Howard A. Schneiderman

Histological analysis of the dynamics of growth of imaginal discs and histoblast nests during the larval development ofDrosophila melanogaster

Summary1.Histological analyses were made of imaginal discs and histoblasts during the larval development ofDrosophila melanogaster to determine the number of cells, the patterns of cell division and the growth dynamics in these adult primordia. Histological studies were also made of the imaginal rings which are the primordia of the adult salivary gland, fore-and hindgut, the anlage cells of the midgut and several larval and embryonic tissues.2.In the newly-hatched larva, the immature eye-antenna, wing, haltere, leg and genital discs contain about 70, 38, 20, 36–45 and 64 cells respectively. These numbers include cells destined to form cuticular elements as well as peripodial, tracheal and nerve cells and probably the progenitors of adepithelial cells. The number of cells counted in the various imaginal disc anlagen is 1.5 to 4 times higher than the numbers deduced from genetic mosaic analyses by other investigators and reasons for these differences are given.3.About 12 h after fertilization, mitosis ceases in all tissues of the embryo except the nervous system. After the larva hatches, mitosis resumes in most of the imaginal anlagen and in some larval tissues. The time of resumption of mitosis in the imaginal anlagen was determined after treating the larvae with colchicine for 2 h.4.Among the imaginal discs, the eye disc is the first to begin cell division, at about 13–15 h after the hatching of the larva (first instar) followed by the wing (15–17 h), the haltere (18–20 h), the antenna, leg, and genitalia (24–26 h, early second instar), and finally the labial and dorsal prothoracic discs (52–54 h, early third instar). The cell doubling time for various discs was calculated from cell counts and the times agree closely with the doubling times deduced from clonal analyses by other workers: e.g., 7.5 h for the cells of the wing disc.5.The imaginal ring of the hindgut first shows cell division early in the second instar. The imaginal rings of the foregut and salivary glands, the anlage cells of the midgut and the cells of the segmental lateral tracheal branches begin to divide early in the third instar.6.The histoblasts which are the anlagen of the integument of the adult abdomen do not increase in number from the time of larval hatching until about 5 h after pupation when they begin to divide. Their behaviour contrasts with that of the histoblasts of the other dipterans such asCalliphora, Musca andDacus, which begin to divide during the second instar.7.The histoblasts are an integral part of the larval abdominal epidermis and, unlike imaginal disc cells, secrete cuticle during larval life. Each hemisegment consists of an anterior dorsal, a posterior dorsal, and a ventral histoblast nest containing about 13, 6 and 12 cells respectively. The 62 histoblasts in each larval segment represent about 7–8% of the total number of cells that form the integument of that segment.8.The number of cells in a particular type of histoblast nest was constant for both male and female larvae and among the different abdominal segments, except that the anterior dorsal group of the first and the seventh segments contains fewer cells than those of the other segments. Although the male and female adultDrosophila lack the first abdominal sternite and the male lacks the seventh abdominal tergite and sternite, the ventral histoblast nests of the first and the dorsal and ventral nests of the seventh abdominal segments are present in the larval stages as well as in the prepupa and have the same morphology and cell number as similar nests in the rest of the abdominal segments.9.The cells of the imaginal discs increase in volume about six-fold and their nuclei increase in volume three-fold between the time of hatching and the initiation of mitosis. The histoblasts increase in volume about 60-fold and their nuclei increase in volume about 25-fold between larval hatching and pupariation.10.Prior to each cell division, the nuclei of the columnar cells of the disc epithelium and of the histoblasts appear to migrate toward the apical surface of the epithelium. The cells round up and shift toward the apical region where mitosis occurs. After cytokinesis, the daughter cells move back to deeper positions in the epithelium. Because the nuclei of the non-dividing cells continue to lie deep in the epithelium, this intermitotic migration of nuclei gives these epithelia a pseudostratified appearance.11.Analyses of the growth of larval cells and of organs confirmed the observations of earlier investigators that cell division occurs only in a few larval tissues, whereas growth in the rest of the larval tissues is by cell enlargement and polyteny. During larval life, cell division was detected only in the central nervous system, gonads, prothoracic glands, lymph glands and haemocytes. Each tissue began mitosis at a characteristic stage in larval life. The larval cells that did not divide, grew enormously, e.g., epidermal cells increased in volume 150-fold and their nuclei increased in volume 80-fold.12.The adepithelial cells, which give rise to some of the imaginal muscles, were first identified between the thick side of the imaginal dise epithelium and the basement membrane at the beginning of the third larval instar (50–52 h). The origin of these precursors of mesodermal structures was analysed and evidence is presented that the adepithelial cells come from the disc epithelium. The question of the origin of the mesoderm of cyclorrhaphan Diptera is reviewed and it is suggested that the imaginal disc ectoderm may become segregated from the rest of the embryo before gastrulation has occurred, that is before the mesoderm has been established.

