ATR is a multifunctional regulator of male mouse meiosis

Abstract

Meiotic cells undergo genetic exchange between homologs through programmed DNA double-strand break (DSB) formation, recombination and synapsis. In mice, the DNA damage-regulated phosphatidylinositol-3-kinase-like kinase (PIKK) ATM regulates all of these processes. However, the meiotic functions of the PIKK ATR have remained elusive, because germline-specific depletion of this kinase is challenging. Here we uncover roles for ATR in male mouse prophase I progression. ATR deletion causes chromosome axis fragmentation and germ cell elimination at mid pachynema. This elimination cannot be rescued by deletion of ATM and the third DNA damage-regulated PIKK, PRKDC, consistent with the existence of a PIKK-independent surveillance mechanism in the mammalian germline. ATR is required for synapsis, in a manner genetically dissociable from DSB formation. ATR also regulates loading of recombinases RAD51 and DMC1 to DSBs and recombination focus dynamics on synapsed and asynapsed chromosomes. Our studies reveal ATR as a critical regulator of mouse meiosis.

Fig. 1

A conditional strategy for efficient depletion of ATR during male mouse meiosis. P30 testis and body weights (a), testis western blots (b), and periodic acid–Schiff and hemotoxylin/eosin-stained stage IV testis sections (c) in Atr^fl/+ Stra8-Cre males (n = 9 males), Atr^fl/– Stra8-Cre males (n = 10 males), Atr^fl/+ Ngn3-Cre males (n = 13 males) and Atr^fl/– Ngn3-Cre males (n = 13 males; means and p values for a indicated; unpaired t-test). Testis weights in Atr^fl/+ Stra8-Cre and Atr^fl/+ Ngn3-Cre males are not significantly different from those in Atr^fl/+ Cre negative males derived from the same crosses (n = 17 Atr^fl/+ Cre negative males from the Stra8-Cre cross, p = 0.37, n = 11 Atr^fl/+ Cre negative males from the Ngn3-Cre cross, p = 0.06). Inset in c shows presence of elongating spermatids in some tubules from Atr^fl/– Stra8-Cre males. d, e ATR (magenta) and SYCP3 (green) staining in Atr^fl/+ Ngn3-Cre (denoted Atr^fl/+) and Atr^fl/– Ngn3-Cre (denoted Atr^fl/–) males. In Atr^fl/+ males (n = 2 males) ATR is observed as foci (see insets) in 85% of leptotene cells (n = 20 cells) and 82% of zygotene cells (n = 28 cells), and on the asynapsed region of the XY bivalent in 100% of pachytene cells (n = 30 cells). In Atr^fl/– males (n = 2 males) ATR is observed in no cells at these three stages (n = 21, 30 and 32 cells at leptonema, zygonema and pachynema, respectively). f Validation of ATR depletion in Atr^fl/– by analysis of MSCI. In Atr^fl/+ males, all pachytene cells show coating of the X and Y chromosome (arrowheads) but not the pseuodautosomal regions (PAR, arrowhead) by γH2AX (magenta; left panels; XY bivalent in box magnified in smaller panels). This coating causes silencing of the X-chromosome gene Scml2 (magenta; right panels) and compartmentalization of the XY bivalent (labeled with HORMAD2; green) in the sex body (n = one male, 29 cells; magnified in smaller panels). g In Atr^fl/– males, XY chromosome γH2AX coating and sex body compartmentalization do not occur. As a result, Scml2 expression (arrow) persists in all early pachytene cells (n = one male; 30 cells). Scale bar in (c) 20 µm, other scale bars 10 µm

Fig. 2

Mid-pachytene germ cell elimination is preserved in mice deficient in the PIKKs. Western blot showing a ATR and b ATM depletion in mice with different PIKK mutations. c γH2AX (magenta) and SYCP3 (green) immunostaining (n = 2 males, 25 cells for each stage) and d stage IV elimination in Atr^fl/– Atm^−/− males. e γH2AX and SYCP3 immunostaining (n = 2 males, 25 cells for each stage) and f stage IV elimination in Atr^fl/– Atm ^fl/– Prkdc^scid/scid males. Scale bar 20 µm in d, f and 10 µm in c, e

Fig. 3

ATR is required for homologous synapsis. Examples of early pachytene synaptic outcomes in a Atr^fl/+ males (n = 3 males) and b Atr^fl/– males (n = 3 males) assessed using HORMAD2 (magenta), SYCP3 (green) and subsequent DNA FISH using Slx and Sly probes (labeled in insets as X chromosome in green and Y chromosome in white). The Slx probe hybridizes to a sub-region of the X chromosome (circled), while the Sly probe coats the majority of the Y chromosome. In Atr^fl/+ males (a), both the autosomes and the XY PARs are synapsed, while the non-homologous regions of the XY pair are asynapsed. In Atr^fl/+ males (n = 104 cells, 2 males), 90 cells had normal synapsis, 12 cells had asynapsis of both the XY and autosomes, 1 cell had asynapsis only of the XY and 1 cell had asynapsis only of the autosomes. b Three examples of synaptic defects in Atr^fl/– males, each described above respective image. In Atr^fl/– males (n = 107 cells, 2 males), 21 cells had normal synapsis, 59 cells had asynapsis of both the XY and autosomes, 23 cells had asynapsis only of the XY, 4 cells had asynapsis only of the autosomes, 10 cells had X self-synapsis, 11 cells had Y self-synapsis and 7 had non-homologous synapsis between the X and/or Y and autosomes. c, d Comparison of pachytene synaptic outcomes in c Spo11^−/− and d Atr^fl/– Spo11^−/− males using immunostaining for SYCE2 (magenta) and SYCP3 (green). e Quantitation of complete, partial and focal SC in Spo11^−/− males (n = 2 males; 88 cells) and Atr^fl/– Spo11^−/− males (n = 2 males; 96 cells). Means and p values (unpaired t-test) indicated. f–h Epistasis analysis of SPO11 and ATR in synapsis using the same markers plus immunostaining for centromeres (CENT; white) to determine CS number. For each cell, colocalizing SYCE2-CENT signals are indicated with dashed circles, and resulting CS numbers are shown. i CS number in Atr^fl/+ males (n = 2 males; 55 cells), Atr^fl/– males (n = 2 males; 37 cells), Spo11^−/− males (n = 2 males; 114 cells) and Atr^fl/– Spo11^−/− males (n = 2 males; 64 cells). Mean values and p values (Mann–Whitney test) indicated. Scale bars 10 µm

