Incomplete cre-mediated excision leads to phenotypic differences between Stra8-iCre; Mov10l1(lox/lox) and Stra8-iCre; Mov10l1(lox/Δ) mice

Abstract

In the Cre-loxp system, expression level and activity of Cre recombinase in a Cre deleter line are critical because these determine not only the cell specificity of gene knockout (KO), but also the efficiency of Cre-mediated excision in a specific cell lineage. Although the spatiotemporal expression pattern of a Cre transgene is usually defined upon the generation of the mouse line, the Cre excision efficiency in a specific targeted cell lineage is rarely evaluated and often assumed to be 100%. Incomplete excision can lead to highly variable phenotypes owing to mosaicism (i.e., coexistence of cells with the flox or the recombined flox allele) and this problem has long been overlooked. Here, we report that Stra8-codon-improved Cre recombinase (iCre), a transgenic allele expressing iCre under the control of the male germ cell-specific Stra8 promoter, could efficiently delete one Mov10l1 flox allele in spermatogenic cells, whereas the excision was incomplete when two Mov10l1 flox alleles were present. The incomplete Cre-mediated excision led to a testicular phenotype that was much less severe than that in the true conditional KO (inactivation, 100%) mice. Our findings suggest that it is essential to determine the efficiency of Cre excision when Cre-loxp system is used for deleting genes in a specific cell lineage and the Cre; gene(lox) (/)(Δ) genotype should be used to evaluate phenotypes instead of Cre; gene(lox/lox) owing to the fact that the latter usually bears incomplete deletion of the flox allele(s).

Keywords: Cre; conditional gene knockout; germ line; loxp; mosaicism; phenotype; piRNA; testis.

Figure 1. Visualization of Stra8-iCre expression and iCre activity in developing testes by crossing the Stra8-iCre line with the Rosa26mTmG^tg/tg reporter transgenic mouse line

Confocal microscopic analyses of testicular cryosections from Stra8-iCre; mTmG^+/tg mice at postnatal day 4 (P4), P8, P14, P21, P60, respectively. Representative fluorescent images counterstained with DAPI at various ages are shown and specific membrane-tagged eGFP (mG) signals are only observed within the seminiferous tubules. At P4 and P8, some germ cells exhibit strong mG signals (arrows), while the rest display weak membrane-tagged tomato red (mT) signals (arrowheads). At P14, germ cells with stronger mG expression are spermatocytes. Levels of mG signals on spermatogonia are much lower compared to those on spermatocytes and spermatids in P21 and P60 testes. Scale bar=50μm.

Figure 2. Generation and phenotype of Stra8-iCre; Mov10l1^+/lox (control), Stra8-iCre; Mov10l1^lox/lox (mosaic) and Stra8-iCre; Mov10l1^lox/Δ (true cKO) mice

(a) Male Stra8-iCre; Mov10l1^+/lox mice were crossed with Mov10l1^lox/lox females to produce Stra8-iCre;Mov10l1^lox/Δ male mice. (b) Stra8-iCre; Mov10l1^lox/lox males were generated by mating female Stra8-iCre; Mov10l1^+/lox with Mov10l1^lox/lox males. (c) Representative PCR gel images showing the different genotypes as indicated. Upper panel: wild type and flox allele bands were detected using the first set of primers (FW1 and Rev) at ~160bp and 400bp, respectively (arrows), and both bands were absent after Stra8-iCre-mediated excision of Mov10l1 flox allele (Δ); Lower panel: a ~500bp band indicates the recombined floxed allele (arrowheads). (d) Comparison of the testis weight at postnatal day 60 among the Stra8-iCre; Mov10l1^+/lox, Stra8-iCre; Mov10l1^Δ/lox and Stra8-iCre; Mov10l1^lox/lox males (*: P<0.05; **: P<0.01). (e) Gross morphology and testis size at postnatal day 60 in Stra8-iCre; Mov10l1^+/lox and Stra8-iCre; Mov10l1^lox/Δ male mice. (f) Gross morphology and testis size at postnatal day 60 in Stra8-iCre; Mov10l1^+/lox and Stra8-iCre;Mov10l1^lox/lox male mice (Scale bar=2mm).

Figure 3. Testicular and epididymal histology of Stra8-iCre; Mov10l1^+/lox (control), Stra8-iCre; Mov10l1^lox/lox (mosaic) and Stra8-iCre; Mov10l1^lox/Δ (true cKO) mice

All testis and epididymis samples were collected from postnatal day 60 males. (a, b, c) Micrographs of hematoxylin-eosin (HE)-stained testicular cross-sections of mice with the three genotypes. (d) Magnified view of framed area in panel c. (e) Diameter of seminiferous tubules in the testes of the three genotypes. 50 round or close to round tubules in the cross sections as illustrated in the inset were calculated for statistical analyses (n=50, *: P<0.01). (f) Percentage of seminiferous tubules displaying a typical spermatogenic stage (a total of 12 stages in the wild type mouse) among HE stained cross sections derived from mice with the three genotypes (n=50). Note that no cross sections showing any of the 12 typical stages of the seminiferous epithelial cycle are present in Stra8-iCre; Mov10l1 ^lox/Δ males. (g, h, i) HE-stained cauda epididymal cross-sections from mice of the three genotypes. Note that mature sperm are completely absent in the Stra8-iCre; Mov10l1 ^lox/Δ epididymis, whereas abundant mature sperm are present in the epididymides with the other two genotypes. Scale bar=50μm.

Figure 4. Spermatogenic disruptions in Stra8-iCre; Mov10^lox/lox (mosaic) mice

(a) Normal spermatogenesis in Stra8-iCre; Mov10^+/lox testes at postnatal day 60. (b) Enlarged image of the framed area in panel a, showing spermatogenic stages (marked with Roman numerals) of the seminiferous epithelial cycle in Stra8-iCre; Mov10^+/lox testes at postnatal day 60. All sections were stained using the periodic acid-Schiff's (PAS) reagent that allows accurate staging based upon the shape of the developing acrosome in spermatids. (c-h) Highly variable spermatogenic disruptions in Stra8-iCre; Mov10^lox/lox seminiferous tubules. Panels d-h are magnified images corresponding to the framed area in panel c. (d) Arrows and arrowheads indicate abnormal zygotene spermatocytes and round spermatids, respectively; (e) Arrows and arrowheads demonstrate zygotene-like and pachytene spermatocytes, respectively; Sporadic vacuoles are present, which are generally located in the cytoplasm of Sertoli cells and are indicative of active germ cell depletion. (f) A morphologically normal tubule cross-section containing round spermatids (step 6) (arrows). (g) A seminiferous tubule contains no elongating/elongated spermatids, suggesting a block in the elongation step of spermiogenesis. Arrows indicate the arrested step 4 spermatids. (h) A relatively morphologically normal seminiferous tubule. Scale bar=20μm.

Figure 5. Partial Cre-mediated excision in spermatogenic cells of Stra8-iCre; Mov10^lox/lox testes as evidenced by detection of the Mov10l1 flox allele in cauda epididymal sperm

(a, b, c) Phase contract microscopic images of cauda epididymal contents in mice of the three genotypes indicated (Scale bar=100μm). (d) Semi-qPCR analyses of relative levels of the flox and the recombined flox (Δ) alleles in sperm of Stra8-iCre; Mov10^lox/lox testes. Two sets of primers were used and the first one (FW1 + Rev) was used to detect the flox allele, whereas the other (FW2+Rev) was employed for determining the recombined flox (knockout, Δ) allele.