In Adult Skeletal Muscles, the Co-Receptors of Canonical Wnt Signaling, Lrp5 and Lrp6, Determine the Distribution and Size of Fiber Types, and Structure and Function of Neuromuscular Junctions

Canonical Wnt signaling is involved in skeletal muscle cell biology. The exact way in which this pathway exerts its contribution to myogenesis or neuromuscular junctions (NMJ) is a matter of debate. Next to the common co-receptors of canonical Wnt signaling, Lrp5 and Lrp6, the receptor tyrosine kinase MuSK was reported to bind at NMJs WNT glycoproteins by its extracellular cysteine-rich domain. Previously, we reported canonical Wnt signaling being active in fast muscle fiber types. Here, we used conditional Lrp5 or Lrp6 knockout mice to investigate the role of these receptors in muscle cells. Conditional double knockout mice died around E13 likely due to ectopic expression of the Cre recombinase. Phenotypes of single conditional knockout mice point to a very divergent role for the two receptors. First, muscle fiber type distribution and size were changed. Second, canonical Wnt signaling reporter mice suggested less signaling activity in the absence of Lrps. Third, expression of several myogenic marker genes was changed. Fourth, NMJs were of fragmented phenotype. Fifth, recordings revealed impaired neuromuscular transmission. In sum, our data show fundamental differences in absence of each of the Lrp co-receptors and suggest a differentiated view of canonical Wnt signaling pathway involvement in adult skeletal muscle cells.


Introduction
In vertebrates, WNT glycoproteins are a family of signaling molecules also involved in myogenesis and, through both canonical and non-canonical Wnt pathways, regulate muscle formation and maintenance of adult tissue homeostasis [1]. Canonical Wnt signaling is initiated by WNT glycoproteins which bind to Frizzled (FZD) and low-density lipoprotein receptor-related protein (Lrp) receptor complex, thereby leading to the inactivation of glycogen synthase kinase 3β (GSK3β) through dishevelled (DSH). In the absence of WNT stimulation, β-catenin forms a destruction complex with adenomatosis polyposis coli (APC), AXIN1/AXIN2 and GSK3β [2]. Phosphorylation of β-catenin by CK1 and GSK3β causes ubiquitination and proteasome-mediated degradation of β-catenin. WNT stimulation results in the activation of DSH, which leads to phosphorylation-dependent recruitment of AXIN1/AXIN2 to the Lrp5/6 receptors and disassembly of the β-catenin destruction complex. Stabilized β-catenin accumulates in the cytoplasm and translocates to the nucleus. There, it complexes with T cell factor/lymphoid enhancer factor (TCF/LEF) transcription factors and acts as a transcriptional coactivator to induce the context-dependent expression of WNT/β-catenin target genes [3].
Interestingly, MuSK contains in its extracellular region a Frizzled-like domain (cysteinerich domain [CRD]) mediating its interaction with several WNTs, including WNT4, WNT11, and WNT9a in vitro [35][36][37]. Activation of the MuSK-Lrp4 complex regulates the prepatterning step, before muscle innervation, during which AChRs begin to aggregate in a central synaptic region of the muscle [38][39][40]. Moreover, in vivo knockdown of WNT4 and WNT11 affects muscle prepatterning and axon guidance, indicating a role for Wnt signaling [35,36,41]. Upon innervation, the MuSK-Lrp4 complex is further stimulated by neural AGRIN, which induces multiple signaling pathways leading to clustering and remodeling of aneural AChR clusters [42,43]. In addition to their role in prepatterning, WNTs have been shown to regulate AGRIN-induced AChR clustering in vitro [32,44].
Altogether, the role of Lrp5 and Lrp6 appears almost not understood in skeletal muscle cells, especially the way in which and whether WNT ligands signal by Lrp5 and Lrp6, or by MuSK CRD. The muscle-specific ablation of Lrp5 and Lrp6 using conditional mouse models might help to understand the role of canonical Wnt and its link to Hippo signaling during myogenesis, in adult myofibers and at NMJs. Here, we conditionally knocked out Lrp5, Lrp6, or both in the skeletal muscle lineage. While the individual knockouts were viable, the double knockout mice died prenatally due to ectopic expression of the Cre recombinase. We characterized both single knockout mice and identified very divergent roles for Lrp5 and Lrp6 in skeletal muscle cells.

