A novel adaptation of FRAP quantifies the movement of Drosophila Basement Membrane Collagen in vivo

doi:10.21203/rs.3.rs-4870374/v1

Download PDF

Research Article

A novel adaptation of FRAP quantifies the movement of Drosophila Basement Membrane Collagen in vivo

https://doi.org/10.21203/rs.3.rs-4870374/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

A wealth of knowledge regarding the functions of Extracellular Matrix (ECM) macromolecules from in vitro or disease models strives for validation in intact animals. In particular, the progression of cardiomyopathies is tightly linked to pathological changes in the heart ECM. To address this in the Drosophila model, we developed a novel adaptation of fluorescence recovery after photobleaching (FRAP), which allows us to assess ECM protein incorporation during growth in living, intact larvae. Recovery of fluorescently tagged protein is a proxy for addition or relocation of ECM protein. We focus on Collagen IVα (Viking), a conserved protein thought to be a stable component of the basement membrane (BM). We established a time course for Vkg-GFP fluorescence accretion in three different BMs through larval development, under normal conditions and when Matrix Metalloprotease or its inhibitor, TIMP is overexpressed. We demonstrate that the gain and loss of Collagen trimers from the basement membrane changes over developmental time and between tissues. High variability in measured fluorescence reduced the sensitivity of this approach. During growth, a strong phasic wave of Vkg accumulation was detected at the second to third instar ecdysis, potentially supporting growth of the new instar. Between organs, flux of Vkg was high in somatic muscle, intermediate in the heart and low in trachea. Heart-specific overexpression of mmp2 and its inhibitor timp, modified the dynamics of Vkg-GFP flux. We find that MMPs are positive regulators of Vkg/Col IV turnover in the ECM, in alignment with current models of ECM regulation.

Extracellular matrix (ECM)

basement membrane (BM)

fluorescence recovery after photobleaching (FRAP)

Collagen IV/Viking

protein turnover

heart/cardiac

Drosophila melanogaster

The heart is an organ that experiences constant mechanical stress. This stress is moderated by the cardiac extracellular matrix (ECM), a dynamic network of proteins, growth factors, and signaling molecules. The heart undergoes fundamental changes, such as size increase and contractility during growth and disease, but how does the ECM accommodate this? Here we explore a novel method to determine how new molecules enter and exit a beating heart in an intact model organism, to better understand the dynamic molecular nature of the cardiac ECM and compare this with other tissues.

Regulation of ECM integration and remodeling during growth is understudied, despite being critical for all organ systems, including heart growth and function. A central limitation is tracking protein turnover in vivo. Recent studies in intact C. elegans employed proteomics and fluorescence sampling to reveal that different collagens are stabilised or turned over during aging (1, 2). In contrast to potentially 181 Collagen genes in C. elegans, there are four in Drosophila (Reinhardt et al., 2023; Teuscher et al., 2019). Over the short term, Fluorescence Recovery After Photobleaching (FRAP) experiments in C. elegans reveal no turnover of basement membrane Collagen IVα1, while ECM cross-linking components such as Perxodasin, γ-Laminin and Nidogen recover over several minutes (1). Collagen IV and Laminin are known to form interconnected scaffolds early in BM formation, and are therefore more stable than other, more mobile ECM proteins (1). FRAP experiments in Drosophila to date have also been short-term, and performed in immobile, transparent embryos, dissected pupae, ovaries or imaginal discs (5–8, 8–12). Greig and Bulgakova (2021) performed FRAP in the epidermis of live late-stage Drosophila embryos to demonstrate that E-cadherin signal recovers by both diffusional and endocytic recycling mechanisms. Similarly, FRAP in Drosophila embryos has revealed turnover of myosin during epithelial wound repair (7).

In Drosophila embryos, hemocytes are predominantly responsible for secretion of ECM proteins Laminin, Collagen, SPARC, and others (13, 14). Post-embryonically, epithelial cells secrete ECM proteins such as Laminin and Syndecan to guide formation of their overlying ECM, but body-wide production of Collagen is primarily from the fat-body (Broadie et al., 2011; Pastor-Pareja & Xu, 2011; Zang et al., 2015, Supplementary table 4).

Post-embryonically, FRAP has been employed in Drosophila imaginal discs to evaluate morphogen trafficking (6, 12). In the Drosophila ovary, FRAP has been used to study dynamics of ER-derived fusome membranes as well as cortical actin organization dynamics (9, 11). To our knowledge, FRAP has never been used to assess protein accretion or turnover in the living Drosophila larva, which is physically active and grows considerably from hatching to pupation.

We developed a novel adaptation of FRAP to measure incorporation of Col IVα2 (Viking, Vkg) during larval growth stages in live Drosophila, employing the Vkg-GFP protein trap developed by Morin et al., (2001). Enhanced GFP (eGFP) is easily and irreversibly photobleached (19, 20). In this model, the endogenous vkg gene is modified with a GFP sequence, so gene copy number or enhancer properties are not modified as in other transgenic approaches. For example, a pulse of a vkg-GFP transgene might facilitate protein tracking, but with the consequence of over-supply of one protein in a trimer. A change in fluorescence of endogenous Vkg is a proxy for accretion or relocation of Col IV. Interpreting a change in Collagen fluorescence as ‘turnover’ is imprecise in an open system like the basement membrane (BM), as change may reflect movement of the trimer laterally in the BM, exchange with free trimer in the hemolymph, digestion, denaturation or shape change of the supported tissue. To this end, we developed a simple mathematical model to estimate the contribution of Col IV loss and gain, or flux, by comparing the values obtained from FRAP with control BM. Collectively, changes in ECM content or organisation have been termed "turnover", or "remodeling" although multiple processes are involved.

