrow aspiration smears were obtained. At least 100 blast cells from each case were subjected to analysis randomly with an image cytometer using Leica Quantimet 600 software (Cambridge, U.K.). The area, convex area, length, width, perimeter, convex perimeter, roundness, total optical density, average optical density and pixel grey value variance of the nuclei were measured by random selection of cells using a 40:1 objective (1 pixel = 0.446 micron). RESULTS: Mann Whitney's nonparametric test showed that there was considerable overlap of morphometric variables between the 3 subtypes. Though statistical significance was found in "roundness" between blasts of AML-M1 and ALL-L2, power analyses (sample size of 100 blasts of each subtype) did not show sufficient power for this variable. However, between blasts of ALL-L2 and AML-M2, "average optical density" and "pixel grey value variance" were statistically significant with full power using power analyses. CONCLUSION: Image morphometry may be helpful in differentiating blasts from lymphoid and myeloid leukemic subtypes.     

Cytoplasmic changes in relation to nuclear maturation and early embryo developmental potential of porcine oocytes: effects of gonadotropins, cumulus cells, follicular size, and protein synthesis inhibition

Morphological and biochemical changes indicative of cytoplasmic maturation in relation to nuclear maturation progression and early embryo developmental potential was studied. Fluorescently labeled microfilaments and cortical granules were visualized by using laser scanning confocal microscopy. The mitogen-activated protein (MAP) kinase phosphorylation and cyclin B1 levels were revealed by Western blot. With the maturation of oocytes, cortical granules and microfilaments were localized at the cell cortex. A cortical granule-free domain (CGFD) and an actin-thickening area were observed over both the MII spindle of a mature oocyte and chromosomes of a nocodazole-treated oocyte, suggesting that chromosomes, but not the spindle, determined the localization of CGFD and actin-thickening area. In oocytes that are incompetent to resume meiosis, as indicated by the failure of germinal vesicle breakdown (GVBD), peripheral localization of cortical granules and microfilaments, phosphorylation of MAP kinase and synthesis of cyclin B1 did not occur after 44 hr in vitro. These cytoplasmic changes were also blocked when GVBD of meiotically competent oocytes was inhibited by cycloheximide. Culture of oocytes in a chemically defined medium showed that biological factors such as gonadotropins, cumulus cells and follicle size affected both nuclear and cytoplasmic maturation as well as embryo developmental potential. Absence of gonadotropins or removal of cumulus cells alone did not significantly influence GVBD or cyclin B1 levels, but decreased the final maturation and developmental ability of oocytes. A combination of gonadotropin absence and cumulus removal decreased GVBD, MAP kinase phosphorylation and embryo development. A high proportion of oocytes derived from small follicles were able to resume meiosis, synthesize cyclin B(1), phosphorylate MAP kinase and translocate CGs, but their maturation and embryo developmental ability were limited. Removal of cumulus cells from small follicle-derived oocytes severely affected their ability to undergo cytoplasmic and nuclear maturation.     

Detection and characterization of a B cell stimulatory factor (BSF-TC) derived from a bone marrow stromal cell line

Stromal cell lines derived from murine bone marrow support the growth of immature pre-B cells and produce cytokines that affect the growth and differentiation of other hematopoietic precursors. Conditioned medium (CM) from one such line (TC-1) stimulated marked proliferation of B cells previously activated by anti-Ig (anti-Ig blasts). Proliferation of anti-Ig blasts was not induced by purified cytokines known to be produced by TC-1 (CSF-1, GM-CSF, or G-CSF) or by IL-1, IL-2, IL-3, IL-4, IL-5, or IL-6. Furthermore, IL-2, IL-4, and IL-5, alone or in combination, failed to support proliferation or differentiation of anti-Ig blasts. TC-1 CM enhanced proliferation of B cells that were co-cultured with LPS, anti-Ig, or dextran sulfate; co-stimulation with anti-Ig was unaffected by the presence of monoclonal anti-IL-4. Proliferation of low, but not high, density B cells isolated from spleen was directly stimulated by TC-1 CM. These results suggest that bone marrow stromal cells produce a novel B cell stimulatory factor (BSF-TC) that induces proliferation of activated B cells.     