Mouse Procedures and Genotyping
Mouse experiments were performed in accordance with animal welfare laws and approved by the responsible local committees (animal protection officer, Sachgebiet Tierschutzangelegenheiten, FAU Erlangen-Nürnberg, AZ: I/39/EE006 and TS-07/11) and government bodies (Regierung von Unterfranken). Floxed Lrp5 and Lrp6 mice were kindly provided by Dr. Gabriela G Loots. HSA-Cre mice were described before [45,46]. Mice were housed in cages that were maintained in a room with temperature 22 ± 1 • C and relative humidity 50-60% on a 12 h light/dark cycle. Water and food were provided ad libitum. Mouse mating and genotyping were performed as previously described [47]. Mice were genotyped by PCR analysis of ear biopsies DNA [45]. Mendelian frequencies were calculated using a simple MS EXCEL workbook [48]. Muscle force of the mice was measured with all four limbs by a Grip Strength Test Meter (Bioseb) [49]. All mice used were 4-7 month of age. No haploinsufficiency was detected in heterozygote floxed Lrp5/6 mice independent from the presence or absence of HSA-Cre. Throughout the manuscript, the term "control" is used for homozygous floxed Lrp5/6 mice without HSA-Cre, and "mutant" is used for homozygous floxed Lrp5 or Lrp6 mice with HSA-Cre.

RNA Extraction, Reverse Transcription, PCR
Total RNA was extracted from mouse tissues with TRIzol reagent (Thermo Fisher Scientific, Schwerte, Germany, 15596026) [45] and reverse transcribed with M-MuLV Reverse Transcriptase (New England Biolabs, Frankfurt am Main, Germany, M0253) according to the manufacturer's instructions. cDNAs were used with mouse-specific primers (Supplementary Table S1) for quantitative PCR reactions using the PowerUp SYBR Green Master Mix (Thermo Fisher Scientific, Schwerte, A25743) and the C1000 Thermal Cycler with the CFX96 Real-Time PCR Detection System (Bio-Rad) according to the manufacturer's instructions. After the PCR run, sizes of amplified DNA products were verified by agarose gel electrophoresis. Ct values of the genes of interest were normalized to Ct values of the internal control (Rpl8 gene) (normalized expression = 2 −∆CT ) or additionally related to the control sample (fold change = 2 −∆∆CT ) [50,51].
For quantitative 3D morphometrical imaging, mice soleus or diaphragm muscle was dissected and fixed in 2% PFA for 2 h at 4 • C. For detection of AChRs, muscle bundles containing 5-10 fibers were prepared and stained with BTX (Invitrogen, Darmstadt, Germany, BTX; 1:2.500) for 1 h at room temperature. Stained bundles were washed three times for 5 min in phosphate buffered saline (PBS) and embedded in Mowiol. Then, 3D images of NMJs were taken with a 63 × 1.4 numerical aperture oil objective (Zeiss Examiner Z1, Carl Zeiss MicroImaging, Göttingen, Germany) at 55 ms exposure time. Images were deconvoluted and analyzed using different modules in AxioVision software (ZEISS AxioVision Release 4.9, Carl Zeiss MicroImaging, Göttingen, Germany). The following parameters were determined for each NMJ: volume, surface, grey sum, grey mean and number of fragments. For each genotype, more than 50 NMJs were analyzed [49]. Fluorescence grey sum depicts the total of grey values for each pixel in the acquired image, whereas fluorescence grey mean is fluorescence grey sum divided by the number of pixels in the acquired image [53].
For X-gal staining, whole soleus muscles were fixed for 1 h in PFA, 3 × 5 min washed in PBS, and incubated in X-Gal staining solution at 37 • C, consisting of 0.75 mg/mL X-Gal, 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 0.01% sodium deoxycholate, 0.02% NP-40, 2 mM magnesium chloride and 20 mM Tris in phosphate buffered saline. Stained muscle tissues were visualized using the Zeiss Discovery V8 stereo microscope equipped with an AxioCam HRm camera and the Zeiss ZEN blue software Release 3.6 (Carl Zeiss MicroImaging, Göttingen, Germany). For quantification of blue staining, the "Color Deconvolution 2" Fiji plugin [54,55] together with Fiji was used to separate the blue image. A threshold for intensity has been attributed to the blue image, which was correlated with the amount of X-Gal staining for each image. After selecting the image of the whole muscle by the polygon button, the integrated density was measured.

