Myh6 promoter-driven Cre recombinase excises floxed DNA fragments in a subset of male germ line cells

Abstract

Myh6-Cre transgenic mouse line was known to express Cre recombinase only in the heart. Nevertheless, during breeding Myh6-Cre to Rosa26^fstdTom reporter (tdTom) mouse line, we observed that a significant part of their F2 tdTom/+ offspring had tdTom reporter gene universally activated. Our results show that Myh6-Cre transgenic mice have Cre recombinase activity in a subpopulation of the male germ line cells, and that Myh6 gene transcripts are enriched in the interstitial Leydig cells and the undifferentiated spermatogonia stem cells. In summary, the current study confirms that the previously known “heart-specific” Myh6 promoter drives Cre expression in the testis.

Graphical Abstract

Introduction

Cre recombinase (Cre) is a protein that recognizes pairs of 34 bp loxP DNA sequences and the Cre-loxP system has been a gold standard for tissue-specific gene function studies. Controlled by a tissue-specific promoter, the Cre recombinase is able to recognize paired LoxP sequences and mediate DNA recombination events in a certain cell type, thus allowing for tissue-specific inactivation, activation or mutation of genes of interest.

In transgenic Cre mouse lines, although the expression of Cre is usally limited to a certain cell type, unexpected transient expression of Cre recombinase may occur in the germline or during early development. This is because many genes are normally expressed at low levels in the germline cells during gametes formation so that their transient expression is generally not noticed in these cells. If the promoters of these genes are used to drive Cre expression in transgenic mouse lines, the promoters may dictate the transient expression of Cre recombinase in the germline cells [1]. For example, Tie2-Cre has been used to generate endothelium specific knockout mice [2], and Cre-recombinase activity was detected in the Tie2-Cre mice female germline cells [3]. When using a tissue-specific transgenic Cre mouse lines to perform genetic studies, it is therfore essential to ascertain whether this Cre mouse line has Cre recombinase activity in the germline cells, which serves as a scrutiny step to avoid potential pitfalls caused by unexpected Cre expression.

Myosin heavy chain 6 promoter-driven Cre (Myh6-Cre) mouse line, generated in 1997 by Michael D. Schneider group (hereafter referred to as MDS Myh6-Cre), has been widely used to study gene function in the postnatal heart [4]. Nevertheless, during breeding of Myh6-Cre to Cre-activated Rosa26^fstdTom reporter (tdTom) mouse line [5], we found that a portion of their tdTom offspring had observable red skin, indication of germline activation of the tdTom reporter gene. In this study, we showed that Myh6 gene and its protein product were expressed in male germline cells and confirmed that transgenic and knock-in Myh6-Cre mice have Cre recombinase activity in male germline cells.

Material and methods

Detail materials and methods are available in the supplementary information file.

Mice

All animal procedures were approved by the Institutional Animal Care and Use Committee of Masonic Medical Research Institute.

Results

Male Myh6-Cre +/−;Rosa26^fstdTom mice produce a portion of offspring that universally express tdTom.

MDS Myh6-Cre +/− mice were crossed with Rosa26^fstdTom to create male MDS Myh6-Cre+/−; tdTom/+ mice, which were then crossed with wild type (WT) female mice (Fig. 1A). Surprisingly, some of their offspring had red skin (Fig. 1B). Genotyping results showed that all of the red mice had the tdTom allele and might or might not have the Myh6-Cre allele (Fig. 1C), suggesting that germ line activation of tdTom reporter happened in their parent generation. To test whether tdTom was whole body activated in these red mice, we performed whole-body fluorescent imaging and ex vivo imaging of internal organs. As shown in Fig. 1D and 1E, the skin and internal organs of the normal tdTom/+ mouse displayed no tdTom fluorescence; however, the counterparts of the red tdTom/+ mouse displayed strong tdTom fluorescence. These data suggest that a subset of germ line cells from MDS Myh6-Cre+/− mice have Cre recombinase activity.

Figure 1. tdTom reporter is universally activated in a portion of the offspring derived from male Myh6-Cre;tdTom/+ mice.

