Actl7b deficiency leads to mislocalization of LC8 type dynein light chains and disruption of murine spermatogenesis.

Actin-related proteins (Arps) are classified according to their similarity to actin and are involved in diverse cellular processes. ACTL7B is a testis-specific Arp, and is highly conserved in rodents and primates. ACTL7B is specifically expressed in round and elongating spermatids during spermiogenesis. Here, we have generated an Actl7b-null allele in mice to unravel the role of ACTL7B in sperm formation. Male mice homozygous for the Actl7b-null allele (Actl7b-/-) were infertile, whereas heterozygous males (Actl7b+/-) were fertile. Severe spermatid defects, such as detached acrosomes, disrupted membranes and flagella malformations start to appear after spermiogenesis step 9 in Actl7b-/- mice, finally resulting in spermatogenic arrest. Abnormal spermatids were degraded and levels of autophagy markers were increased. Co-immunoprecipitation with mass spectrometry experiments identified an interaction between ACTL7B and the LC8 dynein light chains DYNLL1 and DYNLL2, which are first detected in step 9 spermatids and mislocalized when ACTL7B is absent. Our data unequivocally establish that mutations in ACTL7B are directly related to male infertility, pressing for additional research in humans.

Specificity and Redundancy of Profilin 1 and 2 Function in Brain Development and Neuronal Structure.

Profilin functions have been discussed in numerous cellular processes, including actin polymerization. One puzzling aspect is the concomitant expression of more than one profilin isoform in most tissues. In neuronal precursors and in neurons, profilin 1 and profilin 2 are co-expressed, but their specific and redundant functions in brain morphogenesis are still unclear. Using a conditional knockout mouse model to inactivate both profilins in the developing CNS, we found that threshold levels of profilin are necessary for the maintenance of the neuronal stem-cell compartment and the generation of the differentiated neurons, irrespective of the specific isoform. During embryonic development, profilin 1 is more abundant than profilin 2; consequently, modulation of profilin 1 levels resulted in a more severe phenotype than depletion of profilin 2. Interestingly, the relevance of the isoforms was reversed in the postnatal brain. Morphology of mature neurons showed a stronger dependence on profilin 2, since this is the predominant isoform in neurons. Our data highlight redundant functions of profilins in neuronal precursor expansion and differentiation, as well as in the maintenance of pyramidal neuron dendritic arborization. The specific profilin isoform is less relevant; however, a threshold profilin level is essential. We propose that the common activity of profilin 1 and profilin 2 in actin dynamics is responsible for the observed compensatory effects.

Severe protein aggregate myopathy in a knockout mouse model points to an essential role of cofilin2 in sarcomeric actin exchange and muscle maintenance.

Mutations in the human actin depolymerizing factor cofilin2 result in an autosomal dominant form of nemaline myopathy. Here, we report on the targeted ablation of murine cofilin2, which leads to a severe skeletal muscle specific phenotype within the first two weeks after birth. Apart from skeletal muscle, cofilin2 is also expressed in heart and CNS, however the pathology was restricted to skeletal muscle. The two close family members of cofilin2 - ADF and cofilin1 - were co-expressed in muscle, but unable to compensate for the loss of cofilin2. While primary myofibril assembly and muscle development were unaffected in cofilin2 mutant mice, progressive muscle degeneration was observed between postnatal days 3 and 7. Muscle pathology was characterized by sarcoplasmic protein aggregates, fiber size disproportion, mitochondrial abnormalities and internal nuclei. The observed muscle pathology differed from nemaline myopathy, but showed combined features of actin-associated myopathy and myofibrillar myopathy. In cofilin2 mutant mice, the postnatal expression pattern and turnover of sarcomeric α-actin isoforms were altered. Levels of smooth muscle α-actin were increased and remained high in developing muscles, suggesting that cofilin2 plays a crucial role during the exchange of α-actin isoforms during the early postnatal remodeling of the sarcomere.

Dual reporter approaches for identification of Cre efficacy and astrocyte heterogeneity.

Gene inactivation reporters are powerful tools to circumvent limitations of the widely used Cre/loxP system of conditional mutagenesis. With new conditional transgenic mouse lines expressing the enhanced cyan fluorescent protein (ECFP) instead of connexin43 (Cx43) after Cre-mediated recombination, we demonstrate dual reporter approaches to simultaneously examine astrocyte subpopulations expressing different connexins, identify compensatory up-regulation within gene families, and quantify Cre-mediated deletion at the allelic level. Analysis of a newly generated Cx43 knock-in ECFP mouse revealed an unexpected heterogeneity of Cx43-expressing astrocytes across brain areas.

Connexin32 can restore hearing in connexin26 deficient mice.

Functional gap junction channels composed of certain connexin proteins are essential for the function of the cochlea. Homozygous deficiency in the Gjb2 (mice) or GJB2 (human) gene coding for connexin26 (Cx26) in the cochlea leads to hearing impairment in mice and humans, respectively. Here we have studied the functional equivalence of Cx26 and connexin32 (Cx32) isoforms in the cochlea. We analyzed a conditional mouse mutant in which the Gjb2 coding DNA was exchanged by LacZ DNA coding for the reporter protein beta-galactosidase. This allowed us to follow the unrestricted and cell type specific expression of Gjb2 promoter activity. After inner ear specific, Otogelin-Cre recombinase mediated deletion of the loxP-site-flanked LacZ coding DNA, transcription of the Gjb1 gene, coding for Cx32 was activated by the Gjb2 promoter. Interbreeding of these mice with conditional Gjb2 null mice resulted in animals in which Cx32 instead of Cx26 protein is expressed in the non-sensory epithelial network of the cochlea. When we analyzed the auditory function in these mice, we found that the expression of Cx32 protein is sufficient to support hearing in the absence of Cx26. Thus Cx32 can functionally replace Cx26 in the mouse cochlea resulting in almost normal hearing.

Ablation of gap junctional communication in hepatocytes of transgenic mice does not lead to disrupted cellular homeostasis or increased spontaneous tumourigenesis.

Gap junctions between murine hepatocytes are composed of two subunit proteins, connexin26 (Cx26) and connexin32 (Cx32). Previously, we found increased formation of chemically induced liver tumours but no increase in spontaneous development of preneoplastic hepatic foci in mice that lacked Cx32 and expressed decreased amounts of Cx26. In order to clarify this tumour-suppressive effect and to overcome embryonic lethality of constitutive Cx26-deficient mice, cell type-specific targeting of the Cx26 gene was performed. Mice with loxP-flanked Cx26 coding DNA were crossed with mice expressing the Cre recombinase exclusively in hepatocytes. Progeny mice lacking Cx26 in the liver were viable and fertile with no obvious signs of phenotypic alterations. To generate mice that totally lack gap junctional intercellular coupling, these mice were crossed with constitutive Cx32-deficient mice. We found no increase in spontaneously induced liver tumour formation in Cx26 and double deficient Cx26/Cx32 mice. Occasionally, double deficient livers exhibited morphological alterations, like amyloidosis, and a slightly increased basal proliferation rate of hepatocytes. Although the absence of gap junction channels led to altered expression of adhesion-related proteins like E-cadherin and actin, microarray analyses of total liver transcripts yielded only few differences between Cx26-deficient and double deficient livers compared to control samples. Our results suggest that total lack of gap junctional communication due to hepatocytic ablation of Cx26 and Cx32 does not drastically alter basal hepatocytic function and does not lead to increased spontaneous liver tumour formation.