Research

Drosophila adult muscle development

Movie showing the flight muscles (green), which are attaching to tendons (orange); tension is built up and tendons form long extensions.

We use the Drosophila adult muscles, in particular the flight muscles to image muscle-tendon attachment and myofibrillogenesis in intact developing animals. We found that myotubes first attach to tendon cells at both myotube tips and only after being stably attached, myofibrillogenesis is triggered throughout the entire muscle. Using in vivo laser cutting experiments we discovered that mechanical tension is generated after muscles have attached to tendons. Interestingly, this tension build-up is required for ordered myofibrillogenesis (Weitkunat et al. 2014).

Quantifying forces across molecules in vivo

(A-C) Functional principle of FRET-based tension sensors. We insert these sensors into proteins to determine the force across proteins by quantifying the life time of the fluorescence (FLIM). (D) A tissue under mechanical strain: flight muscles (green) pull on stably attached to tendons (red). Regular myofibrils (green) start to assemble.

(A-C) Functional principle of FRET-based tension sensors. We insert these sensors into proteins to determine the force across proteins by quantifying the life time of the fluorescence (FLIM). (D) A tissue under mechanical strain: flight muscles (green) pull on stably attached to tendons (red). Regular myofibrils (green) start to assemble.

During myofibril and sarcomere assembly a quasi-crystalline pattern of myosin motors (thick filaments), actin tracks (thin filaments) linked by gigantic titin molecules (C-filament) is built in every muscle. Using in vivo imaging we found that myofibrillar pattern organises simultaneously across the entire muscle fiber after tension has been built up. We hypothesize that tension is used as a molecular compass to direct myofibril assembly, confirming that each myofibril spans across the entire muscle fiber. This guarantees effective muscle contractions in the mature muscle. We are testing this hypothesis by applying molecular force sensors, which enable us to quantify tension across individual molecules within developing myofibrils. We are also applying novel high-resolution microscopy techniques to monitor myofibril assembly at the nanoscale.

To walk or to fly

Endogenous Spalt protein tagged with GFP (green) by CRISPR-mediated HDR followed by RMCE is functional and localises to flight muscle nuclei. Trachea in white, myofibrils in red.

Wild-type fibrillar flight muscles but not tubular leg muscles express a particular Titin-related isoform (green), which is lost upon knock-down of spalt or arrest. Note the transformation of flight muscle to leg muscle morphology after loss of spalt.

Flight muscles harbour a specialised ‘fibrillar’ contractile apparatus to power fast wing oscillations at 200 Hz. This requires a very stiff muscle fiber type, related to vertebrate heart muscle. In contrast, slowly moving leg muscles display a tubular muscle architecture, closely related to striated vertebrate body muscles. We identified the conserved transcription factor Spalt as myofibril selector gene that instructs fibrillar muscle morphogenesis by inducing expression and alternative splicing of key sarcomeric components (Schönbauer et al. 2011). We found that downstream of Spalt the muscle specific splicing program is regulated by the RNA binding protein Arrest (Bruno) (Spletter et al. 2015). Thus as in vertebrate muscle, the biomechanics of a fiber-type specific contractile apparatus is regulated by transcription and alternative splicing of specific sarcomeric components. Currently, we are investigating the mechanism of fiber-type specific alternative splicing as well as the individual impact of the regulated components on myofibril assembly and muscle biomechanics.

CRISPRing the fly

Endogenous Spalt protein tagged with GFP (green) by CRISPR-mediated HDR followed by RMCE is functional and localises to flight muscle nuclei. Trachea in white, myofibrils in red.

Endogenous Spalt protein tagged with GFP (green) by CRISPR-mediated HDR followed by RMCE is functional and localises to flight muscle nuclei. Trachea in white, myofibrils in red.

We have developed an efficient two-step genome engineering protocol to manipulate any Drosophila gene of interest at its endogenous locus (Zhang et al. 2014). In step 1, we apply CRISPR/Cas9-mediated homology directed repair to replace a target region with a selectable marker (red fluorescent eyes), which is flanked by two attP sites. In step 2, we replace the inserted marker with any DNA of choice using ΦC31-mediate cassette exchange (RMCE). This enables flexible and efficient engineering of the locus, for example to generate a tagged allele, or to insert point mutations of choice.

The Drosophila TransgeneOme

LamininA-GFP expression in the adult thorax. Trachea and motor neurons are particularly well visible.

LamininA-GFP expression in the adult thorax. Trachea and motor neurons are particularly well visible.

In collaboration with the Tomancak, Sarov and VijayRaghavan labs we generated a genome-wide resource for the analysis of protein localisation in Drosophila. We tagged 10,000 proteins by inserting GFP into large genomic FlyFos clones. For about 900 clones we generated transgenic flies, which can be used to assess the in vivo dynamics and subcellular localisation of the tagged protein. For many of the tagged proteins functional antibodies or live imaging tools had not been available before. All transgenic lines are available from VDRC and thus can be used by the fly community. A preprint of the manuscript can be found here.