Nerve Muscle Preparation and Electrophysiological Recordings
Diaphragm-phrenic nerve preparations were maintained ex vivo in Liley's solution gassed with 95% O 2 , 5% CO 2 at room temperature [56]. The recording chamber had a volume of approximately 1 mL and was perfused at a rate of 1 mL/min. The nerve was drawn up into a suction electrode for stimulation with pulses of 0.1 msec duration. The preparation was placed on the stage of a Zeiss Axio Examiner Z1 microscope (Carl Zeiss Mi-croImaging, Göttingen, Germany) fitted with incident light fluorescence illumination with filters for 547 nm/red (Zeiss filter set 20) fluorescing fluorophore (Carl Zeiss MicroImaging, Göttingen, Germany). The compound muscle action potential (CMAP) was recorded using a micropipette with a tip diameter of approximately 10 µm filled with bathing solution. The electrode was positioned so that the latency of the major negative peak was minimized. The electrode was then positioned 100 µm above the surface of the muscle, and CMAP was recorded. Trains of repetitive nerve stimulations (5 Hz) were performed at 2 min intervals, and the ratio of CMAP amplitudes (mean (20th-25th)/2nd) was calculated [49,57]. To block muscle action potentials so that EPPs (endplate potentials) and EPCs (endplate currents) could be recorded at 1 Hz for 100 s [58,59], µ-conotoxin GIIIB (µ-CTX, 2 µM; Peptide Institute, Osaka, Japan) was added to Lilly's solution. EPPs were recorded at 5 Hz for 5 s and at 20 Hz for 10 s. Decrements of EPPs were calculated employing the mean of the first and the last five recordings. Concurrently, clustered AChRs at NMJs were labeled by adding 0.5 × 10 −8 M of BTX (Life Technologies, Darmstadt, Germany) to the same Lilly solution.
In some experiments, the effect of the toxin wore off after 1-2 h, and contractions resumed in response to nerve stimulation. These preparations were then exposed to the toxin for the second time. Two intracellular electrodes (resistance 10-15 MΩ) were inserted within 50 µm of the NMJs under visual inspection [53]. Current was passed through one electrode to maintain the membrane potential within 2 mV of −75 mV, while voltage transients were recorded with the other. Signals were amplified by an Axoclamp 900 A and digitized at 40 kHz by a Digidata 1440 A under the control of pCLAMP 10 (Molecular Devices, Sunny Vale, CA, USA). Voltage records were filtered at 3 kHz and current records at 1 kHz (8-pole Bessel filter). Current transients were recorded using the two-electrode voltage-clamp facility of the Axoclamp 900 A. Clamp gains were usually 300-1000, reducing the voltage transients to <3% of their unclamped amplitudes. At most NMJs, 50-100 spontaneous quantal events were recorded during a period of 1 min. Records were analyzed using pCLAMP 10. Spontaneous events were extracted using the "template search" facility and edited by eye to remove obvious artifacts. Events recorded from each NMJ were averaged, and the amplitude and frequency were determined [49].

Statistical Analysis
Statistical analysis was performed in GraphPad Prism 9 (GraphPad software, San Diego, USA) as indicated. Data are presented as mean values, and the error bars indicate S.D. The number of biological replicates per experimental variable (n) is usually n > 5 or as indicated in the figure legends. For all data with mice, a minimum of 3 mice were studied. The significance was calculated by unpaired 2-tailed student t test, or as indicated by the figure legends, and provided as real p values that are believed to be categorized for different significance levels, **** p < 0.0001, *** p < 0.001, ** p < 0.01, or * p < 0.05.