Pattern formation and determination in the antenna of the homoeotic mutant Antennapedia of Drosophila melanogaster

Abstract The development of the antenna in the antennal-leg homoeotic mutant Antennapedia (Antp R ) was investigated using somatic crossing-over to mark clones of cells in Antp R antennal appendages. Antp R antennae ranged from a nearly normal antenna to a nearly normal leg. The arrangement of clones of marked bristles and cuticle in the more antennalike antennae was similar to the wild type antenna, and that of the leglike antennae was similar to the wild-type leg. The contiguity of clones argued against extensive individual cell migration. The regions occupied by homoeotic leg varied considerably between different Antp R antennae. Observation of Antp R antennae in these phenotypic mosaics showed that specific leg parts replaced specific antennal parts. Even small groups of leg sensilla appeared only at precise locations in the antenna. These results suggest that homoeotic leg cells and antennal cells can both respond to the same positional information or prepattern. An analysis of clone size provided estimates for cell number in the Antp R antenna. It was found that cell numbers in the wild-type and Antp R antennae are about the same until the third instar, when the Antp R cells start dividing more rapidly than wild type. Previous work had shown that clonal inheritance of a commitment for homoeotic leg also did not occur prior to the early third instar. It is suggested that determination for homoeotic leg occurs in the early third instar, and that thereafter this commitment is inherited by the progeny of the determined cells. The increase in growth rate is probably due to a faster growth rate in cells with a leg commitment than in cells with an antennal commitment. The results suggest that, once initiated, determination may be of two types—a clonally inherited determination (for example, to be homoeotic leg) and an environmental determination (for example, to be a specific part of a homoeotic leg). Clonal inheritance of determination in normal embryonic development and in sex determination in intersexes is discussed.