The Drosophila heart is a linear pump, with valves and an open circulatory system, and is in near constant motion as larval feeds and moves by peristalsis (21). All these features require both elasticity and tensile strength, suggesting a need for consistent protein flux. For contrast, we examined tissues with other biophysical challenges. Somatic muscle, inserted in the body wall, experiences intermittent stretch in one direction. The trachea requires an inelastic non-contractile ECM and is the most rigid in the body cavity (Hayashi & Kondo, 2018; Öztürk-Çolak et al., 2016).

Further, we elected to estimate flux at times in larval development when ECM is modified or replaced. Molting is the process through which an animal casts off old feathers, skin, hair, or shell to make way for new growth. In reptiles, arthropods such as Drosophila, and other insects, molting is specifically known as ecdysis. Ecdysis is controlled by hormones, namely the steroid hormone 20-hydroxyecdysone (20E), as well as several neuropeptides that modulate neural activity (24, 25). During ecdysis, tissues and their associated ECM are thought to undergo substantial remodelling to meet the demands of a rapidly growing larval body (Bonnans et al., 2014; Hughes, 2018; Hughes et al., 2020; Page-McCaw et al., 2007). Gene expression for replacement of cuticle as well as proteolytic activity of matrix metalloproteases (MMPs) are upregulated during ecdysis in many insects (30–33).

Here we reveal a developmentally dependent time course of Vkg-GFP accumulation in the cardiac and somatic body wall muscle ECM of Drosophila larvae. We also show that mRNA transcript levels derived from whole-body larval preps do not reveal a crucial role for gene transcription in the regulation of ECM accumulation and turnover during larval growth.

Many enzymes that contribute to ECM assembly and turnover have been identified, but their in vivo activity is not well characterised. We have applied the FRAP approach to quantify changes in Collagen levels after modifying the activity of proteases known to participate in remodeling of the ECM (34–36). Many key enzymes, such as matrix metalloproteinases (MMPs), a disintegrin and metalloproteinases (ADAMs), and ADAM with thrombospondin motifs (ADAMTSs) as well as their inhibitors, tissue inhibitors of matrix metalloproteinases (TIMPs), regulate proteolysis and hence contributes to turnover of ECM components (37, 38). Here we focus on one of the two Drosophila MMPs, MMP2, and the single Drosophila TIMP, a broad spectrum MMP inhibitor. Previous work in our lab investigated the roles of both MMP1 and MMP2 on heart growth and ECM dynamics. Reduced MMP1 function in the embryo resulted in myofibrillar disorganization. Manipulating its activity in larvae did not perturb heart or ECM morphology. Decreasing the level of MMP2 activity only during larval stages resulted in increased density and aberrant localization of Pericardin (Prc) and Vkg Collagens (Hughes et al., 2020). Conversely, increased TIMP expression phenocopied MMP1/2 depletion. These findings prompted us to investigate the impact of manipulated MMP2 and TIMP activity on Vkg turnover at larval stages, using FRAP. We show that mmp2 and timp overexpression driven by the cardiac and heart-adjacent pericardial cells alter accumulation and turnover of Vkg in the cardiac ECM.

Here we report the adaptation of FRAP to live, intact Drosophila larvae in second and third instar larvae. Through a novel adaptation of FRAP in a well characterised model we show that recovery of fluorescence is a proxy for accumulation of Vkg-GFP into a photobleached zone and can be used to estimate protein gain and loss. We report the gain and loss of Collagen trimers from the basement membrane, which changes over developmental time and between tissues.

Novel FRAP does not alter organization of ECM proteins

Recent work suggesting tangible turnover of ECM protein, including Collagen, has relied upon studies of in vitro, in ovo, or fixed tissue (Greig & Bulgakova, 2021; Van De Bor et al., 2021; Teuscher et al., 2024). Of particular interest to our study was the cardiac ECM, implicated in cardiac growth and cardiomyopathies. We aimed to track Collagen IV turnover in vivo, using an adaptation of fluorescence recovery after photobleaching (FRAP). We have developed a protocol enabling the visualisation, photobleaching, and subsequent re-imaging of tissues close to the cuticle of intact larvae and established that third instar larvae robustly recover over a 24hr recovery period (Fig. 1).

We first sought to establish that photobleaching does not alter the structure of the Collagen matrix in the basement membrane (BM) of intact animals. We photobleached third instar larvae of the genotype Vkg ^cc00791, and subsequently dissected and immunolabeled them with an anti-Vkg primary antibody (Fig. 2). Vkg^cc00791, hereafter Vkg-GFP, is a ‘protein trap’ allele of Drosophila in which GFP is inserted as an exon that fluorescently tags Collagen in vivo (Buczsak et al 2007). Strong (52mW) laser irradiation was sufficient to damage immunodetectable Collagen and elicit hemocyte targeting to the injury (Fig. 2A). In contrast, low (1.95mW) intensity irradiation was sufficient to photobleach Vkg-GFP and leave Collagen organisation unaltered. We also examined the possible impact of FRAP on Pericardin (Prc), a fibrous Collagen-like protein located in the cardiac interstitial matrix. Pericardin is required post-embryonically to maintain cardiac architecture and connection to adjacent tissue (Cevik et al., 2019; Rotstein & Paululat, 2016). Immunolabel of fibrous Prc revealed a preserved network of fibrils, unaltered in their pattern of length, branching or alignment (Fig. 3). These findings reveal that FRAP does not alter the in vivo organization of Vkg or Prc in the cardiac ECM.