A-C. Male MDS Myh6-Cre+/−; tdTom/+ mice generate whole body red offspring. A, breeding scheme that leads to the production of whole body red mice. B, a representative image of a normal tdTom/+ mouse and a red tdTom/+ mouse. C, representative tail DNA genotyping of a litter of offspring derived from male Myh6-Cre+/−; tdTom/+ and female WT mice. Red colored letters indicate red tdTom/+ mice. In these offspring, the presence of red skin is Myh6-Cre independent. D. Whole-body fluorescent imaging. E. Ex vivo fluorescent imaging of different organs. H, heart; S, spleen; K, kidney; P, pancreas, B, brain; Ll, liver; Lu, lung. D and E, fluorescent imaging was performed with Perkin Elmer IVIS Spectrum system. Color range indicates radiant efficiency ([p/s/cm²/sr]/[μW/cm²]). F. Distribution of red tdTom/+ mice in the tdTom/+ offspring of different genotype parents. tdTom/+ mouse numbers were indicated in each pie chart. Red slices represent red tdTom/+ mice; grey slices indicate normal tdTom/+ mice. G. Comparison between observed and expected red tdTom/+ mice occuring frequency in the offspring of indicated mating pairs. Chi-square analysis, ****, P<0.0001. NS, not significant. H. Percentage of red tdTom/+ mice in the total tdTom/+ genotype mice. Results from two different mating strageties were compared. Each dot indicates the red tdTom/+ pups percentage in one litter offspring. Offspring derived from male Myh6-Cre+/−; tdTom/+, n=12 litters; offspring derived from female Myh6-Cre+/−; tdTom/+, n=9 litters. Student t test, ***, P<0.001. J. Summary of the genotype distribution and the red tdTom/+ mice occuring frequency in 22 litters pups. Mouse numbers for each genotype and the number of observed red tdTom/+ mice were listed. G, H, and J, the offspring derived from male MDS Myh6-Cre+/−; tdTom/+ and AUTR Myh6-Cre+/−; tdTom/+ mice were combined.

To further corroborate this phenomenon and investigate the underlying mechanism, we checked the offspring of four different groups of parents (Fig. 1F). In the first group, male MDS Myh6-Cre +/−; tdTom/+ mice were crossed to WT females, and 11 out of 25 of their tdTom/+ pups were red. In the second group, we performed the same genetic labeling experiment with another independent Myh6-Cre transgenic mouse line, which has the same Myh6 promoter but has an additional β-actin untranslated region (AUTR) in the 3 prime side of Cre (hereafter referred as AUTR Myh6-Cre) [6]. Similar with MDS Myh6-Cre +/−;tdTom/+ mice, male AUTR Myh6-Cre +/−;tdTom/+ mice also had red tdTom/+ pups (6 out of 15). Because both MDS and AUTR Myh6-Cre are transgenic lines with the Myh6-Cre cassette randomly inserted into the genome, and both lines share the same Myh6 promoter that drives Cre expression, these observations indicate that the Myh6 promoter rather than transgene locus is responsible for Cre expression in germ line cells. In the third group, to test whether only the male Myh6-Cre germ line cells express Cre, we crossed female MDS Myh6-Cre +/−; tdTom/+ with WT mice. Out of 24 tdTom/+ pups, one had red skin, suggesting that Myh6-Cre has minimum activity in female germ line cells. At last, we used an irrelevant transgenic mouse line Ucp1-Cre, which expresses Cre in brown adipocytes [7], to perform the same genetic labeling experiment, and we did not find any red Ucp1-Cre;tdTom/+ offspring (Fig. 1F).

These observations prompted us to analyze the freqeuncy of red tdTom/+ pups in the offspring-derived from two mating strategies: i) male Myh6-Cre +/−;tdTom/+ cross with WT female; and ii) WT male cross with female Myh6-Cre +/−;tdTom/+. Because male MDS Myh6-Cre +/−; tdTom/+ and male AUTR Myh6-Cre;tdTom/+ generated red tdTom/+ pups to a similar extend, we combined the two data sets together in the following analysis. For the offspring derived from both mating strategies, the expected freqeuncy of red tdTom/+ pups should be zero because Myh6-Cre was not suppose to trigger DNA recombination in the germ line cells. Compared with the expected red tdTom/+ pups frequency, observed red tdTom/+ pups frequency from mating strategy i) and not from mating strategy ii) displayed robust statistic difference (Fig. 1G). Furthermore, in the tdTom/+ offspring derived from mating strategy i) and ii), 40.77% and 1.235% were red, respectively (Fig. 1H). Fig. 1I presented the summary of the pups generated from these two mating strategies.