Developmental capacities of benign and malignant neoplasms ofDrosophila

SummaryThe developmental properties of three neoplasms found in lethal mutants ofDrosophila melanogaster were studied and compared with the development of the corresponding wild-type organs. Two of these neoplasms are found in the late larval lethal mutant,lethal (2) giant larvae4 (l(2)gl4) and its allele,lethal (2) giant larvae (l(2)gl). The third neoplasm occurs in the hemizygous lethal male embryos of the mutantNotch8 (Df(1)N8).The mutantsl(2)gl4 andl(2)gl were shown to possess defects involving imaginal primordia of ectodermal origin, such as the imaginal discs of the adult integument and the imaginal optic primordia in the larval brain. They also have enlarged lymph glands and abnormal gonads, salivary and ring glands. Thel(2)gl4 andl(2)gl mutations transform the imaginal discs into noninvasive, lethal neoplasms and the imaginal optic primordia in the larval brain into an invasive and lethal neuroblastoma. Both neoplasms were serially subcultured in female adult hosts where they grew rapidly and killed their hosts in 7–14 days. The neoplastic development of thel(2)gl4 imaginal discs and brain proved autonomous at all stages that were investigated from 10 hour old embryos to mature larvae. When exposed to the hormonal conditions of metamorphosis, the neoplastic tissues,in situ as well as those that had been culturedin vivo, ceased to grow but failed to metamorphose into parts of the adult integument or brain. Furthermore, in almost all cases they failed to resume their growth in the adult fly after metamorphosis.Thel(2)gl4 brain neuroblastoma changed after prolonged subculturein vivo so that although the cells stopped dividing temporarily at the time of metamorphosis of the host, they resumed their growth shortly thereafter and continued to divide throughout adult development and in the emerged adult. Many of thel(2)gl4 neuroblastoma cells showed abnormal karyotypes, shapes and sizes after prolonged culturein vivo.Thel(2)gl4 imaginal disc tissue resembles, in both structure and behavior, certain atelotypic tissue sublines which arise from wild-type imaginal discs after prolonged culturein vivo: the ultrastructure of both types of cells are similar and both cease to grow when exposed to hormonal conditions of metamorphosis. Apparently the epigenetic processes which transform wild-type imaginal discs afterin vivo subculture into atelotypic neoplasms have the same phenotypic expression as the genetic processes at work inl(2)gl4 imaginal discs.An analysis of several other late larval and larval-pupal mutants with defective imaginal discs, such asl(2)gd andl(1)d.lg.-1 revealed that they also had defects in parts of the brain destined to form adult structures. This observation indicates that mutations that affect imaginal discs of the adult integument also affect the imaginal primordia of the adult brain.The neoplasm in the hemizygous male embryo of the embryonic lethalNotch8 is a teratoma-like growth which is lethal and invasive; when implanted into female adult hosts it gives rise to a mass of tissue which consists of most cell types present in the larval nervous system and a number of other unidentified cell types. It too can be subculturedin vivo for many transfer generations.The relative infrequency of neoplasms in insects is analyzed and an attempt is made to explain this infrequency and the occurrence of neoplasms in specific tissues of insects at specific developmental stages. Genetic and epigenetic neoplasms ofDrosophila are compared with neoplasms of other insects and of vertebrates and it is concluded that both genetic and epigenetic neoplasms ofDrosophila share the essential features of vertebrate neoplasms.ZusammenfassungIn drei letalenDrosophila melanogaster-Mutanten wurde die Entwicklungspotenz von Neoplasmen untersucht und mit der Entwicklungspotenz der entsprechenden Wildtyp-Organe verglichen. Zwei dieser Neoplasmen wurden in den spätlarval-letalen, allelen Mutantenlethal (2) giant larvae4 (l(2)gl4) undlethal (2) giant larvae (l(2)gl) gefunden. Das dritte Neoplasma kommt in den letalen, hemizygoten, männlichen Embryonen der MutanteNotch8 (Df(1)N8) vor. Die Mutantel(2)gl4 bzw. ihr Allell(2)gl, zeigen Schäden in den imaginalen Primordien ektodermaler Herkunft, so z.B. in den Imaginalscheiben des adulten Integuments und in den imaginalen optischen Primordien im larvalen Gehirn.l(2)gl4 undl(2)gl besitzen zusätzlich noch vergrößerte Lymphdrüsen und abnormale Gonaden, Speichel- und Ringdrüsen. Sie transformieren die Imaginalscheiben zu benignen, letalen Neoplasmen und die imaginalen optischen Primordien im larvalen Gehirn zu einem malignen und letalen Neuroblastom. Beide Neoplasmen wurdenin vivo über viele Jahre in Abdomen adulter Weibchen gezüchtet, wo sie den Wirt binnen 7–14 Tage töteten. Die neoplastische Entwicklungsweise derl(2)gl4 Imaginalscheiben als auch desl(2)gl4 Gehirns erwies sich bei allen untersuchten Entwicklungsstadien (vom 10 Std alten Embryo bis zur erwachsenen Larve) als autonom. Im Metamorphosetest erfolgte keinerlei Differenzierung. Es wurde jedoch beobachtet, daß unter dem Einfluß der Metamorphosehormone die Zellen des Implantats sich nicht mehr teilten. Die Teilungsunfähigkeit blieb auch nach Abklingen der hormonalen Wirkung erhalten. Eine Ausnahme macht dasl(2)gl4 Neuroblastom nach längerer Kulturin vivo. Diese Neuroblasten wachsen auch noch weiter, nachdem der hormonelle Einfluß erloschen ist. Sie töten den Wirt kurz nach dem Schlupfen oder bereits im Puppen-Stadium. Nach längerer Kulturin vivo zeigen vielel(2)gl4 Neuroblasten abnormale Karyotypen und Zellgrößen.Dasl(2)gl4 Imaginalscheiben-Epithel ähnelt in Aufbau und Verhalten Gewebe-Sublinien, die aus Wildtyp-Imaginalscheiben nach Kulturin vivo hervorgehen. Offenbar hat der epigenetische Transformationsprozeß bei denl(2)gl4 Imaginalscheiben.Die Analyse einiger anderer spätlarvaler, letaler Mutanten, wiel(2)gd undl(1)d.lg.-1, die abnormale Imaginalscheiben besitzen, zeigte, daß sie auch Defekte in den imaginalen optischen Primordien im larvalen Gehirn verursachen.Das Neoplasma im hemizygoten, männlichen Embryo der embryonal-letalen MutanteNotch8 (Df(1)N8), stellt eine Teratoma-ähnliche Wucherung dar, welche aus fast allen Zelltypen des larvalen Nervensystems, zusammengesetzt ist. In Abdomen adulter Weibchen implantiert verhalten sich diese Zellen maligne und letal.Das relativ seltene Auftreten von Neoplasmen in Insekten wird diskutiert. Es wird versucht, das Vorkommen von Neoplasmen in bestimmten Insektengeweben mit den Eigenheiten der Insektenentwicklung zu korrelieren. Genetische und epigenetische Neoplasmen beiDrosophila werden mit Neoplasmen anderer Insekten und Vertebraten-Neoplasmen verglichen.