Fig. 4

ATR ablation does not alter DSB levels but leads to reduction in leptotene recombinase counts. a–d Analysis of SPO11-oligo complexes in P13 testes. SPO11 was immunoprecipitated from whole-testis extracts and SPO11-associated oligos were end-labeled with terminal deoxynucleotidyl transferase and [α-³²P] dCTP, then either separated on SDS-PAGE gels followed by autoradiography (a) and western blotting with anti-SPO11 antibody (b) or digested with proteinase K and resolved on denaturing polyacrylamide sequencing gels (c; background-subtracted lane traces in d). A representative experiment is shown. An additional Atr^fl/+ Cre– control is shown to demonstrate that the Ngn3-Cre transgene does not influence SPO11-oligo levels. In a, b, the bar indicates SPO11-oligo complexes, arrowhead indicates the immunoglobulin heavy chain and asterisk marks non-specific labeling; ND not determined. Each panel shows lanes from the same exposure of a single western blot or autoradiograph, with intervening lanes omitted. For quantitation, SPO11-oligo complex signals were background-subtracted and normalized to Atr^fl/+ Cre– (n = 2) controls. SPO11-oligo quantitation: Atr^fl/+ (1.05 ± 0.07-fold, mean and s.d., n = 2 males), Atr^fl/– (1.17 ± 0.22-fold, n = 4 males), Atm^−/− (14.58 ± 5.24-fold, n = 3 males) and Atr^fl/– Atm^−/− males (11.38 ± 13.15-fold). The reduced SPO11 protein levels in Atm^−/− were previously documented^, , but the molecular basis is not understood. e–j Analysis of leptotene focus counts using SYCP3 (green) and early recombination markers (magenta): RPA (e, f), RAD51 (g, h) and DMC1 (i, j) in Atr^fl/+ males (n = 2 males; 27 cells for RPA, 19 cells for RAD51, 14 cells for DMC1), Atr^fl/– males (n = 2 males; 21 cells for RPA, 19 cells for RAD51, 13 cells for DMC1), Atm^−/− males (n = 2 males; 22 cells for RPA, 15 cells for RAD51, 13 cells for DMC1) and Atr^fl/– Atm^−/− males (n = 2 males; 10 cells for RPA, 20 cells for RAD51, 13 cells for DMC1). RPA counts are not significantly different between Atr^fl/– males (mean 249 foci) and Atr^fl/+ males without the Ngn3-Cre transgene (mean 227 foci, n = 2 males, 17 cells), and are thus not influenced by Atr haploinsufficiency. Mean and p values (Mann–Whitney test) indicated. Scale bars 10 µm

Fig. 5

ATR regulates DSB marker counts during zygonema and pachynema. a, b RPA (magenta) and SYCP3 (green) immunostaining in Atr^fl/+ and Atr^fl/– males. c RPA, RAD51 and DMC1 counts at mid zygonema in Atr^fl/+ males (n = 2 males; 16 cells for RPA, 15 cells for RAD51, 15 cells for DMC1) and Atr^fl/– males (n = 2 males; 16 cells for RPA, 15 cells for RAD51, 15 cells for DMC1). d, e Examples of early pachytene cells from Atr^fl/+ and Atr^fl/– males with normal autosomal synapsis. Two representative autosomes (aut 1 and 2) are boxed in upper panels and magnified in lower panels. f autosomal RPA, RAD51 and DMC1 counts at early pachynema in Atr^fl/+ males (n = 2 males; 49 cells for RPA, 19 cells for RAD51, 33 cells for DMC1) and Atr^fl/– males (n = 2 males; 43 cells for RPA, 19 cells for RAD51, 35 cells for DMC1). g, h Early pachytene Atr^fl/+ and Atr^fl/– cells with normal autosomal synapsis and the X chromosome (boxed in upper panels) identified by Slx DNA FISH (dashed circles in lower panels). i Early pachytene Atr^fl/– cell with autosomal asynapsis. j X-chromosome RPA, RAD51 and DMC1 counts at early pachynema in Atr^fl/+ males (n = 2 males; 64 cells for RPA, 49 cells for RAD51, 59 cells for DMC1) and Atr^fl/– males (n = 2 males; 62 cells for RPA in autosomal synapsis category, 35 cells for RPA in autosomal asynapsis category, 36 cells for RAD51, 53 cells for DMC1). Mean and p values (Mann–Whitney test) indicated. Scale bars 10 µm

Fig. 6

ATR regulates RNF212 counts during pachynema. a–c RNF212 (magenta) and SYCP3 (green) immunostaining at early pachynema in Atr^fl/+ and Atr^fl/– males. d RNF212 counts in Atr^fl/+ male (n = 1 male; 29 cells) and Atr^fl/– male with normal autosomal synapsis (n = 1 male; 15 cells) or autosomal asynapsis (n = 1 males; 25 cells). Mean and p values (Mann–Whitney test) indicated. Scale bars 10 µm