Regulation of Vkg-GFP accretion is stage- and tissue-specific

Having established the feasibility of in vivo FRAP, we enquired whether growth and maturation of the ECM during larval development is reflected by changes in Vkg-GFP fluorescence in a standardized region of interest (ROI). Fluorescence change was measured by comparing initial fluorescence intensity to fluorescence intensity after 24h in both the cardiac and somatic body wall muscle ECM. Both tissues experience growth and mechanical stress, although heart movement is more continuous and occurs in more directions (Fig. 4). We focused on larval muscles 21, 22, and 23, also known as lateral transverse muscles (LT) 1–3, in segment A7, (Supplementary Fig. 1). These muscles were selected because are easily and reliably identified beneath the cuticle via confocal microscopy (41, 42). Although somatic muscles generate forces needed for feeding, locomotion, and hatching, the specific roles for larval muscles 21, 22, and 23 remain unclear (41, 43).

Effective photobleaching within the 66 x 33 µm ROI was affected by occasional movement of the animal under light anesthesia. To control for this and possible lateral diffusion of Collagen in the ECM, we quantified a smaller ROI of 33 x 33 µm nested within the original ROI (Supplementary Fig. 2). Average values for the post-bleach time point were far lower when measured in the nested ROI (33 x 33 µm) relative to the larger original ROI, which more accurately reflects the extent of photobleaching. As anticipated, the averaged fluorescence of the larger ROI was greater. Therefore, we based our analysis on the smaller, enclosed ROI.

Anticipating that tissue remodeling is elevated during the molt from second to third instar, we compared FRAP on late second instar (24–26 hours after hatching) with larval growth during early and mid-third instar stages. We measured Vkg-GFP fluorescence accretion after 24 hr in all experiments, having determined that recovery was more variable if larvae were handled and imaged after shorter recoveries. In contrast to the continuous record for in vitro FRAP, a single 24 hr 'snapshot' averages turnover, underestimating short term rates of change. The snapshot included molting for the late second instar group. We also measured fluorescence accretion in “sham” controls (Fig. 4) to estimate Vkg accumulation in unbleached, equivalent tissue. In cardiac sham control tissue, the normalized change in fluorescence over 24h did not change significantly from late second to mid third instar (Fig. 4M). Net new fluorescence over 24h in the cardiac ECM was highest at the early third instar stage in photobleached (FRAP) zones and in the control zone (Fig. 4M). This suggests Collagen turnover without net gain. In contrast, somatic muscle revealed the largest fluorescence gain in late second instar, dropping through later stages, suggesting net increase in muscle Collagen (Fig. 4M). These estimates were not corrected for animal growth, thereby underestimating the fluorescence gain. Growth of recovered ROIs was variable, averaging a 6% underestimate of gain. These findings indicate tissue- and stage-dependent regulation of Vkg-GFP accumulation. Therefore, we asked if it is possible to attribute differences to different rates of loss and addition of Collagen, reflective of dynamic flux?

Model for estimation of fluorescence loss and gain

Change in fluorescence of Vkg-GFP, a proxy for net change in Vkg protein levels, reflects net new fluorescence (residual plus new protein), including possible lateral movement within the BM and not accounting for potential fluorescence decay in situ. The FRAP model has the potential to estimate both gain and loss of Vkg-GFP. For both heart and muscle there is net new fluorescence after FRAP. However, control tissue can reveal a net loss over the same period. Assuming that Vkg-GFP is lost and gained at the same rate for both the photobleached and control ROI, the apparent contradiction may reflect different rates of addition and loss. All the Vkg leaving the control ROI is fluorescent, and only a fraction of the Vkg leaving the FRAP ROI is still fluorescent (Fig. 5B). This fraction is estimated by dividing the fluorescent signal after photobleaching by the pre-bleach fluorescence intensity value. Using the fluorescence levels at pre-bleach, immediately post bleach, and 24 hours post bleach timepoints, we can calculate the proportion of fluorescence lost and gained by solving for two equations with two variables (Fig. 5A). This approach cannot detect gain that is also lost within the 24-hr recovery. Given that bleached ROIs at 24 hr fluoresce less than controls, the underestimation of gain and loss should be small.

Vkg-GFP turnover is stage- and tissue-specific

We calculated loss and gain of fluorescence and interpreted them as proxies for loss and gain of Vkg protein. In cardiac tissue, the rate of loss appeared to be stable, and gain dropped significantly from early to the mid-third instar stage (Fig. 5C). Similarly, the rate of gain decreased significantly, and the rate of loss increased in somatic body wall muscle ECM throughout larval development (Fig. 5C). The decline in Vkg signal throughout larval growth in the body wall muscle aligns with the poor net fluorescence recovery seen in the third instar body wall muscle ECM.

We next extended our analysis the ECM of an inelastic tissue, the larval trachea. The branched tracheal network spans the length and width of the larval body. We qualitatively observed poor recovery of Vkg-GFP in photobleached branches from the main tracheal trunks near the heart and subsequently determined fluorescence recovery of Vkg-GFP as previously described. In early third instar female larvae, fluorescence accumulation was highest in the tracheal ECM in sham control tissue (Fig. 6D). Conversely, Vkg-GFP accretion was lowest in FRAP tissue of the tracheal ECM, both suggestive of low rates of Vkg loss (Fig. 6D). Vkg-GFP flux was highest in the cardiac ECM and lowest in the tracheal ECM (Fig. 6D).

Altered matrix protease activity modifies Collagen turnover in the cardiac ECM

ECM regulatory enzymes matrix metalloproteinases (MMPs), which digest ECM protein, and their inhibitor tissue inhibitor of matrix metalloproteinases (TIMP), are key to maintaining the balance between protein degradation and synthesis or deposition. Altered MMP and TIMP activity can lead to pathological states through modified ECM protein composition, organization, and remodelling (44). We used the heart-specific driver Hand-GAL4 to overexpress mmp2 and timp in third instar larvae. Second instar larvae were considerably smaller, and experienced decreased viability following FRAP relative to third instars. Combined with the reduced survival of larvae overexpressing mmp2 and timp, we focused exclusively on more robust third instar larvae (Hughes et al., 2020).