The stem cell antigens Sca-1 and Sca-2 subdivide thymic and peripheral T lymphocytes into unique subsets

Stem cell Ag 1 and 2 (Sca-1 and Sca-2), so named due to their expression by mouse bone marrow stem cells, were evaluated for expression by populations of cells within the thymus. Immunohistochemical analysis demonstrated that Sca-1 was expressed by cells in the thymic medulla and by some subcapsular blast cells, as well as by the thymic blood vessels and capsule. Sca-2 expression, which was limited to the thymic cortex, could be associated with large cycling thymic blast cells. Both Sca-1 and Sca-2 were expressed on a sub-population of CD4-CD8- thymocytes, and this subpopulation was entirely contained within the Ly-1lo progenitor fraction of cells. Sca-1 expression by a phenotypically mature subset of CD4+CD8- thymocytes was also noted. Conversely, Sca-2 expression was observed on a phenotypically immature or nonmature subpopulation of CD4-CD8- thymocytes. MEL-14, an antibody that defines functional expression of a lymphocyte homing molecule, identified a small population of thymocytes that contained all four major thymic subsets. Sca-2 split the MEL-14hi thymocyte subset into two Sca-2+ non-mature/immature phenotype fractions and two Sca-2- mature phenotype fractions. In peripheral lymphoid organs, Sca-1 identified a sub-population of mature T lymphocytes that is predominantly CD4+CD8-, in agreement with the thymic distribution of Sca-1. Peripheral T cells of the CD4-CD8+ phenotype were predominantly Sca-1-. In contrast, Sca-2 did not appear to stain peripheral T lymphocytes, but recognized only a subset of B lymphocytes which could be localized by immunohistochemistry to germinal centers. Thus, expression of Sca-1 is observed throughout T cell ontogeny, whereas Sca-2 is expressed by some subsets of thymocytes, including at least one half of thymic blasts, but not by mature peripheral T lymphocytes.     

Morphological heterogeneity of secretory granules of rat Clara cells: an immunocytochemical study.

The secretory granules of rat bronchiolar Clara cells were classified into different types by their ultrastructural appearances followed by immunocytochemistry using anti-rat 10 kDa Clara cell-specific protein (10 kDa CCSP) antibody. One predominant type was the oval to round granule (type A granule), of which the matrix was composed of a map-like mixture of electron-dense and less electron-dense material. Another predominant type was the rod-shaped granule (type B granule). The content of type B granules varied from a finely fibrillar (type B1 granule) to an electron-dense, rod-like (type B3 granule) structure. Various intermediate types (type B2 granule) between type B1 and B3 granules were also found. Small cytoplasmic vesicles were found occasionally in close proximity to type B2 or B3 granule. Another type of granule (type C granule) was large, up to 8 microns in diameter, and contained a moderately electron-dense amorphous matrix. Both type A and C granules stained at a similar density with the antibody. The nascent form of type A granules, which was found in the vicinity to the trans face of the Golgi apparatus, was also labeled. On the other hand, the labeling density of type B granules varied: type B1 granules were almost devoid of immunolabeling, whereas type B3 granules were intensely labeled. Type B2 granules stained with the antibody; however, the labeling density was less than that of type B3 granules. The small cytoplasmic vesicles of type B2 granules were labeled.(ABSTRACT TRUNCATED AT 250 WORDS)     

Possible sequential transformation of a series of acidophils in the pituitary autografts in the renal capsule of male rats in association with ACTH cell.

