/ . Embryo!, exp. Morph. Vol. 33, 4, pp. 1003-1011, 1975 1003Printed in Great Britain

Restoration of the capacity to form pole cellsin u.v.-irradiated Drosophila embryos

By RICHARD WARN1

From the MRC Laboratory of Molecular Biology, Cambridgeand the Genetics Laboratory, Department of Biochemistry, Oxford

SUMMARY

Injection of pole plasm into u.v.-irradiated posterior poles of early Drosophila embryosleads to the restoration of the capacity to form pole cells in nearly half of the recipients. Theeffect is specific, since cytoplasm from the anterior tip has no such result. In most cases onlya small number (between 1 and 5) of discrete pole cells are formed. However, a large numberof pole cell fragments with or without nuclei occur. Occasionally pole cells were formedoutside the area of the originally irradiated pole plasm. This happened when material wasinjected more anteriorly than usual. Thus polar cytoplasm contains some factor(s) necessaryfor the formation of pole cells.

INTRODUCTION

The pole plasm of insect embryos has long been believed to contain specialsubstances required to influence the activity of the nuclei which subsequentlyenter it (Hegner, 1914). In Drosophila the pole cells which are formed from theposterior region of the egg are precociously budded off f h prior to the cellu-larization of the rest of the embryo. They have a characteristic morphologywhich is very different from that of the other cells which are subsequentlyformed. Thus, even at the time of their formation, pole cells have certaincharacteristics which mark them off from the cells surrounding them. The cellswhich come to contain pole plasm have a precisely determined fate. The germcell line comes only from these cells.

Destruction of pole cells by u.v.-irradiation (Geigy, 1931; Aboim, 1945) ledto the absence of sex cells in the adult gonads and showed that regulation ofthis region of the embryo did not occur, even early in development. More recentstudies (Poulson & Waterhouse, 1960; Hathaway & Selman, 1961; Warn,1972; Graziosi & Micali, 1974) have shown that the loss or destruction of poleplasm in the newly fertilized egg leads to a failure of pole cell formation andsubsequently to adults with sterile gonads.

So far little is known of the nature of pole cell determinants, although it hasbeen hypothesized (Mahowald, 1968) that they are contained as maternal mRNA

1 Author's address: Genetics Laboratory, Department of Biochemistry, South Parks Road,Oxford 0X1 3QU, U.K.

1004 R. WARN

within the polar granules. These organelles are characteristic of the pole cells ofDrosophila (Huettner, 1923). In order to identify the constituents of pole plasmwhich are necessary for its specific biological effects it is necessary to have adirect assay for these effects. Smith (1966) first used this kind of assay system todemonstrate the biological properties of the germinal plasm of Rana pipiens.In this paper the results of injecting the pole plasm into irradiated posteriorpoles of early-stage Drosophila embryos are described. The results agree wellwith the recent papers of Okada, Kleinman & Schneiderman (19746) andIllmensee & Mahowald (1974).

MATERIALS AND METHODS

Stocks and collection of eggs

Eggs were collected from standard collecting boxes containing about10000 flies, 4-8 days old. To obtain approximately synchronized eggs, it wasnecessary to pre-feed the flies for at least 1 h with collecting dishes containingan agar medium rinsed with a solution of 7 % acetic acid and smeared witha thin layer of yeast. Eggs were then collected at 10 min intervals using similarcollecting dishes filled with yeast agar medium topped with live yeast butwithout acetic acid.

U.v.-irradiation

The chorion was first removed mechanically. The eggs were lifted out of thecollecting dish, immobilized on double-sided 'Scotch' tape, and dried for2-3 min in a current of warm air from a hair drier. The chorion then becamerather brittle and could be cracked open with a mounted needle, freeing theembryo.

U.v.-irradiation was carried out with the short-wavelength region of a Minera-lite UVSL-15 lamp (Ultra-violet Products Inc.), lacking the filter. This givesout light with a wavelength of 254 nm. The method of irradiation was basedon the technique of Graziosi & Micali (1974). A block of agar (3 % in Becker'ssaline and blackened by the addition of powdered charcoal) 1 mm thick wasplaced on a microscope slide and cut in two. The eggs were oriented with theirposterior poles upwards along the cut surface of one half of the agar block,and the two halves were then fitted together. The embryos were placed 30 mmfrom the u.v. source. Prior to each experiment the output of the u.v. lamp waschecked with a dosimeter (Black-Ray, Ultra-violet Products Inc.). A dose of3000 ergs/mm2, delivered over 15 sec, was adopted. At this dose a high pro-portion of embryos did not show any generalized irradiation damage, but it isalso well above the limit where any pole cells were seen to form (1000 ergs/mm2).The total time to prepare the eggs and carry out the irradiation was about10 min.