Heart-specific overexpression of mmp2 did not significantly alter Vkg-GFP fluorescence recovery relative to wildtype (vkg-GFP heterozygotes) in the cardiac ECM of early and mid-third instar larvae (Fig. 7G). A non-significant increase in Vkg-GFP gain and loss was evident in the early third instar. (Fig. 7H). In contrast, Vkg-GFP fluorescence recovery and flux significantly decreased after timp overexpression relative to controls in the mid-third instar cardiac ECM, and to a lesser degree in early third instars, suggesting that TIMP is implicated in stabilizing Collagen turnover (Fig. 7G and H). A possible effect of the homozygous vkg-GFP allele or Hand-GAL4 insert on fluorescence change was assessed and found to be not significant (Supplementary Fig. 5).

We evaluated possible distant effects of heart-specific mmp2 and timp overexpression on fluorescence recovery and Vkg-GFP turnover in body wall muscle. Overexpression of mmp2 did not affect Vkg-GFP fluorescence recovery or Vkg-GFP flux in the body wall muscle of early and mid-third instar larvae (Fig. 8G and H). In contrast, heart-specific overexpression of timp increased Vkg-GFP accumulation in sham control tissue and significantly decreased calculated loss at both stages (Fig. 8G and H). The UAS-timp insert has ectopic or “leaky”, expression, affecting Vkg turnover in the body wall muscle, as noticed in previous studies (Hughes et al., 2020). We assessed Vkg loss and gain with driverless timp overexpression and similarly observed reduced Vkg turnover, although this difference was not statistically significant (Supplementary Fig. 4D).From these results we can surmise that timp overexpression affects homeostatic Vkg turnover by decreasing Vkg loss.

ECM and ECM-associated gene transcript levels vary little throughout larval development

The expression of ECM and ECM-related genes fluctuates in disease states, and throughout development and growth (45, 46). We sought to assess this by examining whether changes in gene expression underlie the differences in Vkg accretion and estimated turnover between larval stages. To this end, we quantified changes in mRNA transcript levels in late second, early-third and mid-third instar larvae, by qPCR on RNA isolated from whole larval bodies.

Transcript levels of ECM structural proteins: Laminin, Integrin and Perlecan (lanA, mys, trol, wb), Collagens (also structural; Cg25C, prc, vkg), enzymes active in the ECM: matrix metalloprotease and tissue inhibitor of metalloprotease (mmp1, mmp2, timp), Collagen modulators (loxl1, SPARC), and the moulting initiator (ETH) were determined (Fig. 9). See Supplementary table 5 for a description of all targets and their human orthologs. Expression of structural genes (lanA, mys, trol, and wb) as well as Collagen Cg25C were highest at late second instar (LL2) and lowest at early third instar (EL3), suggestive of decreased expression between molts. Transcript levels of ECM proteases mmp1 and mmp2 and their inhibitor timp were highest at mid-third and lowest in early third instar, indicating decreased expression between molts and increased expression prior to pupation. In contrast, both vkg and prc increased as larvae grew into third instar. None of these transcript level changes were statistically significant, suggesting that transcription or transcript half-life are not key regulators of change in the larval ECM.

Transcript expression levels for vkg were significantly higher in larvae of the genotype Vkg-GFP (Vkg^cc00791) relative to yw at each developmental stage (p-values < 0.0001) (Supplementary Fig. 3). The Vkg-GFP protein trap consists of a transposed GFP exon in an intron of the endogenous Vkg gene, so that Vkg coding an N terminal GFP is under control of the vkg promoter. Vkg-GFP exhibits subcellular and tissue localization akin to endogenous Vkg, and is believed to not disrupt upstream or downstream splicing events (18). It is possible that the location of the GFP insert within the Vkg sequence is limiting degradation of the vkg transcript.

We had included ETH (ecdysis triggering hormone) in the hope of correlating molting with our transcript levels. ETH is considered a master regulating hormone of ecdysis, and the developmental timepoints selected for this assay span ecdysis (47). FlyBase’s RNA-Seq By Region tool reports high ETH levels in pooled L2 larvae (Öztürk-Çolak et al., 2024). Our data suggests this drops before ecdysis. Further comparisons with the Flybase dataset were limited by a lack of statistics and different sampling windows.

Incorporation of BM proteins into a functional matrix over the course of organismal growth is poorly understood. To better compare the addition and turnover of the Drosophila Collagen IVα1, or Vkg across different tissues, we developed a novel adaptation of FRAP for live, intact Drosophila larvae. This technique allows us to measure fluorescence recovery of the protein trap Vkg-GFP, which we use as a proxy for accretion of the Vkg protein over a 24-hour recovery period, extending the time-course of previous photobleaching studies. Studies in C. elegans, which lack a heart, employed FRAP, or FRAP-adjacent techniques to study protein turnover in vivo (1, 48, 49). They revealed that rates of protein turnover vary between tissues and decrease with age. Pulse-chase tracking of ECM in Drosophila embryos suggests a short half-life for BM Collagen Matsubayashi et al., 2020). The role of ECM proteases in metamorphosis has been tracked in intact pupae (Davis et al., 2022). Here we provide the first report of in vivo ECM protein turnover in intact Drosophila larvae.

Vkg-GFP content of basement membranes varies with developmental time and tissue type

Our results indicate that an Argon laser intensity of 1.95 mW is sufficient to photobleach the GFP portion of the Vkg-GFP protein trap without affecting the immunoreactivity or ECM incorporation of the Vkg protein. Our findings also indicate that the organization of Prc, a fibrillar Collagen in the same matrix, is not perturbed by FRAP. We next evaluated whether maturation of the ECM during larval growth and development can be tracked through changes in Vkg-GFP fluorescence in a region of interest (ROI). Ecdysis (or molting) between larval instars provides a phasic ECM remodeling event to analyse ECM turnover. The tissues characterised here are each linked indirectly to the new cuticle. The heart and alary muscle reorganize their insertions, which may entail increased remodeling (50, 51).