Altogether, these data suggest that the Myh6 promoter drives Cre expression in a subset of male germ line cells.

Myh6 is expressed in male germ line cells.

If Myh6 promoter drives Cre expression in male germ line cells, we should be able to detect tdTom fluorescence in the testes of Myh6-Cre;tdTom/+ mice. Indeed, ex vivo imaging showed that both MDS Myh6-Cre;tdTom/+ and AUTR Myh6-Cre;tdTom/+ testes, and not the normal tdTom/+ testes had tdTom fluorescence (Fig. 2A). We then examined testes sections for the location of tdTom expression. In normal tdTom/+ testes, no observable tdTom fluorescence was detected. In the Myh6-Cre;tdTom/+ testes, we found strong tdTom fluorescence in the inner regions of some semniferous tubules (Fig. 2B), which are the places for spermatids to differentiate and mature [8].

Figure 2. Myh6 gene and its protein products are expressed in testis.

A-B. Myh6-Cre activates tdTom in testes. A, ex vivo fluorescent imaging of testes. B, representive fluorescent images of testes sections. Scale bar = 50 μm. C. PCR test of Cre-mediated DNA recombination. M, molecular marker. D. qRT-PCR measurement of Myh6 gene expression. Cycle threshhold (Ct) values were used to display Myh6 mRNA levels in heart, testis and liver. 1μg total RNA was used for revers transcription. For each organ, n=3. One-Way ANOVA, *, p<0.05; ***, p<0.001. E. Western blot of testis MYH6. GAPDH was used as loading control. F. Immunofluorescence images of testes sections stained with MYH6 antibody (ABclonal). IgG stained testes sections were used as negative controls. Phalloidin (Cayman Chemical) was used to label spermatids. Scale bar = 50 μm. G–I. Single cell RNA sequencing analysis of Myh6 expression in different testis cell types. G, Myh6 levels in different testis cell types. H, Myh6 levels in different germ cell types. SPG, spermatogonia; Prelap, preleptotene cells; ES, elongating spermatids; SCytes, spermatocytes; STids, post-meiotic haploid round spermatids. I, Myh6 levels in four types of spermatogonia cells. G-I, Myh6 expression data were retrieved from published adult mouse testis single cell RNA sequencing results (GSE112393). J. Summary of the current finding.

In the Myh6-Cre transgenic mice, the Cre gene expression may not fully recapitulate the physiological expression pattern of endogenous Myh6 gene, and therefore the current observation may be due to transgene-caused ectopic Cre expression in the testes. To test this possibility, we crossed the newly published Myh6-Cre knockin (KI) mouse [9] with Rosa26^fstdTom to generate male Myh6-Cre (KI);tdTom/+ mice. In the testes of Myh6-Cre (KI);tdTom/+ mice, we also observed the activation of tdTom reporter (Fig. 2B). We further corroborated our observation by confirming the presence of Cre-meidated DNA recombination events in the testes. As shown in Fig. 1C, tdTom allele DNA recombination occured in both the hearts and the testes of the adult male Myh6-Cre (Tg);tdTom/+ and Myh6-Cre (KI);tdTom/+ mice. To exclude that this Myh6-Cre mediated testes DNA recombination was specific for tdTom allele, we extracted DNA from the testis of our recently published Myh6-Cre;Vgll4^fl/fl (Vgll4^cKO) mice [10], and confirmed that Myh6-cre caused Vgll4 flox allele recombination in both the heart and the testes of the the Vgll4^cKO mice (Suppl. Fig. 1A). These data together suggest that Myh6 promoter is active in the testes and that it drives Cre expression in male germ line cells in both transgenic and knockin Myh6-Cre mouse lines.

Driven by these observations, we further asked whether MYH6 protein itself was expressed in the testes. In the original publication of MDS Myh6-Cre, the LacZ reporter assay indicates that these mice do not have Cre activity in their livers [4]. We therefore used mouse liver as negative control and heart as positive control to further examine the expression of MYH6 in testis. The qRT-PCR results showed that testis had significantly higher Myh6 mRNA level than the liver (Fig. 2D), and western blot showed that MYH6 protein was expressed in the heart and testis but not in the liver (Fig. 2E). We further performed MYH6 immunofluorescence staining with the testes and found that MYH6 protein was enriched in the inner layer of seminiferous tubules (Fig. 2F), and that MYH6 signals partially overlaped with the Phalloidin-labeled spermatids.