U.v.-irradiated Drosophila embryos 1005

Micro-injection

Micro-injection was carried out using a glass micropipette held in a Singer'micro-dissector', and attached to an Agla syringe (as described by Elsdale,Gurdon & Fischberg, 1960). The pipettes were prepared from BDH hard glasscapillary tubes. After an initial pulling-out over a small Bunsen burner, thepipettes were tapered to a final external diameter of about 10 fim and an internaldiameter of about 5 /tm and had a very short tip. Smoothing of this tip wasaccomplished by melting the edge against a heated wire and carefully pulling itaway. (A detailed description of the method of drawing and smoothing micro-pipettes is given by Gurdon, 1974.) The whole shaft of the pipette was filled withparaffin oil so as to transmit small changes in pressure without delay. However,as injections were carried out in paraffin oil, a tiny bubble of air had to beintroduced into the very tip to permit identification of its position.

Transfer of cytoplasm

A group of about 15 embryos was irradiated while ca. 10 donor eggs were leftdrying on the 'Scotch' tape. The irradiated embryos were transferred to aFalcon Petri dish and dried for 6 min at 25 °C in a calcium chloride desiccator.This reduced the turgor pressure so that cytoplasm could be introduced withoutloss of fluid. Similarly, the donor embryos were dechorionated and desiccated.After this period liquid paraffin (Boots' Liquid Paraffin) was poured into thedishes and the embryos examined under transmitted light. Following theobservations of Imaizumi (1954) it was possible to determine the age of earlyembryos accurately and all stages older than 20 min were discarded. The re-maining irradiated embryos were divided into two lots; one group was trans-ferred into another Petri dish to act as recipients whilst the others served ascontrols. The donor embryos were embedded in deep grooves to hold themfirm, prior to the addition of the irradiated recipients which were placed inshallower hollows to allow easier subsequent handling.

Between 0-1 and 0-2 nl of fluid (which represents about 1 % of the egg volume)was removed from directly under the vitelline membrane at the posterior pole.As the cytoplasm was slowly sucked out the angle of the pipette was changedslightly so that surrounding pole plasm and not the underlying yolky cytoplasmwas taken up. It was then introduced into the recipient embryo at the dorsalmargin of its pole plasm, so as to introduce new material without disturbingunnecessarily the organization of the original pole plasm. Any embryos wherecytoplasm exuded from the injection wound after the operation were removed,as were all the donor embryos. In general, cytoplasm from one donor wasinjected into one recipient. The whole operation took about 30 min. A controlseries, injected with cytoplasm from the anterior tip of the egg, was carried outin exactly similar fashion. After injection, the embryos were cultured at roomtemperature in darkness to avoid the possibility of photoreactivation. At three

1006 R. WARN

Table 1. Results of cytoplasmic transfer into the posterior poles ofirradiated embryos o/"Drosophila

Treatment ofirradiated embryos

A. Injected with posterior pole plasmControl, no injection

B. Injected with anterior tip cytoplasmControl, no injection

Table 2. Results of each pole plasm transfer experiment(from first line of Table 1)

Withpole cells

180

00

Withoutpole cells

2122

1616

Total no. embryos injectedNo. embryos with pole cellsNo. embryos lackingpole cells

No. pole cells in eachpositive case

1

431

51

43

2

413

1—

—

3

541

2532

Experiment

4

413

2—

—

5

321

54

—

no.

6

422

33

—

7

725

11

—

8

404

——

—

9

431

3132

hours of development they were removed and sections were cut to look for thepresence or absence of pole cells.

Histology

After removal from the oil the embryos were placed for 30 min in a form-aldehyde-alcohol-acetic acid mixture (with proportions of 6:16:1 plus 30 partswater) to fix the exterior of the eggs, thereby reducing their fragility. They werethen punctured with a fine glass micro-needle and full-strength solution (lackingthe water) was substituted. To this 0-5 % Triton X detergent had been added.Sayles, Procunier & Browder (1973) found that this detergent was successfulin making eggs permeable, and a combination of Triton X and puncture resultedin very good fixation of embryos. After impregnation with Paraplast (SherwoodMedical Industries Inc.), sections were cut at 6/tm and stained with Heiden-hain's iron haematoxylin.

RESULTS

In Tables 1 and 2 a summary of the principal findings has been drawn up.A high percentage (nearly half) of embryos injected with pole plasm containedmorphologically normal pole cells, whereas the embryos injected with anteriortip cytoplasm as well as the irradiated controls showed no sign of pole cell

U.v.-irradiated Drosophila embryos 1007

Fig. 1. Posterior pole region of an unmanipulated mid-blastema embryo. x2200.b.n., blastema nucleus; p.c, pole cell; p.g., polar granules.

hn

Fig. 2. Posterior pole of a mid-blastema stage embryo lacking pole cells afterirradiation of this region at about the 1st cleavage stage, x 1900. b.n., blastemanucleus; v.m., vitelline membrane.