The difference in Collagen deposition between body wall muscle and the heart is most pronounced during molt to the third instar. Net gain was greater in body wall muscle, possibly reflecting the remodeling of muscle and tendon cells as they insert into new cuticle. During growth stages in the third instar, somatic muscle experienced a net loss of Vkg, while heart Collagen fluorescence revealed no consistent trend. This was unexpected as we anticipated that growth and the constant movement of the cardiac ECM would require more remodeling, including more removal of photobleached Collagen. Instead, the results suggests that the cardiac ECM is more stable at all stages, with lower Vkg turnover relative to the body wall muscle ECM. This might be attributable to increased crosslinking in heart ECM. Growth of the tissue in the ROI after photobleaching led to an estimated 6% allometric underestimation of Vkg accretion. Analysis of net change did not uncover the balance of loss versus gain of Collagen molecules during remodeling, which prompted us to seek a model to track Vkg movement.

Vkg-GFP addition and loss are stage- and tissue- specific

The assessment of fluorescence change of Vkg-GFP is a proxy for net change in Vkg protein levels but does not allow for direct measurement of protein turnover. Turnover includes both addition and loss, and if balanced, would not alter measured Collagen fluorescence. To determine the rate of turnover, we developed a mathematical model to calculate the estimated loss and gain of Vkg over our 24-hour recovery period. By photobleaching Collagen in one region of interest (ROI) and considering protein addition or loss as equivalent to a control ROI, we can estimate the rate of loss of photobleached Collagen by contrasting the change in fluorescence between both ROIs. Loss of photobleached Vkg-GFP would not change fluorescence. Anticipating that Vkg turnover would be highest during the second instar molt, we compared calculated loss and gain of Vkg between late second, early, and mid-third instar larvae. The rate of Vkg loss appeared fixed in cardiac tissue. Calculated addition of new Collagen was significantly lower in mid-third instars relative to late second and early third instars. We observed a net increase in Vkg-GFP fluorescence in body wall muscle ECM after molting of the second instar larvae. The gain of Vkg becomes less after ecdysis in both tissues, perhaps indicating a decreased need for protein deposition. While the rate of loss did not change in the heart across time, the loss of Vkg protein increased in muscle.

We extended our assessment of Vkg fluorescence recovery and turnover to the larval trachea, an robust network of inelastic epithelial tissue often used as a model to study tissue morphogenesis (52, 53). Vkg-GFP fluorescence recovery and calculated turnover in photobleached ROIs were substantially lower in the tracheal ECM relative to other tissues. Interestingly, Vkg accretion in control tracheal tissue was found to be elevated relative to that seen the cardiac and body wall muscle ECM. Together, the mathematical model interprets this as a low rate of Vkg loss in the tracheal ECM. This is consistent with higher levels of Collagen crosslinking in rigid trachea (22, 54).

ECM and ECM-associated gene transcript levels change throughout larval development

Rates of Collagen gain and loss appear to be correlated with tissue type, elasticity, and developmental stage. Nevertheless, multiple tissues with different ECMs obtain most or all of their Collagen from a central source, the Drosophila fat body, which secretes Collagen into the hemolymph for distribution (16). Insertion and crosslinking of Collagen from the hemolymph to diverse tissues is local, and after embryogenesis, not mediated by fibroblast-like cells as for vertebrates (55–57). How much of the regulation of ECM production can be at the genetic level? We sought to detect changes in ECM and ECM-associated gene transcript levels through larval development to determine how much change in ECM production is regulated transcriptionally. Unexpectedly, whole-body mRNA levels for most ECM genes of interest, including many that are cell autonomous (Laminins, Integrins, crosslinkers and proteases) were not statistically different across the three developmental stages assayed (1, 58–61). As the hemolymph remains a reservoir for unincorporated Collagen IV trimer, local factors, as opposed to supply, may regulate the rate of Vkg accretion. Some locally synthesized ECM proteins, such as Laminin will exhibit tissue-specific changes in transcript expression that would not be detected with a whole organism approach. Organ-specific mRNA isolation is impractical with Drosophila – particularly with the heart, which is associated with polyploid non-contractile pericardial cells (21).

We detected higher mRNA expression of vkg in larvae of the genotype vkg-GFP compared to wildtype larvae at each larval stage. It is plausible that the location of insertion of the GFP exon within the vkg coding sequence interferes with the degradation of vkg transcripts (18).

Taken together, our results show that ecdysis accelerates Vkg accretion in both the cardiac and body wall muscle ECM. Our data on calculated Vkg gain and loss reveal that each flux is differentially regulated between cardiac and body wall muscle and tracheal ECM, as well as between larval developmental stages. Developmental changes are not strongly linked to variation in matrisome transcript levels, and likely reflects changes in local kinetics of protein exchange between the hemolymph and the ECM.

Altered matrix protease activity modifies Collagen turnover in the cardiac ECM

Post-embryonic remodeling of the ECM in living animals is not well studied. Recently, Teuscher et al. (2024) revealed that coordinated ECM remodeling through mechanotransduction is required and sufficient to extend lifespan in C. elegans. Our work reveals that rates of addition and loss of Collagen from the ECM varies by tissue and developmental stage. Proteases are believed to be the major players responsible for protein loss from the ECM. During Drosophila embryogenesis, MMP1 is required for incorporation of new Collagen into the nascent basement membrane and is thought to maintain pliability of the Collagen matrix (49). MMP1 has also been shown to be crucial for accumulation of Vkg at the leading edge during wound closure in Drosophila embryos and is therefore important for wound closure and re-epithelialization. Expression of TIMP inhibits this process (62). We have previously determined that the modified MMP1 activity had little effect on the larval heart, while MMP2 is required for post-embryonic heart development and removal of excess Collagen. Similarly, this is inhibited by TIMP (Hughes et al., 2020). We sought to extend our understanding to the level of Collagen turnover. Our findings provide further evidence that heart-specific overexpression of mmp2 and timp perturbs the intrinsic turnover of Vkg in the third instar cardiac ECM. For instance, overexpression of mmp2 results in greater Vkg-GFP fluorescence recovery and Vkg flux relative to wildtype in the early third instar cardiac ECM. In the process of removing Collagen, it appears that MMPs reveal openings to assemble new ECM networks.