Differential Impairment of Viability and Body Weight in Conditional Skeletal Muscle-Specific Lrp5, Lrp6, or Double Knockout Mice
The overall topology of the canonical Wnt co-receptors Lrp5 and Lrp6 is very similar and they are believed to be mandatory for induction of canonical Wnt signaling, but their Wnt signaling capabilities are reported being not equivalent [6]. To study their role regarding canonical Wnt signaling in adult muscle fibers and at NMJs, we generated conditional Lrp5 and Lrp6 knockout mice by breeding them with HSA-Cre mice, which express Cre recombinase under the control of the human skeletal actin (HSA) promoter [60]. We used two previously described conditional floxed alleles for Lrp5 and Lrp6 [61]. PCRbased genotyping analysis ascertained identification of heterozygous and homozygous floxed alleles ( Figure 1A). Accordingly, Lrp5 and Lrp6 protein amounts were significantly reduced in hind limb muscle lysates of conditional Lrp5 or Lrp6 knockout mice ( Figure 1B). Note that not all offspring genotypes followed mendelian distribution ( Figure 1E). While individual single and double mutant mice were detectable at early embryonic stage, double knockout mice died around embryonic day E13. It appears rather unlikely that midembryonal death of conditional double knockout mice is related to a skeletal muscle phenotype, but the observed lethality might be related to previously reported ectopic expression of the Cre recombinase [60] and demands to be investigated using different muscle-specific Cre mice. Here, the analysis of the double knockout is beyond the scope of this manuscript. We focused on characterization of the conditional single knockout mice in adult skeletal muscle fibers. At adulthood, body weight of Lrp6 knockout mice was slightly, but significantly, lower in comparison with control mice ( Figure 1C). Muscle strength of individual Lrp5 or Lrp6 knockout mice appeared not to be different in comparison with control mice ( Figure 1D). In total, these data demonstrate little significant reduction of body weight in conditional Lrp6 knockout mice, and mid-embryonal lethality of double knockout mice. slightly, but significantly, lower in comparison with control mice ( Figure 1C). Muscle strength of individual Lrp5 or Lrp6 knockout mice appeared not to be different in comparison with control mice ( Figure 1D). In total, these data demonstrate little significant reduction of body weight in conditional Lrp6 knockout mice, and mid-embryonal lethality of double knockout mice. (B) Images of X-ray exposure of Western blot membrane of muscle lysates (GAPDH) or immunoprecipitations of Lrp5 or Lrp6 using muscle hind limb lysates of control and mutant mice as indicated. (C) The graph reflects the body weight of control and single knockout mice at adulthood (4-7 months of age). Note that in comparison with controls, body weight of conditional Lrp6 knockouts is slightly reduced (n = 8-16 mice per genotype). (D) Muscle force is presented by a graph summarizing force per weight for control and conditional single knockout mice. (E) Distribution of offspring of conditional Lrp5 and Lrp6 knockout mice (n = 8-16 mice per genotype). Note that while conditional single knockout mice are born in agreement with mendelian distribution, double knockout die at embryonic stage. All labeling of graph bars represents genotype of mice as indicated in (D). * p <0.05.

Oxidative Metabolism Is Reduced in the Skeletal Muscles of Conditional Lrp5 or Lrp6 Knockout Mice
Canonical Wnt signaling was identified in type IIa and IIx adult muscle fibers and being linked to muscle fiber cross-sectional area [22]. To obtain a first impression of any impairments of muscle cross section histology in conditional Lrp5 or Lrp6 knockout mice, we performed typical hematoxylin and eosin staining using hind limb muscles, gastrocnemius and soleus (Figure 2A). No apparent difference of histology was detectable between control and conditional Lrp5 or Lrp6 knockout mice (Figure 2A). We also did not

Oxidative Metabolism Is Reduced in the Skeletal Muscles of Conditional Lrp5 or Lrp6 Knockout Mice