In our electron microscopy, acidiphils in the pituitary autografts placed in the renal capsules of immature male rats underwent a sequential transformation with the lapse of time: Within 3 and 6 days, all the somaotrophs packed with the large granules of about 350 mmu diameter dispersed. The size and number of the granules in somatotrophs were quickly and markedly reduced with severe modification of cell shape. There was evidence during this time course that Siperstein's or Moriarty's corticotrophs might be synonymous with the stellate shape of acidophils with the arrangement of small granules 150-200 mmu in diameter along the cell acidophils. The "acidophils of the small granule type" possibly related to ACTH production according to Yoshimura et al. (1974) were frequently detected in the grafts as elongated or irregularly shaped cells. Their minute granules 100-150 mmu in diameter were also distributed in row in the cytoplasmic peripheral area. Gradual loss of the minute granules below 100 mmu in diameter eventually made the acidophils to transform into agranular cells. Our own idea that ACTH secretion might correlate with a series of cells transforming along the acidophil-axis was indirectly supported by the present observation on pituitary grafts. On the other hand, basophils rapidly degenerated and died away. Ten and 20 days after autografting, the graft cells which might be principally composed of the cells of acidophil origin enormously proliferated through mitotic division, showing the homologous fine structure, without the normal cell individuality. They always contained three different size and shape of granules simulataneously. Significance of such a rapid and strong response of acidophils to the ectopic replacement in the immature male rats was discussed from the view-point of hypothalamic regulation to simple protein hormones.     

Loss of Rna and Protein, And Changes in Dna During A 30-Hour Cold Perchloric Acid Extraction of Cultured Cells

KB cells derived from human carcinoma were fixed in acetic-alcohol (1:3) and extracted with 10% perchloric acid (PCA) at 4 C for 1, 3, 6, 9, 12, 24 and 30 hr. Cells were then washed in water and stained for nucleic acids, proteins, polysaccharides, and lipids. Control cells were kept in water for 30 hr prior to staining. Acridine orange (AO) fluorochroming revealed color changes in residual cytoplasmic and nucleolar RNA as well as DNA during extraction--interpreted as indicative of molecular alterations. All nucleic acid stains (AO, gallocyanin chromalum, and azure B bromide) demonstrated a differential extraction of RNA, with cytoplasmic RNA being removed in about 6 hr and nucleolar RNA requiring 6 more hours for complete extraction. Large granules appeared early in nuclei. These were positive for DNA by azure B, gallocyanin chromalum, Feulgen, and fluorescent-Feulgen. These same granules stained for protein by mercuric bromphenol blue and alkaline Biebrich scarlet. At 24 hr, there was visual and Feulgen-cytophotometric evidence for a slight loss of DNA, which may amount to 10-20%. There was a progressive loss of cytoplasmic and nuclear but not nucleolar protein during PCA treatment. Concurrently, large protein-positive granules appeared in the cytoplasm. Apparently, PCA treatment in combination with an aqueous wash was responsible for some protein loss. Glycogen was gradually lost (fluorescent PAS) and redistributed in cells. Lipids were unaffected (Sudan black B).     

Visualizing DNA replication sites in the cell nucleus.

Studies of DNA replication associated with the nuclear matrix have led to a radically new view of replication at the macroscopic level. It is proposed that individual replicons and their associated replicational assemblies (replisomes) are clustered together during active replication by attachment to the nuclear matrix at special sites termed 'clustersomes'. Direct visualization of replication sites in permeabilized cells by fluorescence microscopy following biotin-11-dUTP incorporation provides support for this model. Discrete replication granules are observed with sizes and numbers consistent with each granule being a site of replicon cluster synthesis. Distinct patterns of these sites are seen in different periods of S-phase. Both the individual granules and their early and late S-phase dependent patterns are strikingly maintained following extraction of the cells for in situ nuclear matrix structures. Similar results were obtained when probing in vivo sites of replication following incorporation of 5-bromodeoxyuridine. The three-dimensional organization of these replicational granules (clustersomes) is studied using confocal light microscopy and an appropriate multidimensional image analysis system.     

Leukemia-related morphological features in blast cells

This paper investigates the use of image-processing methods to detect leukemia-related morphological differences in mononuclear blast cells. Routinely prepared Pappenheim-stained blood smears were scanned in a high-resolution color TV-microscope system. Eleven blast-cell classes (OMSBC, T-ALL, OMS, ALL, LBL, IBL, AUL, AML, AMOL, AMMOL, and CML) were analyzed with the nonparametric statistical software program "Classification and Regression Trees" (CART). This paper documents the initial statistical evaluation of 62 leukemia-related morphological features that directly measure and analyze the cell-related quantifiable differences occurring in the various blast cells. The 62 cell image features include both common cytophotometric features, and new texture and color features developed for this project. This study found that each leukemia specimen contains a dominant class of blasts that correlates with the specific leukemia, plus a distribution of blasts from related diseases. The present data suggest the existence of a distribution fingerprint pattern for each leukemia.     

