Proc. Natl. Acad. Sci. USAVol. 77, No. 11, pp. 6720-6723, November 1980Genetics

Synaptonemal complexes at premeiotic interphase in the mousespermatocyte

(DNA replication/synapsis/recombination/electron microscopy/meiosis)

R. F. GRELL, E. F. OAKBERG, AND E. E. GENEROSOBiology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37830

Communicated by William L. Russell, July 31, 1980

ABSTRACT Male mice were injected intraperitoneally with125 ACi (1 Ci = 3.7 X 1010 becquerels) of [3H~thymidine at 1-hrintervals and killed 1 hr after the second injection. Testes wereprepared for bright-field and electron microscopic autoradi-ography. Primary spermatocytes, identified by light microscopyto be at the premeiotic interphase stage, were found to beheavily labeled. Electron microscopic examination disclosedthe coincidental occurrence of synaptonemal complexes andlabel within the nuclei of premeiotic interphase spermatocytes,indicating synapses of homologues had begun during the Sphase. The significance of this finding for the traditional viewof meiosis is discussed.

In eukaryotes the synaptonemal complex (SC) is present duringa restricted early portion of the first meiotic division and isusually shed from the bivalent when the repulsion phase setsin at diplonema. It is visualized in electron micrographs as atripartite, ribbon-like structure, approximately 200 nm wide,which normally lies between synapsed homologues along theirlongitudinal axis. The function of the SC is not fully understood,although its position coaxial to each bivalent suggests that itserves to hold homologues in close register so as to facilitate theintimate pairing required for recombination (1). Regardless ofits role, its presence in the meiocyte of a diploid organismprovides unequivocal evidence that homologues are sy-napsed.The aspect that this study is concerned with is the time when

the SC initially appears. According to the traditional model ofmeiosis (2), its formation necessarily coincides with the zygotenestage of prophase, when homologous pairing is assumed to begin(1). There is considerable evidence to suggest, however, thatsynapsis may precede prophase and that the fine, leptonemalthreads which become apparent at earliest prophase are alreadypaired (ref. 3; for a review see ref. 4). It this latter view is correct,the SC could serve as an ultrastructural marker of synapsisduring premeiotic interphase. In fact, recent studies show thatthe SC is seen during the meiotic S phase in several organisms,including Drosophila (5, 6), wheat (7), and yeast (8). Thepresent work was undertaken to localize the time of its forma-tion in still another genetically well-known organism. Themouse is favorable for such a study because successive stagesof spermatogenesis are recognizable and have been clearlydefined (9, 10). We present here electron microscopic (EM)autoradiographs that reveal the presence of SCs during the Sphase of premeiotic interphase in the mouse.

The publication costs of this article were defrayed in part by pagecharge payment. This article must therefore be hereby marked "ad-vertisement" in accordance with 18 U. S. C. §1734 solely to indicatethis fact.

6720

EXPERIMENTAL PROCEDUREBright-Field Autoradiography. The mice used in this study

were 12-week-old 101 X C3H hybrid males. Two males weregiven two intraperitoneal injections, 1 hr apart, of 125 MCi (1Ci = 3.7 X 1010 becquerels) of [3H]thymidine (Schwarz/Mann;specific activity, 60 Ci/mmol) in 0.25 ml of distilled water(total, 250 MCi per mouse). The mice were killed 1 hr after thesecond injection (total exposure time, 2 hr). One testis from eachmouse was prepared for bright-field autoradiography byfixation in Zenker/formol for 8 hr, parafin sectioning (5 Mm),and staining in periodic acid/Schiff. Slides were coated withKodak NTB2 liquid emulsion, exposed for 1 wk at 4VC, de-veloped in Kodak-D170, counter-stained in Ehrlich's hema-toxylin, and mounted in Canada balsam. Stages of the cycle ofseminiferous epithelium were identified according to the sys-tem of Oakberg and Huckins (10), and stage 1 was examinedfor the presence of labeled spermatocytes which are unique tothis stage.EM Autoradiography. The remaining two testes were pre-