Canonical Wnt signaling was identified in type IIa and IIx adult muscle fibers and being linked to muscle fiber cross-sectional area [22]. To obtain a first impression of any impairments of muscle cross section histology in conditional Lrp5 or Lrp6 knockout mice, we performed typical hematoxylin and eosin staining using hind limb muscles, gastrocnemius and soleus (Figure 2A). No apparent difference of histology was detectable between control and conditional Lrp5 or Lrp6 knockout mice (Figure 2A). We also did not observe any signs of myopathy in the individual conditional knockout mice, as the number of fibers with centrally located nuclei or apoptotic cells was similar in the muscles of control and conditional knockout mice ( Figure 2B; data not shown). Cytochrome oxidase (COX) histochemical staining of muscle cross sections typically labels fibers with high mitochondrial content, such as slow fiber types (type I) or fast oxidative type fibers (type IIa), with a darker color. At first glance, we did not observe any differences visually inspecting COX-stained hind limb cross sections (Figure 2A). We employed color deconvolution using Fiji software to quantify COX staining. Interestingly, more COX staining was detected in conditional Lrp6 knockout muscle fibers type I ( Figure 2C) and gradually less COX activity was quantified in bright-colored fibers of soleus muscle, most likely type IIa fibers, of conditional Lrp6 knockout mice in comparison with controls ( Figure 2D). observe any signs of myopathy in the individual conditional knockout mice, as the number of fibers with centrally located nuclei or apoptotic cells was similar in the muscles of control and conditional knockout mice ( Figure 2B; data not shown). Cytochrome oxidase (COX) histochemical staining of muscle cross sections typically labels fibers with high mitochondrial content, such as slow fiber types (type I) or fast oxidative type fibers (type IIa), with a darker color. At first glance, we did not observe any differences visually inspecting COX-stained hind limb cross sections (Figure 2A). We employed color deconvolution using Fiji software to quantify COX staining. Interestingly, more COX staining was detected in conditional Lrp6 knockout muscle fibers type I ( Figure 2C) and gradually less COX activity was quantified in bright-colored fibers of soleus muscle, most likely type IIa fibers, of conditional Lrp6 knockout mice in comparison with controls ( Figure 2D). For color deconvolution, DAB filter was used and a threshold of 0 to 50 generated a grey scale image. Raw integrated density was measured and normalized by total fiber area. All labeling of graph bars represents genotype of mice as indicated in (D). *** p < 0.001, * p < 0.05.

The Numbers and Cross-Sectional Areas of Skeletal Muscle Fibers Are Altered in Conditional Lrp5 or Lrp6 Knockout Mice
Previously, our lab detected active canonical Wnt signaling in type IIa and type IIx muscle fibers of the hind limbs and in diaphragms [22]. We asked whether the same fiber types are impaired in conditional Lrp5 or Lrp6 knockout mice. Using cross sections and immunofluorescence stainings with myosin heavy chain-specific antibodies of hind limb muscles of the control and mutant mice several differences were detected ( Figure 3). Unexpectedly, not the same fiber types being equipped with high activity of canonical Wnt signaling were compromised in conditional Lrp5 or Lrp6 knockout muscle fibers. Previously, low but detectable canonical Wnt signaling activity was analyzed in soleus muscles in comparison with other hind limb muscles where significantly more canonical Wnt signaling activity was detectable [22]. First, the number of slow type I fibers increased in conditional Lrp6, but not in Lrp5 knockout mice ( Figure 3A,B). Second, the number of fast type IIa fibers decreased in conditional Lrp6, but not in Lrp5 knockout mice ( Figure 3A,B). Third, cross-sectional areas of slow type I fibers were significantly lower in conditional Lrp6, but not changed in conditional Lrp5 knockout mice ( Figure 3A,C). Fourth, cross-sectional areas of type IIa fibers were reduced in both, conditional Lrp5 and Lrp6, knockout mice ( Figure 3A,C). Altogether, these data demonstrate changes of muscle fiber type distribution and cross-sectional areas, mainly for conditional Lrp6 knockout mice.

