Phosphoinositide regulation of integrin trafficking required for muscle attachment and maintenance

Abstract

Muscles must maintain cell compartmentalization when remodeled during development and use. How spatially restricted adhesions are regulated with muscle remodeling is largely unexplored. We show that the myotubularin (mtm) phosphoinositide phosphatase is required for integrin-mediated myofiber attachments in Drosophila melanogaster, and that mtm-depleted myofibers exhibit hallmarks of human XLMTM myopathy. Depletion of mtm leads to increased integrin turnover at the sarcolemma and an accumulation of integrin with PI(3)P on endosomal-related membrane inclusions, indicating a role for Mtm phosphatase activity in endocytic trafficking. The depletion of Class II, but not Class III, PI3-kinase rescued mtm-dependent defects, identifying an important pathway that regulates integrin recycling. Importantly, similar integrin localization defects found in human XLMTM myofibers signify conserved MTM1 function in muscle membrane trafficking. Our results indicate that regulation of distinct phosphoinositide pools plays a central role in maintaining cell compartmentalization and attachments during muscle remodeling, and they suggest involvement of Class II PI3-kinase in MTM-related disease.

Figure 1. Mtm depletion leads to myofiber detachment and morphological defects common with human myotubular myopathy.

(A) Timeline of fly and muscle development, indicating stages of mtm requirements; days after puparium formation (APF). (B–B′) Normal body wall muscles in OreR and mtm^Δ77/mtm^z2-4747 third instar larvae. F-actin. (C) Percent viable and lethal progeny 13 days after egg lays with 24B-GAL4 and DMef2-GAL4 muscle-targeted mtm RNAi. (D–D′) Pharate dorsal abdominal muscles F-actin. Large IOMs (arrows) and smaller adult muscles (open arrowheads) span tergites (numerals). (D′) Detached IOMs (arrowheads) seen with mtm RNAi. (E–F) IOMs in filleted abdomens with 24B-GAL4 or DMef2-GAL4 expression of 1 or 2 copies of mtm RNAi hairpins. (E) Number IOMs present in tergites 3 and 4, including detached but present IOMs. (F) Number of present, visibly detached IOMs. (G–G′) Timelapse microscopy of GFP in IOMs imaged in same animals 1, 2, 3 and 4 days APF. (G′) With mtm RNAi, normal IOM formation (1d APF), survival upon histolysis of non-persistent muscles (2d APF), myofiber thinning (3d APF) and rethickening (4d APF) preceded detachment (4d APF, arrowheads). (H–H′) Individual IOMs. F-actin, red; DNA, blue. Projections, merged and nuclei images; central z-sections, F-actin. (H) Contractile myofibrils are normally tightly packed (double arrow) around linear aligned nuclei (arrowheads). (H′) With mtm RNAi, intact peripheral myofibrils surround expanded central area (double arrow) with unaligned nuclei (arrowheads). (I) Distance (µm) of nuclei from IOM midline. (J) Number nuclei per IOM. Scale bar 200 µm, except H–H′ 20 µm.

Figure 2. Mtm is required for βPS-integrin flux from intracellular compartments and localization at sarcolemmal adhesions.

(A) Schematic of individual pharate IOM and regions imaged. MTJ, myotendinous junction. (B–B″) βPS-integrin in IOM z-projections. (B) βPS-integrin at MTJs (arrow) and costameres (open arrowheads) in control. (B′–B″) With mtm RNAi, βPS-integrin was absent from detached ends (B″, arrow) and costameres, and detected on abnormal inclusions (arrowheads). (C–C′) IOM sarcolemma highlighting βPS-integrin at costameres in control (C, open arrowheads), absent with mtm RNAi (C′). (D–D′) IOM central z-sections revealing βPS-integrin punctae in control (D), and accumulation on abnormal inclusions with mtm RNAi (D′, arrowheads). DNA, blue. (E–E′) Transmission electron microscopy of IOM cross-sections, showing densely packed central regions in control (E) and large lucent membrane compartments with mtm RNAi (E′, arrowheads). (F) Averaged FRAP recovery curves and mean mobile fraction for larval βPS-integrin:YFP (int/+) in wildtype background (pink) and trans-heterozygous null mtm^Δ77/mtm^z2-4747 (orange). (G) Little to no βPS-integrin present at the sarcolemma in wildtype pupal IOM, 2.5 days APF. (H) βPS-integrin on central inclusions (arrowheads) detected in wildtype pupal IOM, 2.5 days APF. DNA, blue. (I–I′) βPS-integrin (red, and single channel below) at costameres in z-projections of adult abdominal lateral transversal muscles (I), and sporadically absent from costameres and dispersed in regions of myofibers with mtm RNAi (I′) in 6 day old adult flies. GFP, green; DNA, blue. DMef2-GAL4. (J) Heterozygous mtm^Δ77/+ enhanced frequency of adult wing blisters in hemizygous if³/Y flies. Scale bars 10 µm, except E–E′ 1 µm.