Two color light scattering identifies physical differences between lymphocyte subpopulations

Forward angle light scattering of two different wavelengths by cells in a flow cytometer was used to investigate physical differences between lymphocytes of different lineage, functional subclass and developmental stage. Correlation of the ultraviolet (UV: 351 nm and 364 nm) and 488 nm light scattering signals produced by lymphoid cells demonstrated that the two signals were not equivalent and that they placed different emphasis on the physical parameters characterizing lymphocytes. Both small T and B lymphocytes from peripheral lymphoid tissues and mitogenically activated large T and B lymphocyte blasts were discriminated by both wavelengths. Differences between the Lyt-2 negative and Lyt-2 positive T lymphocyte subsets were also apparent. Two color light scattering could also discriminate between immature thymocytes and mature peripheral T cells and between small bone marrow cells and mature peripheral B cells. In bone marrow an increase in UV light scattering coincided with the appearance of cell surface immunoglobulin on small cells. These data establish that two color light scattering is a sensitive probe for distinguishing cells of apparently similar morphology and that it can be used to study the physical changes that occur during lymphoid cell differentiation.     

In vitro apoptotic cell death during erythroid differentiation

Erythropoiesis occurs in bone marrow and it has been shown that during in vivo erythroid differentiation some immature erythroblasts undergo apoptosis. In this regard, it is known that immature erythroblasts are FasL- and TRAIL-sensitive and can be killed by cells expressing these ligand molecules. In the present study, we have investigated the cell death phenomenon that occurs during a common unilineage model of erythroid development. Purified CD34+ human haemopoietic progenitors were cultured in vitro in the presence of SCF, IL-3 and erythropoietin. Their differentiation stages and apoptosis were followed by multiple technical approaches. Flow cytometric evaluation of surface and intracellular molecules revealed that glycophorin A appeared at day 3-4 of incubation and about 75% of viable cells co-expressed high density glycophorin A (Gly(bright)) and adult haemoglobin at day 14 of culture, indicating that this system reasonably recapitulates in vivo normal erythropoiesis. Interestingly, when mature (Gly(bright)) erythroid cells reached their higher percentages (day 14) almost half of cultured cells were apoptotic. Morphological studies indicated that the majority of dead cells contained cytoplasmic granular material typical of basophilic stage, and DNA analysis by flow cytometry and TUNEL reaction revealed nuclear fragmentation. These observations indicate that in vitro unilineage erythroid differentiation, as in vivo, is associated with apoptotic cell death of cells with characteristics of basophilic erythroblasts. We suggest that the interactions between different death receptors on immature basophilic erythroblasts with their ligands on more mature erythroblasts may contribute to induce apoptosis in vitro.     

Morphological heterogeneity of secretory granules of rat Clara cells: an immunocytochemical study

Summary The secretory granules of rat bronchiolar Clara cells were classified into different types by their ultrastructural appearances followed by immunocytochemistry using anti-rat 10 kDa Clara cell-specific protein (10 kDa CCSP) antibody. One predominant type was the oval to round granule (type A granule), of which the matrix was composed of a map-like mixture of electron-dense and less electron-dense material. Another predominant type was the rod-shaped granule (type B granule). The content of type B granules varied from a finely fibrillar (type B1 granule) to an electron-dense, rod-like (type B3 granule) structure. Various intermediate types (type B2 granule) between type B1 and B3 granules were also found. Small cytoplasmic vesicles were found occasionally in close proximity to type B2 or B3 granule. Another type of granule (type C granule) was large, up to 8 m in diameter, and contained a moderately electron-dense amorphous matrix. Both type A and C granules stained at a similar density with the antibody. The nascent form of type A granules, which was found in the vicinity to the trans face of the Golgi apparatus, was also labeled. On the other hand, the labeling density of type B granules varied: type B1 granules were almost devoid of immunolabeling, whereas type B3 granules were intensely labeled. Type B2 granules stained with the antibody; however, the labeling density was less than that of type B3 granules. The small cytoplasmic vesicles of type B2 granules were labeled. From these findings, it is suggested that the granules of rat Clara cells consist of two types of granules of distinct origin; one appears to derive from condensing vacuoles of Golgi origin, whereas the other may be formed by membranefusions with small cytoplasmic vesicles of unknown source.     