pared for EM autoradiography by fixation in ice-cold 3%(vol/vol) glutaraldehyde/0. 1 M phosphate buffer, pH 7.2, for2 hr; they were then washed, left overnight in phosphate bufferat 4VC, and postfixed for 1 hr in 1% osmium tetroxide/0. 1 Mphosphate buffer, pH 7.2. After dehydration the tissue wasinfiltrated overnight in propylene oxide/Epon, 1:1 and em-bedded in a fresh Epon mixture. Sections (2 Am thick) were cutwith an LKB ultramicrotome III and examined by one of us(E.O.) for the presence of stage 1 in the cycle of seminiferousepithelium. Blocks identified as stage 1, on the basis of cell typesin thick sections, were used to cut contiguous thin serial sections(90 nm). The thin sections were stained with uranyl acetate andprepared for autoradiography by being coated with Ilford L4emulsion and exposed in light-tight boxes at 40C for 21-23 days.The emulsion was developed in Microdol-X, acid-fixed, air-dried, and stained with lead citrate. The sections were scannedin a Hitachi HU-11C electron microscope for the presence ofthe SC and label in the preleptotene and early leptotene sper-matocyte.

RESULTSSeveral criteria were used to identify premeiotic interphasespermatocytes. First, each cross section of a seminiferous tubulecan be classified as a particular stage of spermatogenesis by thecharacteristic cellular associations (9, 10). The premeiotic in-terphase spermatocyte is unique to stage 1, as are early leptotene

Abbreviation: SC, synaptonemal complex; EM, electron micro-scopic.

Dow

nloa

ded

by g

uest

on

May

29,

202

0

Proc. Natl. Acad. Sci. USA 77 (1980) 6721

spermatocytes, type A1 spermatogonia, and steps 7, 8, and 16spermatids. Pachytene spermatocytes and type As speimato-gonia are found in all stages from 1 to 6. Second, the composi-tion of the karyoplasm of the interphase spermatocyte has acharacteristic appearance-homogeneous and somewhatgranular, unlike the early leptotene nucleus which shows thebeginning of chromosome condensation. Both contrast sharplywith the highly condensed chromatin of the pachytene nucleus.Third, the interphase spermatocyte typically lies close to thebasement membrane of the tubule, whereas the pachytene celllies closer to the lumen. Fourth, the nuclear diameter of theinterphase spermatocyte is less than that of the other premeioticand meiotic germ-cell types present. Finally, if the premeioticinterphase spermatocyte is undergoing DNA synthesis, tracksrepresenting incorporation of [3H]thymidine are often visiblein the autoradiograph.

Fig. 1, a cross section of two stage-i seminiferous tubulesfrom a mouse injected with [3H]thymidine shows a row ofheavily labeled premeiotic interphase spermatocytes lying closeto the basement membrane of the upper tubule and a row ofearly leptotene nuclei lying next to the basement membraneof the lower tubule. Other cell types present at stage 1 are in-dicated. Fig. 2A is a low-magnification EM autoradiograph ofa section of a seminiferous tubule identified as stage 1 that showsthree interphase spermatocytes lying close to the basementmembrane of the tubule. Each possesses a homogeneous kar-yoplasm with no evidence of chromatin condensation. Labelis evident in each, indicating that DNA replication is occurring.Fig. 2B shows two early leptotene nuclei from stage 1 withchromosome condensation beginning; Fig. 2C shows several

::

W_

w nFt

%

4f_i16

4

tk

@<>, :

.' R*9-

9

16

., + 5't- - g > t°,

*'-'@Q4~, *'v| t {<

k.<

ss'tR'-eX't,8 g

.,dit'.K''' pa,

':9f!; ^, a R A"n..