A clonal analysis of development inDrosophila melanogaster: Morphogenesis, determination, and growth in the wild-type antenna

Abstract Clones of cells which were genetically marked at specific ages by X-ray induced somatic crossing-over were observed on the antenna of Drosophila melanogaster . Clonally related cells come to occupy modified elongate longitudinal stripes on the antenna involving a curve and an enlargement of the clone on the third antennal segment. The twin spot results showed that the clones deriving from the daughters of most cell divisions are aligned preferentially in the proximodistal axis on the second antennal segment, which indicates that oriented cell divisions play a role in antennal morphogenesis. Factors causing the enlargement of the clone on the third segment were found to be local increases in cell size and locally high rates of cell division. Extensive migration of individual cells was not detected. An analysis of the relative amounts of male and female tissue in the antennae of gynandromorphs indicated that about 9 cells initiate the antenna. Independently, estimates from somatic crossing-over indicate that the antenna consists of approximately 7 cells from about 2 hours after egg deposition until the larva hatches, whereupon the cell number increases logarithmically. Animals in which somatic crossing-over was induced after puparium formation showed a decrease in the frequency of marked clones to a minimum at about the time of pupal cuticle secretion followed by a burst of mitosis, and then cessation of divisions about the time of pupal-adult apolysis. Since the number of cells initiating the antenna estimated from gynandromorph data agrees with the independent estimate of the number of cells in the antenna at about 2 hours after egg laying, obtained by somatic crossing-over, it is suggested that the primative antenna is initially delineated at about 2 hours, the time at which the cleavage nuclei enter the cortex of the egg. A clonal analysis provided results which are consistent with a progressive determination of the antennal disc and several factors characteristic of determined cells are discussed. Some literature on insect mosaics is reexamined in the light of recent findings with Drosophila mosaics. Since there are striking similarities in the clonal patterns of a fruitfly, a wasp, a bee, and a moth (Fig. 14), the conclusion is reached that probably all holometabolous insects employ oriented cell divisions and characteristic growth patterns to a greater degree than individual cell migration in adult morphogenesis. The possible reasons for these morphogenetic characteristics of insects are discussed.