1008 R. WARN

Fig. 3. Mid-blastema embryo resulting from an egg irradiated 15 min after fertiliza-tion. Pole plasm was injected at the 4-8 cell stage, x 1900. n.p.c, morphologicallynormal pole cell; a.c, abnormally formed cell; b.n., blastema nucleus.

Fig. 4. Mid-blastema embryo. This was irradiated 15 min after fertilization andinjected with anterior tip cytoplasm. xl300. p.p., posterior pole; b.n., blastemanucleus.

U.v.-irradiated Drosophila embryos 1009

formation. The difference is seen in Figs. 1-4. It proved necessary to definewhat constituted a pole cell because a spectrum of effects was seen in embryoswhich received pole plasm. As shown in Fig. 1, a typical pole cell is circular incross-section in contrast to the columnar epithelial cells of the blastoderm. Theratio of cytoplasm to nucleus is much less, and in suitable preparations polargranules can be seen. Pole cell nuclei are spherical and the chromatin stainsfairly densely as a clump, while the somatic nuclei form ovoid structures witha prominent nucleolus attached to a less densely staining strand of nucleoplasm.

In general only a small number (1-5) of fully formed pole cells were observed(cf. the last line of Table 2). However, quite a large number of cellular frag-ments, with or without nuclei, were seen. Such fragments differed from thedeveloping epithelial cells because they were not included in the blastoderm celllayer. Furthermore, the general structure and nuclear morphology were similarto those of pole cells. Because of this, the results given in Tables 1 and 2 area minimum estimate of the effects of the injected pole plasm. Two embryosinjected with pole plasm contained fragments but no normally formed polecells. They were included among the negative results in Table 1. Fig. 3 showsone of these cell-like fragments on the left-hand side. The nucleus appearsfairly well developed but there is only a small amount of cytoplasm, which is,in fact, still attached to the main mass of the egg. In contrast, four normalpole cells are present in the centre of the field.

Two cases of transfer of posterior pole plasm into irradiated embryos pro-vided more spectacular results. In one 40 pole cells were formed, and in theother 31. These numbers are just below the bottom end of the spread of numbersof pole cells in unmanipulated embryos (41-75, average 58).

Not all embryos contained pole cells solely in the region of the posteriorpole. In two cases pole cells were found in positions some way removed fromthe posterior pole. In both embryos the injection wound was close to the site offormation of the pole cells, suggesting that the material forming the pole cellsdoes not migrate far from the point of its introduction.

How constant are the results of pole plasm transfers ? Nine separate experi-ments were carried out involving 43 recipient embryos. Thirty-nine of thesedeveloped, of which 18 (46%) contained normally formed pole cells. Thesuccess rate per experiment varied, but not significantly, and in most cases atleast two or three positive results were obtained. Only one experiment gave nopositive results. The two embryos with large numbers of pole cells came fromdifferent experiments.

DISCUSSION

The above experiments show that pole plasm has a specific capacity to inducepole cell formation in irradiated posterior poles and sometimes also in moreanterior regions of the embryo. Because of the two cases where pole cells wereformed in more anterior regions of the embryo, pole plasm can act outside the

1010 R. WARN

area of the posterior pole. Its activity is unaffected by its position in the embryo.This result has also been found by Ulmensee & Mahowald (1974) as describedbelow. Pole plasm transfer constitutes the transfer of specific 'determinants'because a particular cell type, as judged by the criteria described, is formedfrom nuclei which can be assumed to be totipotent (Zalokar, 1971; Ulmensee,1972). However, there are other types of cytoplasmic transfer experiment inDrosophila (Garen & Gehring, 1972; Okada, Kleinman & Schneiderman,1974o) where the factor restored promotes further general development ratherthan the appearance of a new cell type. In these cases, the molecules requiredto continue development cannot be considered as 'determinants'.

The results described in this paper depend on morphological criteria toidentify the presence or absence of pole cells. Two recent papers (Okada et ah19746; Ulmensee & Mahowald, 1974) have shown that pole cells resulting fromcytoplasmic transfer can differentiate to form mature gametes. Okada et ah(19746) achieved such results by a similar procedure to that described in thispaper, dissecting the adults that emerged to examine the state of the gonads.Ulmensee & Mahowald (1974) injected the pole plasm into the anterior tips ofembryos of the same age as the donor and observed the subsequent formationof cells very similar to pole cells. Transfer of these cells into the posterior poleof blastoderm stages of different genotype yielded a few imagines which gaveoffspring carrying the genetic markers of the transferred 'pole' cells.