We noted a more dramatic effect of timp overexpression relative to mmp2 overexpression on net Vkg-GFP fluorescence recovery. This is in accordance with a more severe cardiac phenotype with altered timp activity previously reported (Hughes, 2018; Hughes et al., 2020). Drosophila produce a single TIMP protein, which inhibits multiple proteases, including MMP1/2 and ADAMTS (63, 64). In the cardiac ECM of third instars, timp overexpression leads to decreased Vkg-GFP fluorescence recovery. Overexpression of timp in the Drosophila embryo phenocopies loss of mmp1, whereby Vkg accumulation and wound closure are impaired (62). Our findings are consistent with a requirement for protease and protease inhibitor activity to remove and replace Collagen in the basement membrane. When less Collagen is removed, less will be added.

Altogether our findings show that regulation of Vkg protein turnover, and likely also protein crosslinking are tissue and stage dependent. Increasing the activity of the ECM protease mmp2 led to more Vkg turnover in the early third instar cardiac ECM, whereas increased timp appeared to decrease Vkg turnover in the cardiac ECM. It is probable that additional factors, such as protein modification and cross-linking contribute to differences in protease mediated Collagen turnover between organs and stages.

Prospects for ECM studies in vivo

Despite the important value of non-invasive studies of the ECM in vivo, the approach is infrequently chosen. Inevitable tissue movement, animal resampling, re-identification of ROIs, and potential stress from handling and irradiation are disincentives for quantitative study. Repeated sampling from a living organism introduces more sources of variation and reduces the number of statistically significant observations. Individuals may vary in optical diffraction and transparency at different ages, as well as the optical depth to the tissue of interest. There is no endogenous calibration for this, so measured fluorescence for each tissue type is assumed to be proportional to the number of tagged GFP molecules. We also note that natural decay of fluorescence, growth of the tissue, and molecules added and lost within the recovery period all contribute to an underestimation of gain and flux. Most significant outcomes reported here are relative increases in flux, despite this bias.

There are uncertainties with the use of tagged proteins that may bias quantitative interpretation. Our study, and others employ "protein trap" GFP inserts, which modifies only the endogenous gene (1, 18). As Vkg transcript rises through development in both wildtype and Vkg-GFP genotypes, we expect, despite the higher transcript level of Vkg-GFP in larvae relative to wildtype, that the rise seen in our GFP trap larvae reflects wildtype dynamics. However, the half-life for fluorescence in a GFP protein trap is unknown. Estimates from other in vivo models suggest a fluorescence half-life triple our experimental time course (65). The relative stability of Vkg-GFP in the ECM compared with untagged Vkg is also not known.

Our mathematical model is successful in revealing changes in the flux of Collagen. However, it forces high loss and gain when fluorescence intensity values are very different between photobleached and control tissue. Incorporating corrections for the suggested factors that affect GFP fluorescence may compensate for this bias. Development of intramolecular FRET GFP variants as protein traps may enable pulse-chase studies for more direct assays of matrisome turnover (66).

For the first time, turnover of endogenous protein in the ECM of a growing heart and muscle have been characterised in a non-invasive manner. Our findings reveal that there is both loss and accumulation of Collagen during larval development and growth that varies between tissues and correlates with tissue elasticity. Although less precise than studies in vitro or in ovo, it is essential to corroborate and extend insights from biologically constrained models to intact organisms. We believe that our approach can be applied to estimate turnover of other fluorescent proteins in vivo in transparent organisms such as C. elegans, Drosophila and Zebrafish (Danio rerio). Despite the inherent variability of in vivo imaging, this study establishes that the ECM of the basement membrane undergoes constant turnover and suggests a relatively short half-life of Collagen in active, moving tissue. Turnover appears to be greater when tissue remodeling, associated with insect molt, is higher. The model also provides in vivo evidence for protease activity in flux of Collagen in the basement membrane. Other studies have suggested a role for SPARC, Peroxidasin, Nidogen and Lysyl Oxidase in Collagen IV stabilisation that can now be validated and related to ECM pathologies such as cardiac fibrosis.

Fly stock maintenance and stocks

Flies were reared on yeast-based solid food in polystyrene vials at room temperature (22–23°C) unless otherwise indicated. For a list of all fly stocks used, see Supplementary Table 1.

Larval collections

To collect larvae for FRAP and qPCR experiments, flies were placed in collection chambers comprised of a 100 mL plastic chamber, inverted over a 60mm plate of 10 ml standard fly medium. Houses were set up to contain roughly 50–70 flies, using a 2:1 female to male ratio, and placed in a 25°C incubator. After 48h acclimatisation, newly hatched first instar larvae (24h after egg fertilization) were hand-picked and moved to another food plate. Larvae were then collected at three different timepoints: late second instar (24-26h after hatching), early third instar (48-50h after hatching), and mid-third instar (76-78h after hatching). Developmental stages were confirmed with larval mouth hooks. Slight differences in developmental time were noted between genotypes and were accounted for. The genetic background of both Vkg-GFP and yw flies is wildtype Canton S.