The Absence of Lrp5 or Lrp6 in Skeletal Muscle Fibers of Conditional Knockout Mice Determines the Differential Expression of Myogenic and Synaptic Markers
Canonical Wnt signaling is employing transcriptional co-activator β-catenin to recruit members of the TCF/LEF transcription factor family for target gene expression. One such target gene is AXIN2. We wondered how gene expression might be modulated in absence of the canonical Wnt signaling co-receptors Lrp5 or Lrp6. We decided to make use of a previously published canonical Wnt signaling reporter mouse model, Axin2-LacZ [18,22]. By breeding, the Axin2-lacZ reporter was combined with conditional Lrp5 or Lrp6 knockout in adult mice and skeletal muscles of these mice were compared with control mice. Canonical Wnt signaling activity was monitored by blue-colored staining of adult muscle fibers in the extensor digitorum longus and soleus muscles of the mice ( Figure 4A), a strategy which was described previously [22]. Employing color deconvolution to quantify the degree of blue stain, indeed, blue-stained adult muscle fibers appeared gradually less-colored in conditional Lrp5 and Lrp6 knockout mice in comparison with controls ( Figure 4B). Next, we examined different regulation of several myogenesis and synaptic genes in conditional Lrp5 or Lrp6 knockout mice. Total RNA was extracted from diaphragm and extensor digitorum longus muscles of relevant mice and used for qPCR experiments. First, we analyzed transcript amounts of typical common myogenic markers such as Pax7, Myod1 and Myog. We did not detect any gross change of transcript amount for these three markers; Myod1 was a little decreased in extensor digitorum longus from conditional Lrp5 knockout mice ( Figure 4C). Second, we looked for gene expression levels of differentiated myofiber markers, such as myosin heavy chain genes Myh2, Myh3 and Myh7 ( Figure 4D). Interestingly, we detected several modulations regarding transcript amounts of these markers; prominently, Myh7 was changed in both muscles in conditional Lrp6 knockout mice ( Figure 4D). Third, we analyzed synaptic genes Chrna1, Chrng, and Dok7 ( Figure 4E). While Chrng was not changed in the mutants, in comparison to control, Chrna1 transcript amount was upregulated in both conditional Lrp knockout muscles ( Figure 4E). Fourth, by investigating transcript amounts of signaling path members we determined that less transcript amount of a typical target gene of canonical Wnt signaling, Axin2, was detected ( Figure 4F). The transcript level of Ctnnb1 was not different in both muscles of conditional Lrp5 and Lrp6 knockout mice in comparison to controls ( Figure 4F). Transcript amounts of Hippo path members Yap1 and Wwtr1 were not much different; a little reduction of Yap1 transcript amount was detected in extensor digitorum longus muscles of conditional Lrp6 knockout mice ( Figure 4G). Transcript amounts of typical Hippo target genes were mostly not changed; Ankrd1 was a little reduced in extensor digitorum longus muscles of conditional Lrp6 knockout mice ( Figure 4H). In summary, myogenic and synaptic transcriptome of conditional Lrp5 or Lrp6 knockout muscles appeared to be modulated, but partly differently regulated between conditional Lrp5 or Lrp6 knockout mice.

The Endplate Band Widths of Conditional Lrp5 and Lrp6 Knockout Mice Are Enlarged and the Neuromuscular Junctions of Conditional Lrp6 Knockout Mice Are Fragmented and Accompanied by Impaired Neuromuscular Transmission
Active canonical Wnt signaling is believed to play a role at NMJs [22]. Hence, we looked for NMJs of conditional Lrp5 or Lrp6 knockout mice in comparison with controls. We dissected diaphragm and soleus muscles and stained them with BTX ( Figure 5A,B). Endplate band width in conditional Lrp5 and Lrp6 knockout mice gradually increased in comparison with controls ( Figure 5C). After 3D imaging of individual NMJs of soleus fibers, we detected several structural changes ( Figure 5B,D,F-I). We quantified individual BTX-stained NMJ 3D images from dissected soleus fibers in an automated high-throughput fashion. We started by looking for NMJ fragmentation which is believed to represent the quality of NMJs, because pathologies or aging are known to increase the number of fragments that NMJs are composed of [28]. In comparison with controls, soleus of muscle fibers of conditional Lrp5 or Lrp6 knockout mice, and also diaphragm muscle fibers harbored fragmented NMJs ( Figure 5B,D,E). NMJs of conditional Lrp6 knockout mice were more fragmented than those of conditional Lrp5 knockout mice ( Figure 5B,D,E). Quantitative analysis of the 3D imaging data revealed changes in volume and surface area of BTX-stained NMJs of conditional Lrp6 knockout mice in comparison with controls ( Figure 5F,G). While the percentage of NMJs composed of less than five fragments decreased gradually in conditional Lrp5-or Lrp6-deficient soleus muscle fiber bundles in comparison with controls, the percentage of NMJs with six to ten fragments significantly increased in conditional Lrp5, and even more in conditional Lrp6 knockout mice ( Figure 5D). This