Translation and granule localization of mouse mast cell protease-5. Immunodetection with specific antipeptide Ig.

An antibody to mouse mast cell protease-5 (MMCP-5) was obtained by immunizing a rabbit with a 17-residue synthetic peptide corresponding to the unique amino acid sequence at residues 146 to 162 in this serine protease. After affinity purification, anti-MMCP-5(146-162) Ig reacted in SDS-PAGE immunoblots to recombinant MMCP-5 and to the native MMCP-5 protein present in the lysates of mouse serosal mast cells and the MC5 line of Kirsten sarcoma virus-immortalized mouse mast cells. Immunocytochemical staining localized MMCP-5 to the cytoplasmic granules of serosal mast cells and Kirsten sarcoma virus-immortalized mouse mast cells. Because mouse bone marrow-derived mast cells express abundant amounts of MMCP-5 mRNA, anti-MMCP-5(146-162) Ig was used to study the translation and granule accumulation of this protease when progenitor cells differentiate into these immature mouse mast cells. Maximal expression of MMCP-5 mRNA occurred after bone marrow cells had been cultured for 2 wk in IL-3-rich WEHI-3 cell conditioned medium, and MMCP-5 protein was detected in these cells. However, electron-microscopic analysis with gold-labeled antibody revealed that the amount of MMCP-5 in the individual granules of bone marrow-derived mast cells varied. The highest concentration of MMCP-5 was found in the most electron-dense secretory granules of the cells. These studies demonstrate the ultrastructural localization of the earliest transcribed mouse mast cell chymase, MMCP-5, and its granule accumulation during the differentiation of mouse bone marrow progenitor cells into immature mouse mast cells.     

Immunoglobulin gene organisation and expression in haemopoietic stem cell leukaemia

We have analysed the organisation and expression of mu genes in the granulocytic phase and in the lymphoid and myeloid blast crises of Philadelphia chromosome (Ph1) chronic granulocytic leukaemia (CGL), a leukaemia which is known to arise in multipotential stem cells. We find that mu chain gene rearrangement occurs exclusively in lymphoid blast crisis leading in some, but not all, cases to the synthesis of small amounts of cytoplasmic mu chains characteristic of early pre-B lymphocytes. In Southern blots, only one or two rearranged mu chain genes are seen, suggesting that a clonal event leading to blast crisis can occur in a committed B cell precursor rather than in the multipotential stem cell precursor, in which the Ph1 chromosome originated. The pattern of mu gene rearrangement observed in Ph1 CGL blast crisis is compared with that in normal B cells, other B lineage malignancies, myeloid leukaemias and T cell leukaemias.     

Development of ATPase-positive,immature Langerhans cells in the fetal mouse epidermis and their maturation during the early postnatal period

Summary The development and maturation of Langerhans cells during the differentiation of skin was studied in mice from fetal day 13 to adult using 3 indices: (1) ATPase activity; (2) ultrastructure; and (3) quantitative evaluation of the cell population.ATPase-positive Langerhans cells appeared in the epidermis at first at fetal day 16, and they increased in number in the differentiating epidermis during the late fetal period. The earliest appearance of Birbeck granules was at postnatal day 4. Cored tubules were also formed in the Langerhans cells in the dermis at around the same age. The cells containing Birbeck granules or cored tubules are considered to be mature Langerhans cells. In the Langerhans-cell lineage, those cells in the epidermis at stages earlier than postnatal day 4 and not yet containing specific organelles are considered to be immature Langerhans cells. These immature Langerhans cells can be identified ultrastructurally in the epidermis at fetal day 16, coinciding with the appearance of ATPase-positive cells. The increase in the number of immature Langerhans cells during the perinatal period was shown by quantitative analysis of nuclear density and relative Langerhans-cell area on the electron micrographs.It is concluded that ATPase is a marker of the Langerhans-cell lineage from the early development stages, while Birbeck granules and cored tubules are markers that identify mature Langerhans cells in electron micrographs.