The control genotype used for mmp2 and timp overexpression experiments was Hand-GAL4, Vkg-GFP/Cyo-YFP, which, like the experimental crosses, is heterozygous for Vkg-GFP. In Figs. 2–4, homozygous Vkg-GFP was used as the wildtype. We observed that fluorescence recovery in homozygous individuals (Vkg^cc00791) was lower in the cardiac and body wall muscle ECM compared to individuals heterozygous for Vkg-GFP.

Larval freezing for qPCR

For qPCR experiments larvae of the Vkg-GFP and yellow white (yw) genotypes were collected and frozen. Late second instar, early third and mid-third instar larvae were collected for qPCR as described. Nuclease-free snap tubes containing RNAse-free PBS (100 µl) were prepped. For both the early third and mid-third instar collections, five larvae were collected per tube. Ten larvae per tube were collected for late second instars owing to their smaller size. Snap tubes were immediately flash-frozen in liquid nitrogen and stored at -80°C.

Larval dissections

Dissections were performed in early third instars according to the protocol adapted from (67). All steps were performed at room temperature unless stated otherwise. Live larvae were immobilized ventral side up using tungsten dissection pins magnetically adhered to a dissection plate. Larvae were bathed in phosphate-buffered saline (PBS), then fine iridectomy scissors were used to deflate the larva by making the first incision at the anterior end to avoid disrupting the fat bodies and cardiac muscle attachments. The first incision was extended along the ventral midline to the posterior extremity. The cut cuticle was then pinned on either side of the larva to expose the contents of the body cavity. Intestines and posterior fat bodies were removed. Tracheal branches were left intact to limit damage to the alary muscles suspending the heart in the body cavity. Dissected larvae were fixed using 4% formaldehyde in PBS in the dissection plate for five minutes at room temperature, then placed on ice in a 48-well plate containing the same fixative solution. Once all dissections were complete, they were allowed to fix at room temperature in the 48-well plate for 15 minutes prior to immunolabeling.

Immunolabeling

The immunolabeling protocol used was adapted from Alayari et al. (2009). All steps were performed at room temperature unless otherwise specified. After removing the fixative solution, dissections were washed three times for ten minutes in 1X PBT (1X PBS + 0.3% Triton). Dissections were then blocked with 10 µL normal goat serum (NGS) in 150µL PBT for 30 minutes prior to the addition of 5 µL primary antibody (1:30) in 150 µL PBT and left to incubate at 4°C overnight with constant shaking. The following day the dissections were again washed three times in PBT, then blocked with NGS as described. Samples were then incubated with 1 µL of the secondary antibody (1:15) and 2 µL of Alexa 647 Phalloidin (1:75) in 150 µL PBT for one hour at room temperature with continuous shaking. A final round of three PBT washes was performed, followed by a ten-minute wash in 1X PBS to remove the Triton. Upon removal of PBS, dissections were placed in 50% glycerol in PBS at 4°C for three hours or overnight, the transferred to 70% glycerol before imaging. See Supplementary table 3 for a complete list of antibodies used.

Confocal imaging

Frontal stacks Z-stacks were taken at 200 Hz with a step size of 1 µm at a resolution of 1024 x 512 pixels on a Leica SP5 confocal microscope. Settings were kept constant across all experiments. A 20X objective was used in FRAP experiments and to image dissected and immunolabeled samples. A 63X objective was used to image immunolabeled Pericardin to assess fiber orientation. Pinhole sizes of 60 and 95 µm were used with the 20X and 63X objectives, respectively. Sequential scanning was used to avoid crosstalk between channels (488, 647, and 543 nm) when imaging immunolabeled dissections. Only healthy, feeding and motile larvae were re-imaged at the 24h recovery timepoint. All sampling was made over a 4 hr midday window to eliminate potential circadian effects (69).

Projections of stacks were generated using the Leica LAS AF software. Only the Argon laser was used for FRAP experiments. Argon laser power was set to 15%, unless otherwise specified. Unaltered images were used for quantification but were adjusted for brightness for publication.

Fluorescence recovery after photobleaching (FRAP) for live, intact larvae

The FRAP protocol used below was also previously described in (MacDuff, 2019). See Supplementary Information for the complete protocol. First, larvae were anesthetized with chloroform in order to immobilize them, as described by Cevik et al. (2019). Prior to mounting larvae for imaging, a thin layer of halocarbon 27 oil (polymer of chlorotrifuoroethylene (PCTFWE)) was painted between two 1.5mm coverslips, 5mm apart on a microscope slide. This ensured that the larva does not stick to the slide or overlying coverslip. Anesthetized larvae were then laid dorsal side up on the slide, and a third cover slip secured over them with tape. A small drop of immersion oil was dispensed onto the overlying coverslip, and body segment A7 centred in the field of view.

To obtain 3D projection of the pre-bleach stack of images, laser power was set to 15%. Number of images within a stack was between 80–100 for the heart and 50–70 for the muscle to encompass the region of interest, excluding the cuticle or other structures. The ROI (66x33 µm) was then placed over the region of interest at 12X zoom. Laser intensity in the photobleached zone was set to 3% (1.95 mW). Following the photobleaching period, Argon laser power was set to 15%, as before. A post-bleach stack and 3D projection were then generated. After 24h, larvae were re-imaged as described.

To quantify FRAP, fluorescence intensity from original files was measured in grey values, each pixel from 0 to 255. Using the Leica LAS AF software, the average grey value in the sampled 33x33 µm ROI within the photobleached zone (66x33 µm) in segment A7 and in an ROI of the same dimensions in the adjacent, unbleached segment (A6) were obtained at the pre-bleach, post-bleach and 24h post-bleach timepoints. At the 24h timepoint, the ROI was re-identified based on segment and original placement. The bleached zone was often still discernible in some or all Z-steps within the stack. Calculations of change in ROI fluorescence were not corrected for animal growth. Estimates of growth of the bleached area over 24h were variable, averaging to a 6.1% increase (data not shown).