correlation was also evident by quantitative 3D analysis of NMJ volume and surface area ( Figure 5F,G). Total fluorescence intensities and mean fluorescence per pixel were not changed ( Figure 5H,I), arguing for a lack of difference comparing total amount of AChRs or amount of AChR per pixel between control and mutant soleus NMJs. Of note, these structural data confirm different consequences of phenotypes in muscles of conditional Lrp5 and Lrp6 knockout mice. We continued addressing for any correlations of structure to function in conditional Lrp5 or Lrp6 knockout mice. To this end, we recorded extra-and intra-cellular potentials and currents in muscles of mutant mice in comparison with controls to analyze the physiology of neuromuscular transmission at their NMJs ( Figure 6). By recording CMAPs, compound muscle action potentials that are triggered by consecutive nerve stimuli, we neither detected a difference between control and mutant mice with CMAP amplitude ( Figure 6A), nor any significant change of the decrement of amplitudes at 5 and 50 Hz ( Figure 6B,C). Membrane resistance values were comparable between different genotypes arguing for non-affected membrane integrity ( Figure 6D). Moreover, recording of miniature endplate potentials (mEPP) did not reveal a significant change of the frequency ( Figure 6E). Neither mEPP or miniature endplate current (mEPC) amplitudes were different in mutant mice arguing for unaffected local depolarizations around endplates in response to spontaneous acetylcholine release ( Figure 6F,I), nor mEPP and mEPC rise time and decay time constants were changed in either mutant in comparison to controls ( Figure 6G,H,J,K). On the other hand, EPP and EPC amplitudes, local responses at NMJs to nerve stimulation, were decreased in conditional Lrp6 knockout mice ( Figure 6L,O). Run down experiments at 5 and 20 Hz demonstrated a decrease in EPP decrement for conditional Lrp6 knockout mice ( Figure 6M,N). Quantal content, the mean number of quanta that are released to generate an EPP, was not changed in conditional Lrp knockout mice in comparison with controls ( Figure 6P). In sum, no changes in neural transmission were recorded for conditional Lrp5 knockout mice. Altogether, our data demonstrate structural or functional impairments at the NMJs of adult conditional Lrp5 or Lrp6 knockout mice. Interestingly, impairments are more prominent in conditional Lrp6 knockout mice and less strong in conditional Lrp5 knockout mice. Cells 2022, 11, x FOR PEER REVIEW 16 of 21 Hz. Data are represented as mean ± SD; n ≥ 5 mice per genotype and analysis of >20 individual NMJs per muscle. All labeling of graph bars represents genotype of mice as indicated in (P). * p < 0.05.

Discussion
Active canonical Wnt signaling plays a key role in different aspects of skeletal muscle biology. While there is some knowledge available regarding the role of canonical Wnt signaling in early and late myogenesis and active signaling being present at NMJs [22], almost nothing is known regarding the need for the co-receptors of canonical Wnt signaling, Lrp5 and Lrp6, in skeletal muscles. Here, we used conditional Lrp5 or Lrp6 knockout mice to generate skeletal muscle-specific ablation of Lrp5 or Lrp6 expression by Cre recombinase mediated generation of knockout mice. We utilized HSA-Cre mice to ensure knockout of Lrp5 or Lrp6 at developmental stage, where myoblasts fuse to myotubes. With these mice, we investigated the role of Lrp5 and/or Lrp6 in adult muscle fibers or at NMJs. We aimed to understand whether both co-receptors are required for canonical Wnt signaling in adult muscle fibers and at NMJs, and whether we detect any specific changes distinguishing conditional Lrp5 from Lrp6 knockout mice. Previously, a different WNT3A response of Lrp5 or Lrp6 was found in osteoblasts arguing for a more important role of Lrp6 in comparison with Lrp5 [62]. Regarding the role of Lrp5 and/or Lrp6 at NMJs, this is even more exciting because an important regulator of NMJ formation and maintenance, the receptor tyrosine kinase MuSK, was reported to bind canonical Wnt ligands by its CRD and thereby act by itself like a canonical Wnt signaling receptor [35,63]. In patients, evidence accumulated for the pathophysiological importance of the MuSK CRD domain either by being linked to mutations in congenital myasthenic syndromes or by detection of autoantibodies in myasthenia gravis [64,65]. For better understanding, a mouse model lacking the CRD of MuSK was generated and characterized [63]. Interestingly, profound NMJ defects were detected in those mice at embryonic and adult stage culminating in muscle weakness [63]. The effects were reversible using lithium chloride, which is an agonist of the canonical Wnt signaling that can inhibit GSK3β activity and thereby stabilize free cytosolic β-catenin [63]. Obviously, it is interesting to understand the ways in which different canonical Wnt receptors are related to each other and whether the classical Wnt co-receptors, Lrp5 and Lrp6, are involved in canonical Wnt signaling at adult NMJs with or without MuSK CRD domain.
