Activation-Induced Cytidine Deaminase Does Not Impact Murine Meiotic Recombination

Abstract

Activation-induced cytidine deaminase (AID) was first described as the triggering enzyme of the B-cell-specific reactions that edit the immunoglobulin genes, namely somatic hypermutation, gene conversion, and class switch recombination. Over the years, AID was also detected in cells other than lymphocytes, and it has been assigned additional roles in the innate defense against transforming retroviruses, in retrotransposition restriction and in DNA demethylation. Notably, AID expression was found in germline tissues, and in heterologous systems it can induce the double-strand breaks required for the initiation of meiotic recombination and proper gamete formation. However, because AID-deficient mice are fully fertile, the molecule is not essential for meiosis. Thus, the remaining question that we addressed here is whether AID influences the frequency of meiotic recombination in mice. We measured the recombination events in the meiosis of male and female mice F1 hybrids of C57BL/6J and BALB/c, in Aicda^+/+ and Aicda^-/- background by using a panel of single-nucleotide polymorphisms that distinguishes C57BL/6J from BALB/c genome across the 19 autosomes. In agreement with the literature, we found that the frequency of recombination in the female germline was greater than in male germline, both in the Aicda^+/+ and Aicda^-/- backgrounds. No statistical difference was found in the average recombination events between Aicda^+/+ and Aidca^-/- animals, either in females or males. In addition, the recombination frequencies between single-nucleotide polymorphisms flanking the immunoglobulin heavy and immunoglobulin kappa loci was also not different. We conclude that AID has a minor impact, if any, on the overall frequency of meiotic recombination.

Keywords: AID; cytidine deaminase; double-strand breaks; germline; meiotic recombination.

Figure 1

(A) Recombination events for all studied groups. The numbers of recombination events were calculated for all 19 autosomes per offspring of females F1.Aicda^+/+ (FWT, n = 79), females F1. Aicda^−/− (FKO, n = 79), males F1. Aicda^+/+ (MWT, n = 78), and males F1. Aicda^−/− (MKO, n = 78). Each dot represents a sample; lines represent mean ± SEM. A Mann-Whitney test was used to compare sample groups using a Bonferroni correction for multiple testing, * P < 0.0125; n.s., nonsignificant. (B) Average recombination events for centromeric and telomeric regions. For these analyses, only the SNP pairs closer to the regions were included. Thus, for the centromeric region only, the recombination frequencies of the first SNP pair in chromosomes 1, 7, 8, 11, 18, and 19 (n = 6) were used in the comparison between progeny of the female (FWT) and the male (MWT) F1.Aicda^+/+. For the telomeric regions only, the recombination frequencies of the last SNP pair in chromosomes 1, 2, 3, 4, 5, 8, 9, 11, 14, 16, 17, and 19 (n = 12) were used in the comparison between the progeny of the female (FWT) and the male (MWT) F1.Aicda^+/+. Bar and bar error represent mean ± SEM. A Mann-Whitney test was used to compare sample groups using a Bonferroni correction for multiple testing, * P < 0.0125; n.s., nonsignificant. (C) Average recombination events per physical size (Mbp/cM) for Chromosomes 1 and 19 for F1.Aicda^+/+ progeny. Bar and bar error represent mean ± SEM. A Mann-Whitney test was used to compare FWT vs. MWT groups in each chromosome, *P < 0.05.

Figure 2

Proportions of recombination events per SNP pair for all 19 autosomes calculated for offspring of females F1.Aicda^+/+ (FWT, n = 79, white bars), females F1.Aicda^−/− (FKO, n = 79, white bars with stripes), males F1.Aicda^+/+ (MWT, n = 78, gray bars), and males F1.Aicda^−/− (MKO, n = 78, gray bars with stripes). On the x-axis, the order of the SNP pairs is shown. Error bars correspond to 95% confidence intervals as estimated by the Agresti-Coull method. A Fisher-exact test was used to compare sample groups using a Bonferroni correction for multiple testing, *P < 0.000225.

Figure 3

Average recombination frequencies of the IgH locus (chromosome 12), a putative AID target. Recombination frequencies were calculated using genotyping results from SNP in approximate position 108 Mbp and PCR results from VNTR sequence that distinguishes C57BL/6J from BALB/C genome in approximate position 116 Mbp calculated for offspring of FWT (n = 68), FKO (n = 67), MWT (n = 68), and MKO (n = 67). Bar and bar error represent mean ± SEM. A Mann-Whitney test was used to compare sample groups using a Bonferroni correction for multiple testing, *P < 0.0125; n.s. is non-significant.

Figure 4

(A) Aicda/beta-actin expression levels measured by real-time PCR in activated B-cells of B6.Aicda^+/+ and B6.Aicda^−/−, B6.Aicda^+/+ sorted 4N, 2N, and N (WT 4N, 2N, and N, respectively) spermatocytes, and unsorted testicular (WT T) and epididymal (WT E) sperm cells. (B) Aicda/beta-actin expression levels measured by real-time PCR in activated B-cells of B6.Aicda^+/+ and B6.Aicda^−/−, and B6.Aicda^+/+ ovaries from non-superovulated females (WT ovary-non-superovulated), “complete oocyte” samples (WT oocytes + granulosa cells); only cells of the granulosa “oocyte free” (WT granulosa cells) and isolated oocytes (WT oocytes). Error bars correspond to two independent experiments in which the tissues were pooled from 16−20 females per sample. (C) Western blot analysis of activated B-cells of B6.Aicda^+/+ and B6.Aicda^−/−, B6.Aicda^+/+ ovaries from non-superovulated females (WT ovary non-superovulated), “complete oocyte” samples (WT oocytes + granulosa cells); only cells of the granulosa “oocyte free” (WT granulosa cells) and isolated oocytes (WT oocytes).