13 .1

* ~ ~~~.A -i bPi

Wof

FIG. 1. Bright-field autoradiograph of a cross section oftwo ad-jacent stage-1 seminiferous tubules. Above-is mid stage 1 with a rowof labeled spermatocytes at premeiotic interphase (PI) lying close tothe basement membrane (Bm). Below is late stage 1 with early lep-totene spermatocytes (L). Associated cell types at stage 1 are pachy-tene (P), step-8 spermatids, and step-16 spermatids.

-.I

..= C

FIG. 2. EM autoradiographs of spermatocytes. (A) Premeioticinterphase spermatocytes (PI) lying close to the basement (Bm) andshowing homogeneous karyoplasm. (X5400.) (B) Leptonema (L)showing the beginning of chromosome condensation. (X5400.) (C)Pachynema (P) with well-condensed chromosomes. (X5400.) Ar-rowhead, radioactive label.

pachytene nuclei with heavily condensed chromatin. Fig. 3Ais a high magnification of a premeiotic interphase spermatocytefrom Fig. 2A showing both label and a well-defined SC at-tached at one end to the nuclear membrane. Fig. SB is an EMautoradiograph of another spermatocyte at premeiotic inter-phase that contains SC and label within a homogeneous kar-yoplasm. An early leptotene nucleus with SC and label (Fig. 4)

Genetics: Grell et al.

4 WP!r.% V.. .I.-1

t .. e

0 r.- if.4 3i!l

a p

Dow

nloa

ded

by g

uest

on

May

29,

202

0

Proc. Nati. Acad. Sci. USA 77 (1980)

1 pm

B

FIG. 3. EM autoradiographs of spermatocytes at premeiotic in-terphase showing SCs and label. (A) Premeiotic interphase from Fig.2A with well-defined SC terminating at the nuclear membrane.(X14,200.) (B) Premeiotic interphase with SCs and granular homo-geneous karyoplasm. (X14,200.) Arrowhead, SC; arrow, radioactivelabel.

suggests that the 2-hr interval between injection and killing issufficient to permit interphase nuclei undergoing late DNAsynthesis to progress into early leptonema as reported by Monesi(11).

DISCUSSIONThe occurrence of SCs during premeiotic interphase or the Sphase is not peculiar to the mouse. Fawcett's (12) early de-scription of SCs in human and cat spermatocytes reports theirpresence when "nuclei appear quite homogeneous" and "beforechromosomes as such are visible in the electron microscope."Because prophase is defined as beginning when chromosomesbecome visibly distinct, Fawcett's description would seem tofix the time of SC appearance as preleptonema. In Drosophilamelanogaster, SCs are extensive within 3-6 hr of pro-oocyteformation (ref. 13; unpublished data) and throughout theensuing 30-hr S phase (5, 6). In the pollen mother cell of wheat,

4.1,

'p.~~~~,4

* w*

arrow, radioactive label

SCs are found at stage 3, which is the S phase of premeioticinterphase (7); and mn synchronized samples of sporulating yeastcells, SCs are present throughout most of the period of DNAreplication (8).Two alternative explanations have been offered to account

for SCs during DNA replication. The first assumes that the aDNA replication occurs during meiotic, prophase; the secondproposes that synapsis of homologues precedes meiotic pro-phase. In studies of organisms whose early meiotic stages remainrefractory to analysis by light microscopy, several investigatorshave accepted the first alternative as correct. Carpenter (14)identified the 30-hr S phase of the Drosophila pro-oocyte aszygo/pachynema because the SC is visible throughout. Petersonet al. (8) concluded that DNA synthesis (localized between 0and 8-9 hr after transfer to sporulation medium) occurs duringzygonema, pachynema, and as late as diplonema because theSC is detected within 2 hr of transfer. These interpretations arein marked contrast to the overwhelming evidence that DNAreplication takes place predominantly during premeiotic in-terphase (15, 16). The fallacy stems from the widespread as-sumption that the SC is restricted to and identifies zygo!pachynema.The premeiotic stages of primary spermnatocytes in the mouse

can be accurately identified by the criteria given above, andtermination of DNA synthesis prior to meiotic prophase hasbeen repeatedly demonstrated (11, 15). The presence of well-defined SCs in the preleptotene spermatocyte provides an un-equivocal demonstration that synapsis of homologues precedesmeiotic prophase.