Although the results described from the present work do not show that thecells formed as a result of pole plasm transfer have all the properties of pole cells,they do provide a third independent demonstration of the biological effects oftransferring such cytoplasm in Drosophila. Furthermore, examination of em-bryos at the earliest stage when specific effects due to injection of pole plasmcan be seen provides a simple and sure test. More than 90 % of all embryosdevelop to this stage after manipulation. However, during later developmenta high proportion of embryos irradiated or injected at the posterior tip die(Warn, 1972). Thus examination of embryos for the earliest effects of injectionof fractions would be a preferable method of assaying the biological activityof the constituents of pole plasm such as polar granules.

I would like to thank Dr J. B. Gurdon for much help and advice during the course of thisstudy and also Dr D. B. Roberts and Dr G. Graziosi for their help and critical reading ofthe manuscript.

U.v.-irradiated Drosophila embryos 1011

REFERENCES

ABOIM, A. N. (1945). Developpement embryonnaire et postembryonnaire des gonadesnormales et agametiques de Drosophila melanogaster. Revue suisse Zool. 52, 53-154.

ELSDALE, T. R., GURDON, J. B. & FISCHBERG, M. (1960). A description of the technique fornuclear transplantation in Xenopus laevis. J. Embryol. exp. Morph. 8, 437-444.

GAREN, A. & GEHRING, W. (1972). Repair of the lethal developmental defect in deep orangeembryos of Drosophila by injecting egg cytoplasm. Proc. natn. Acad. Sci. U.S.A. 69,2982-2985.

GRAZiosr, G. & MICALI, F. (1974). Differential responses to ultraviolet irradiation of thepolar cytoplasm of Drosophila eggs. Wilhelm Roux Arch. EntwMech. Org. 175, 1-11.

GEIGY, R. (.193.1). Action de l'ultraviolet sur le pole germinal dans l'ceuf de Drosophila melano-gaster. Revue suisse Zool. 38, 187-288.

GURDON, J. B. (1974). The Control of Gene Expression in Animal Development. London:Oxford University Press.

HATHAWAY, D. S. & SELMAN, G. G. (1961). Certain aspects of cell lineage and morpho-genesis studied in embryos of Drosophila melanogaster with an ultraviolet microbeam./ . Embryol. exp. Morph. 9, 310-325.

HEGNER, R. W. (1914). The Germ Cell Cycle in Animals. New York: Macmillan.HUETTNER, A. F. (1923). The origin of the germ cells in Drosophila melanogaster. J. Morph.

39, 249-265.ILLMENSEE, K. (1972). Developmental potencies of nuclei from cleavage, preblastoderm, and

syncytial blastoderm transplanted into unfertilized eggs of Drosophila melanogaster.Wilhelm Roux Arch. EntwMech. Org. 170, 267-298.

ILLMENSEE, K. & MAHOWALD, A. P. (1974). Transplantation of posterior pole plasm inDrosophila. Induction of germ cells at the anterior pole of the egg. Proc. natn. Acad. Sci.U.S.A. 71, 1016-1020.

IMAIZUMI, T. (1954). Recherches sur l'expression des facteurs letaux hereditaires chez l'em-bryon de la drosophile. Protoplasma 44, 1-10.

MAHOWALD, A. P. (1968). Polar granules of Drosophila. II. Ultrastructural changes duringearly embryogenesis. / . exp. Zool. 167, 237-262.

OKADA, M., KLEINMAN, I. A. & SCHNEIDERMAN, H. A. (1974a). Repair of a genetically causeddefect in oogenesis in Drosophila melanogaster by transplantation of cytoplasm from wildtype eggs and by injection of pyrimidine nucleosides. Devi Biol. 37, 55-62.

OKADA, M., KLEINMAN, I. A. & SCHNEIDERMAN, H. A. (19746). Restoration of fertility insterilized Drosophila eggs by transplantation of polar cytoplasm. Devi Biol. 37, 43-54.

POULSON, D. F. & WATERHOUSE, D. F. (1960). Experimental studies on pole cells and midgutdifferentiation in Diptera. Aust. J. biol. Sci. 13, 541-567.

SAYLES, C. D., PROCUNIER, J. D. & BROWDER, L. W. (1973). Radiolabelling of Drosophilaembryos. Nature, New Biol. 241, 215-216.

SMTTH, L. D. (1966). The role of a 'germinal plasm' in the formation of primordial germcells in Rana pipiens. Devi Biol. 14, 330-347.

WARN, R. (1972). Manipulation of the pole plasm of Drosophila melanogaster. Acta Embryol.exp. Suppl. 415-427.

ZALOKAR, M. (1971). Transplantation of nuclei in Drosophila melanogaster. Proc. natn.Acad. Sci. U.S.A. 68, 1539-1541.

{Received 20 November 1974)

63 EMB 33