Figure 3. Integrin adhesions are independent of T-tubules, but share an mtm function for maintained organization.

(A) IOM longitudinal section schematic of alternating sarcolemmal structures: T-tubule membranes (green) marked by Amph and Dlg; costameres (red) with IACs linked to Z-lines of peripheral myofibrils; z-section images from central, central periphery or central perinuclear regions, as shown. (B–B′) Dlg, green; Zormin, red. Dlg detected continuously on longitudinal and transversal T-tubules in control (B), but only on longitudinal tubules and on central inclusions with mtm RNAi (B′, arrowheads). (C–C′) βPS-integrin (red) and dlg1:GFP (green) partially co-localized at longitudinal T-tubules in control (C, open arrowheads), and on abnormal central inclusions with mtm RNAi (C′, arrowheads). Single channels, right. (D) Normal βPS-integrin localization at costameres and (D′) internal punctae in amph²⁶ mutants that lack T-tubules. (E) Persistent βPS-integrin-inclusions with mtm RNAi in amph²⁶ mutant. amph²⁶, UAS-IR-mtm^3.1/amph²⁶; DMef2-GAL4/+. (F) Normal transverse tubule formation and Amph localization in mys¹/mys^ts1 mutants reared at non-permissive temperature, with reduced βPS-integrin function. Scale bars 10 µm.

Figure 4. Mtm depletion disrupts integrin trafficking at endosomal compartments.

(A–A″) GFP:LAMP (green, and single channels) found as punctae throughout IOM controls (A) and localized to inclusions with mtm RNAi (A′, arrowheads; A″). F-actin, red; DNA, blue. (B) GFP:Rab5 (green, and single channels) found as punctae with normally little overlap with βPS-integrin (red) throughout IOM controls. (B′–B″) Rab5 partially co-localized with βPS-integrin on inclusions and accumulated at the plasma membrane and perinuclear with mtm RNAi. DNA, blue. (C–C″) PI(3)P detected by GFP:2xFYVE (green, and single channels) and βPS-integrin (red) exhibited little overlap in control (C), but co-localized on inclusions with mtm RNAi (C′–C″, arrowheads). (D–D″) PI(3)P detected by GFP:2xFYVE (green) and Dlg (red) exhibited little overlap in control (D), but co-localized on inclusions with mtm RNAi (D′–D″, arrowheads). Scale bars 10 µm, except zooms A″, C″, D″ 2.5 µm.

Figure 5. Class II and Class III PI3-kinases affect mtm-dependent integrin adhesions differently.

(A–A′) Pharate abdominal muscles, F-actin. (A) IR-Pi3K68D and (A′) IR-Vps34 single RNAi (top) and mtm co-RNAi (bottom). Arrowheads, detached IOMs. (B) Number of visibly detached IOMs. (C, C′, C″) Sarcolemmal βPS-integrin detected at costameres; (C) control, (C′) IR-Pi3K68D and (C″) IR-Vps34 in single RNAi (top) and mtm co-RNAi (bottom). Only Pi3K68D, mtm co-RNAi restored βPS-integrin at costameres. (D) Percentage IOMs that lack costameres ≥half of myofiber surface. (E, E′, E″) βPS-integrin central z-sections; (E) control, (E′) IR-Pi3K68D and (E″) IR-Vps34 single RNAi (top) and mtm co-RNAi (bottom). Only Pi3K68D, mtm co-RNAi reverted abnormal βPS-integrin-inclusions. (F) Percentage IOMs with βPS-integrin on inclusions. (B,D,F) IOMs in single RNAi (light bars) and mtm co-RNAi (dark bars) conditions. Scale bars 10 µm, except A–A′ 200 µm.

Figure 6. β1-integrin is mislocalized to perinuclear inclusions in XLMTM myopathy.

Cryosections from XLMTM (n = 3) and age-matched control human muscle biopsies were immunostained either with anti-β-dystroglycan or with anti-β1D-integrin. (A–B) β1D-integrin was found only along the sarcolemma in control muscle (A), but was mislocalized to the perinuclear compartment in XLMTM fibers (B, arrows). (C–D) In contrast, Dystroglycan was found normally distributed along the sarcolemma membrane in both control (C) and XLMTM (D) muscle. Scale bar, 25 µm.