This finding has important implications for the time andmechanism of recombination. Extensive studies of the Dro-sophila genome utilizing heat treatment to induce (17) or toenhance (5) meiotic recombination have demonstrated that therecombinational responsive period coincides closely with thetime of DNA replication, whereas treatments following the Sphase are entirely ineffective. Independent confirmation thatrecombination and replication are in some way temporallycoupled comes from studies of a temperature-sensitive re-combination mutant of Drosophila (18); these studies haveshown that the ability of the restrictive temperature to reduce

6722 Genetics: Grell et al.

Dow

nloa

ded

by g

uest

on

May

29,

202

0

Proc. Natl. Acad. Sci. USA 77 (1980) 6723

exchange is again limited to the S phase and is identical totheheat-sensitive period for increasing recombination in the normalgenome. Similarly, commitment to meiotic recombination inSaccharomyces cerevisiae is reported to occur duringpremeiotic S phase, probably at its very early stages, and to failto occur in the absence of premeiotic DNA replication (19).

If the synapsis of homologues were not initiated until thezygo/pachytene stages of meiotic prophase, any model whichproposes to couple recombination with replication would befaced with great difficulty; clearly, a cloie approximation ofhomologous regions is a prerequisite for the exchange of geneticmaterial. The finding that homologues are paired throughoutpremeiotic S stage in four of the best-characterized geneticorganisms dispels this difficulty and requires serious consid-eration of molecular models in which recombination is a con-comitant feature of the replication process.

This research was sponsored by the Office of Health and Environ-mental Research, U.S. Department of Energy, under Contract W-7405-eng-26 with the Union Carbide Corporation.

1. Moses, M. J. (1968) Annu. Rev. Genet. 2, 363-412.2. Winiwarter, H. de (1901) Arch. Biol. 17.3. McDermott, A. (1971) Can. J. Genet. Cytol. 13,536-549.

4. Grell, R. F. (1969) in Genetic Organization, eds. Caspari, E. W.& Ravin, A. W. (Academic, New York), pp. 361-492.

5. Grell, R. F. & Day, J. W. (1974) in Mechanisms in Recombina-tion, ed. Grell, R. F. (Plenum, New York), pp. 327-349.

6. Day, J. W. & Grell, R. F. (1976) Genetics 83,67-79.7. McQuade, H. A. & Bassett, B. (1977) Chromosoma 63, 153-

159.8. Petersen, J. G., Olson, L. W. & Zickler, D. (1978) Carlsberg Res.

Commun. 43, 241-253.9. Oakberg, E. F. (1956) Am. J. Anat. 99,391-414.

10. Oakberg, E. F. & Huckins, C. (1976) in Stem Cells ofRenewingCell Populations, eds. Cairnie, A. B., Lala, P. & Osmond, D. G.(Academic, New York), pp. 287-302.

11. Monesi, V. (1962) J. Cell Biol. 14, 1-18.12. Fawcett, D. W. (1956) J. Biophys. Biochem. Cytol. 2, 402-

406.13. Koch, E. A., Smith, P. A. & King, R. C. (1967) J. Morphol. 121,

55-70.14. Carpenter, A. T. C. (1975) Chromosoma 51, 157-182.15. Swift, H. (1950) Physiol. Zool. 23, 169-198.16. Taylor, J. H. (1957) Am. Nat. 91,209-221.17. Grell, R. F. (1971) Genetics 69,523-527.18. Grell, R. F. (1978) Proc. Natl. Acad. Sci. USA 75,3351-3354.19. Kassir, Y. & Simchen, G. (1978) Genetics 90, 49-68.

Genetics: Grell et al.

Dow

nloa

ded

by g

uest

on

May

29,

202

0