Regulation and metamorphosis of the abdominal histoblasts ofDrosophila melanogaster

SummaryThe development of the adult abdomen ofDrosophila melanogaster was analyzed by histology, microcautery, and genetic strategies. Eight nests of diploid histoblasts were identified in the newly hatched larva among the polytene epidermal cells of each abdominal segment: pairs of anterior dorsal, posterior dorsal, and ventral histoblast nests and a pair of spiracular anlagen. The histoblasts do not divide during larval life but begin dividing rapidly 3 h after pupariation, doubling every 3.6 h. Initially they remain confined to their original area, but 15 h after pupariation the nests enlarge, and histoblasts replace adjacent epidermis cell by cell. The histoblasts cover half the abdomen by 28 h after pupariation and the rest by 36 h. Polytene epidermal cells of the intersegmental margin are replaced last. Cautery of the anterior dorsal nest caused deletion of the whole corresponding hemitergite, whereas cautery of the posterior dorsal nest caused the deletion of the macrochaetae of the posterior of the hemitergite. Cautery of the ventral nest deleted the hemisternite and the pleura, whereas cautery of the spiracular anlagen deleted the spiracle. Results of cautery also revealed that no macrochaetae formed on the tergite in the absence of adjacent microchaetae. Clonal analysis revealed that there were no clonal restrictions within a hemitergite at pupariation. Cautery of polytene epidermal cells other than those of the intersegmental margin failed to affect tergite development. However, cautery of polytene epidermal cells of the intersegmental margin adjacent to either dorsal histoblast nest caused mirror-image duplications of the anterior or posterior of the hemitergite in 10% of the hemitergites. Forty percent of the damaged presumptive hemitergites formed complete hemitergites, indicating extensive pattern regulation and regeneration. Pattern duplication and regeneration were accounted for in terms of intercalation and a model of epimorphic pattern regulation (French et al., 1976). Histoblasts in adjacent segments normally develop independently, but if they are enabled to interact by deleting the polytene epidermal cells of the intersegmental margin, they undergo intercalation which results in duplication or regeneration. The possible role of the intersegmental margin cells of insects in development was analyzed.

Isolation of temperature sensitive mutations blocking clone development inDrosophila melanogaster, and the effects of a temperature sensitive cell lethal mutation on pattern formation in imaginal discs

SummaryA method of isolating temperature-sensitive (ts) mutations blocking clone development, based on the analysis of twin spots produced by X-ray induced somatic recombination is reported. From this screen 10 ts mutations were recovered which caused an absence of the lethal-bearing clone at the restrictive temperature. Eight of these mutations were analyzed. Seven proved to be autonomous ts cell lethals and one was an autonomous ts mutation which reversibly affected cell division and growth of imaginal disc cells and growth of larval cells. The effects on development of one of the cell lethal mutations,l(1)ts-504, are described. Heat pulses (29°C) 24–72 hrs long caused a high frequency (up to 90%) of morphologically abnormal animals. The abnormalities observed were of two major kinds: deficiencies and duplications of imaginal disc derivatives. In addition, alterations of tarsal segmentations occurred. Heat pulses to larvae also delayed pupariation and eclosion by as much as four days. In general, longer pulses led to a greater delay in pupariation and eclosion and a higher frequency of deficiencies and duplications than shorter pulses. Exposure to restrictive temperature early in larval development delayed pupariation and resulted in mostly normal animals; exposure during the second and early third larval instar also delayed pupariation and led to a high frequency of duplications; exposure later in larval life, i.e. mid and late third larval instar, caused no delay in pupariation but led to a high frequency of deficiencies. These results can be explained by the occurrence of areas of cell death, which can be seen in the imaginal discs of larvae exposed to restrictive temperature by staining with trypan blue. This conclusion is further supported by the observation in gynandromorphs of duplications of female nonmutant tissue. These results are discussed in relation to current theories of pattern formation.

The development of the imaginal abdomen of Drosophila melanogaster.