It is obvious to inquire the phenotype of the conditional Lrp5/Lrp6 double knockout mice. As mentioned above, the human skeletal actin promoter was reported to be also ectopically expressed [60]. We assume that ectopic Cre expression is the cause of lethality of conditional Lrp5/Lrp6 double knockout at embryonic stage~E13 ( Figure 1E). The use of other skeletal muscle Cre mice is required for further studies to completely understand the role of canonical Wnt signaling in muscle cells. Alternatively, the extraction of primary muscle satellite cells from mice expressing tamoxifen-inducible Cre recombinase might permit to establish Lrp5/Lrp6 double knockout muscle cells in vitro. Differentiation of such mononuclear myoblasts to multi-nucleated myotubes might allow to monitor the consequences of double knockout muscle cells regarding proliferation, myotube formation, and clustering of AChR upon incubation of myotubes with AGRIN-conditioned cell culture media. These possibilities for investigating the phenotype of the double Lrp5/Lrp6 knockout muscle cells or mice should be considered in future studies.
Here, our data evidence the importance of Lrp5 and Lrp6 acting together, because double knockout mice are lethal before birth, while single knockouts mice are viable ( Figure 1E). This means that the ectopic expression and subsequent knockout of one of the Lrp genes has a less serious effect than the knockout of both Lrp genes in those ectopic cells, where HSA-Cre is expressed. Previous reports indicate a more severe phenotype in Lrp6 in comparison with Lrp5 knockout mice [66,67]. Interestingly, our data analyzing conditional Lrp5 or Lrp6 knockout muscles are in agreement with the previous data, arguing for non-related function of these two co-receptors. Both conditional Lrp5 or Lrp6 knockout mice show profound impairments in adult skeletal muscle fibers and at NMJs. In adult muscle fibers, (1) oxidative metabolism is little reduced in mutant mice in comparison with controls (Figure 2A,C,D), (2) myosin heavy chain distribution and cross-sectional areas are disorganized in conditional Lrp6 knockout mice (Figure 3), (3) canonical Wnt target gene expression is reduced ( Figure 4A,B), and (4) myogenic gene expression profiles are partially modified ( Figure 4C,D). At the NMJs of conditional Lrp5 or Lrp6 knockout mice, we observed (1) slightly altered synaptic gene expression ( Figure 4E), (2) enlarged endplate width in diaphragm muscle ( Figure 5A,C), (3) fragmentation of NMJs ( Figure 5B,D,E), and (4) slightly but significantly impaired neuromuscular transmission ( Figure 6). The data we obtained for adult muscle fibers in mutant mice suggest a potential involvement of canonical Wnt signaling in fiber type distribution and metabolic status of the fibers. At NMJs, it appears that canonical Wnt signaling is influenced by the classical Wnt co-receptors Lrp5 and Lrp6, and their role in relation to MuSK-CRD requires further study. Interestingly, endplate band width is gradually enlarged in conditional Lrp5 or Lrp6 knockout mice ( Figure 5A,C). It might be worth to analyze whether prepatterning is differently affected in conditional Lrp5 or Lrp6 knockout mice by analyzing diaphragms of E13 and E18 embryos. Fragmentation of NMJs looking at conditional Lrp5 or Lrp6 knockout mice ( Figure 5B,D,E) might be related to modified synaptic gene expression ( Figure 4E).
Previously, conditional muscle-specific knockout of β-catenin, the key transcriptional co-activator of canonical Wnt signaling, was shown to consequence in a retrograde regulation of motoneuron differentiation in mice [29]. In our experiments, the gene expression level of β-catenin is unchanged in conditional Lrp5 or Lrp6 knockout mice ( Figure 4F), but the expression level of a typical canonical Wnt target gene, Axin2, is significantly reduced ( Figure 4A,B,F). It is likely that both Lrp5 and Lrp6 should be knocked out to detect retrograde regulation of motoneuron differentiation.
In summary, both Lrp5 and Lrp6 apparently play different roles in adult skeletal muscle fibers and at NMJs; they are not replaceable by each other. Future studies also involving the double knockout and the MuSK CRD mutant mice should help to address many more questions and move towards better understanding of the role of WNT receptors in skeletal muscle biology.