To fully understand which testis cell types express Myh6, we examined the Myh6 gene expression pattern with our previously published adult mouse testis single cell RNA sequencing data [11]. In all the 11 major testis cell types, Myh6 showed highest expression levels in the interstitial Leydig cells, followed by spermatogonia stem cells (Fig. 2G). Based on unsupervised clustering algorithm we previously identified 12 germ cell states, with the clusters 1–3 showing discrete states and the following 9 clusters displaying continuous differentiation trajectories [11]. In these 12 germ cell clusters, Myh6 was enriched in cluster 1 and 3 (Fig. 2H), which represent spermatogonia stem cells (SPG) and meiotic germ line cells (preleptotene cells), respectively. Within the SPG cells, four subgroups (SPG1-4) had been identified based on their differentiation status, and Myh6 was enriched in the undifferentiated SPG-1 cells and the SPG-2 cells which express early differentiating marker genes (Fig. 2I). These single cell RNA sequencing data further confirmed the expression of Myh6 in testis and suggest that Myh6 transcript is enriched in the interstitial Leydig cells and the undifferentiated SPG cells.

Different from male Myh6-Cre+/−;tdTom/+ mice, the female Myh6-Cre+/−;tdTom/+ mice rarely had red tdTom/+ offspring (Fig. 1F), suggesting that Myh6 promoter does not drive Cre expression in the female germ line cells. If this hypothesis is true, Myh6 gene and its protein product should not be expressed in ovary. We then performed MYH6 immunofluorescence staining with adult mouse ovary and found that MYH6 was not detectible in any ovary cell types (Suppl. Fig. 1B). The absence of MYH6 in overy cells, together with the fact that female Myh6-Cre +/−;tdTom/+ mice rarely generate red tdTom/+ offspring, strongly suggest that Myh6 promoter does not drive Cre expression in female germ line cells. The main findings of this study were summarized in Fig. 2J.

Discussion

The current study demonstrates that trangegenic Myh6-Cre triggers the excision of floxed allele in a sub-population of male germ line cells. When studying a gene of interest (GOI), if male Myh6-Cre+/−; GOI^flox/flox mice were used to cross with female GOI^flox/flox, a significant portion of the offspring (~ 40%) might be whole body heterozygous (GOI^flox/−). In most of the cases, the GOI is recessive and therefore no phenotypes can be detected in the undesired GOI^flox/− mice; however, the Myh6-Cre-mediated cardiac specific study can be severely delayed or flawed if the GOI has dominant traits under nomal or stress conditions. For example, several years ago we performed a cardiac-specific YAP gain-of-function study, in which we generated Myh6-Cre [4]; ROSA26^{fs.rtTA/fs.rtTA} [12]; Col1A1^TetOYAP [13] mice (YAP^GOF) that express an activated YAP (YAP S127A) in the heart under the control of doxycycline [14]. We maintained this YAP^GOF line by crossing Myh6-Cre+/−;ROSA26^{fs.rtTA/fs.rtTA} with ROSA26fs.rtTA/fs.rtTA;Col1A1TetO-YAP1/TetO-YAP1. Surprisingly, around half of the YAPGOF mice developed acute diarrhea several days after doxycycline (Dox) treatment and were excluded from our study, which severely delayed the project progression. Later, Barry et al. reported that Dox-induced and intestinal epithelium-specific activation of YAP sickened the mice by causing rapid loss of proliferating crypts [15]. Together with our current findings, these two observations suggest that a big portion of the YAP^GOF mice had the rtTA allele universally activated, which drove YAP expression in the intestine and caused acute diarrhea when Dox was applied.

Here, we also oberved that Myh6-Cre rarely triggers flox allele excision in female germ line cells. Therefore, when the GOI has dominant traits, we recommend using the dams to carry Myh6-Cre for breeding heart-specific conditional knock out mice.

In summary, we discovered that MYH6 is expressed in testis and that Myh6 promoter-driven Cre excises floxed DNA fragments in a subset of male germ line cells. Cautions should be taken to avoid potential pitfalls if Myh6-Cre is used to study the floxed alleles that are haploinsufficiency or have dominant traits.