Abstract The development of the imaginal Drosophila melanogaster was investigated by an analysis of clones of cells marked by somatic crossing-over. The results showed that each abdominal half tergite is initiated by about 11 cells which fail to divide during late embryonic and larval stages. Cell division commences only at the time of pupariation, and continues for the next 32 hr. The spatial pattern of marked clones indicated that each tergite consists of a left and right side—marked cells do not cross the midline. Twin spots were examined and indicated that two daughter cells are aligned preferentially in the transverse orientation. These results are contrasted to data obtained for other imaginal anlage, particularly with regard to the time of cessation of cell division.

Intercellular adhesivity and pupal morphogenesis inDrosophila melanogaster

SummaryLeg and wing imaginal discs of mature larvae ofDrosophila melanogaster when treated with 0.1% trypsin for 5–10 min underwent a change in shape that closely resembled normal pupal morphogenesis. Simultaneously, the cells of the disc epithelium changed in shape from tall columnar to cuboidal. Colcemid eliminated microtubules but was without effect on the shape of the imaginal discs or their cells. Tryptic digestion reduced non-junctional intercellular adhesivity but septate desmosomes and gap junctions remained intact.It is proposed that the structure of imaginal discs permits the packaging of the anlagen of the adult integument so that they can change shape and replace the larval structures in a brief period. Apparently most of the definitive form of the pupal leg is built into the disc and becomes visible within a few minutes as intercellular adhesivity is changed.

Cuticle secretion during larval growth in Drosophila melanogaster

Abstract The moulting cycle and growth of the larval integument of Drosophila melanogaster has been studied by light and electron microscopy. Growth during the first, second and third larval instars is accompanied by 3.0-, 3.4- and 3.7-fold increases in surface area, respectively. Growth in surface area occurs continuously during the larval stages, with no detectable relationship to the moulting cycle. Measurements of the thickness of the cuticular layers show that the endocuticle grows in thickness by apposition and in surface area by stretching. The pre-apolytic epicuticle remains at fairly constant thickness during the increase in surface area, indicating that it grows by intussusception of new components. Post-apolytic epicuticle becomes thinner and increases in surface area by stretching. The epicuticle and pre-ecdysial endocuticle are traversed by filaments, but these do not penetrate the endocuticle secreted after ecdysis. We suggest that the filaments transport breakdown products from the old cuticle inward to the epidermis for reutilization. The growth and deposition of cuticle in two larval growth mutants, lethal (2) giant larvae and Chubby Tubby , involves mechanisms similar to those found in wild-type larvae, but in Chubby Tubby the endocuticle contains inclusions which are ultrastructurally similar to dense epicuticle.

Developmental capacities of immature eye-antennal imaginal discs ofDrosophila melanogaster

SummaryThe location of the immature eye-antennal discs ofDrosophila melanogaster in embryos and young larvae was established by means of transplantation experiments. The developmental capacities of these immature discs was then investigated by implanting them into larvae which were ready to metamorphose, thus bypassing large portions of embryonic and larval development. Imaginal eye-antennal discs of embryos and first instar larvae are unable to synthesize eye pigments or secrete cuticle. The discs acquire the first detectable competence in the middle of the second instar, 32–36 hours after hatching, when the eye region of the disc becomes competent to synthesize ommochrome pigments and the rest of the disc becomes competent to secrete a thin untanned, transparent cuticle. Competence to synthesize pteridine pigments becomes evident later, 36–42 hours after hatching. The competence to produce specific bristle and hair patterns is acquired still later, 42–56 hours after hatching. Different regions of the eye-antennal disc acquire competence at different times and the acquisition of competence seems to occur in a proximo-distal sequence within both eye and antennal regions of the disc. In the eye region of the disc, the competence to produce proximal structures such as facets appears before the competence to produce ocelli. Similarly, in the antennal region of the disc, the competence to produce the first antennal segment appears before the competence to produce the third antennal segment or arista. Also, the acquisition of competence to produce a specific cuticular pattern occurs four to six hours earlier in the eye region of the disc than in the antennal region. It was also found that the temporal sequence in which differentiation events actually occur during adult development is similar to the temporal sequence in which specific competences are acquired by the growing immature eye-antennal discs.