To calculate normalised fluorescent recovery, the 24h post-bleach value was divided by the pre-bleach value. No adjustment was made for the half-life of GFP fluorescence in vivo. In intact Dictytostelium, the half-life of GFP is 70 hours (65). The following tests were performed on the heterozygous vkg-GFP controls and homozygous vkg-GFP datasets using 2023 GraphPad Prism© v10 to demonstrate that fluorescence recovery data is normally distributed: D’Agostino-Pearson, Shapiro-Wilk, and Anderson-Darling. F-tests for equality of variances between developmental stages or genotypes were performed in Microsoft© Excel (2023). ANOVA (GraphPad Prism 10© Version 10.0.0) was performed on normal data. Tests for significance and corrections for multiple comparisons are identified in figure captions 4–8. Outliers were identified using GraphPad’s ROUT method. Fluorescence loss and gain were calculated as described using the Desmos online Graphing Calculator (© 2023 Desmos Studio, PBC). Normality could not be established for calculated loss and gain data between developmental stages or genotypes. The Kruskal-Wallis with Dunn’s multiple comparisons test (GraphPad Prism 10© Version 10.0.0) was then used to determine statistical difference.

Quantification of Pericardin fibre orientation

Early third instar Drosophila larvae underwent dissection immediately after photobleaching (Fig. 3). A method for quantifying ECM patterns, developed by Wershof et al. (2021) was used to determine differences in fibre alignment, branching, endpoints, or curvature between immunolabeled Prc in the photobleached and sham control ROI. The TWOMBLI parameters in the ImageJ plug-in used were: contrast saturation (0.35), minimum line width (7), maximum line width (11), minimum curvature window (40) maximum curvature window (40), minimum branch length (10), maximum display HDM (200), and minimum gap diameter (0). Unpaired t-tests were performed using 2023 GraphPad© Software to test for statistical significance between FRAP and control ROIs. Results were consistent across a range of TWOMBLI parameters.

RNA extraction

Whole-body RNA was extracted using the a magnetic bead protocol, based on the method outlined in Yost et al. (2020) (see Supplementary Information). Larvae were pooled as follows: ten larvae per LL2 sample and five larvae per EL3 or ML3 sample. Expression levels were tested three times per genotype for n = 3.

cDNA synthesis

Following extraction, RNA was reverse transcribed to cDNA using the Applied Biosystems® (Thermo Fisher Scientific) High Capacity cDNA Reverse Transcription kit, according to the manufacturer’s instructions.

qPCR analysis

cDNA primers were selected using the DRSC FlyPrimerBank and the NCBI Primer design tool. Glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was selected as the reference gene. qPCR was performed using the Applied Biosystems StepOnePlus™ Real-Time PCR instrument. Interaction plots of expression data (Δ C_T) presented here were generated in R Statistical Software (v4.0.2) using the following libraries: emmeans (1.7.2) (73), lme4 (1.1–30) (74), ggbeeswarm (0.6.0) (75), and ggplot2 (3.3.6) (76). DCT comparisons employed the Holm-Bonferroni correction for multiple testing.

Data were compared with Flybase (23).

A7	Abdominal segment 7
BM	Basement membrane
Cg25C	Collagen at 25C
Col IV	Collagen IV
DSHB	Developmental Studies Hybridoma Bank
DV	Dorsal vessel
ECM	Extracellular matrix
FRAP	Fluorescence recovery after photobleaching
LanA	Laminin A
Lox	Lysyl oxidase
MMP	Matrix metalloproteinase
Ndg	Nidogen
NGS	Normal goat serum
PBS	Phosphate buffered saline
PBST	Phosphate buffered saline with 0.3% Triton-X-100
Prc	Pericardin
Pxn	Peroxidasin
ROI	Region of interest
SPARC	Secreted protein acidic and cysteine rich
TIMP	Tissue inhibitor of matrix metalloproteinase
Trol	Terribly reduced optic lobes
UAS	Upstream activation sequence
Vkg	Viking
Wb	Wing blister
WT	Wildtype
yw	Yellow white

Author Contribution

D.M. wrote the main manuscript and prepared the figures. R.J. assisted with manuscript writing and figure design and acquired funding.All authors reviewed the manuscript.

Acknowledgement

The authors thank Dr. Ian Dworkin for his statistical assistance and generous sharing of laboratory space, as well as Dr. Rachel Andrews for editorial advice.

Keeley DP, Hastie E, Jayadev R, Kelley LC, Chi Q, Payne SG, et al. Comprehensive Endogenous Tagging of Basement Membrane Components Reveals Dynamic Movement within the Matrix Scaffolding. Developmental Cell. 2020;54(1):60–74.e7.
Teuscher AC, Statzer C, Goyala A, Domenig SA, Schoen I, Hess M, et al. Longevity interventions modulate mechanotransduction and extracellular matrix homeostasis in C. elegans. Nat Commun. 2024;15(1):276.
Reinhardt M, Drechsler M, Paululat A. Drosophila collagens in specialized extracellular matrices.
Teuscher AC, Statzer C, Pantasis S, Bordoli MR, Ewald CY. Assessing Collagen Deposition During Aging in Mammalian Tissue and in Caenorhabditis elegans. Methods Mol Biol. 2019;1944:169–88.
Greig J, Bulgakova NA. Fluorescence Recovery After Photobleaching to Study the Dynamics of Membrane-Bound Proteins In Vivo Using the Drosophila Embryo. In: Campbell K, Theveneau E, editors. The Epithelial-to Mesenchymal Transition: Methods and Protocols [Internet]. New York, NY: Springer US; 2021 [cited 2023 May 6]. p. 145–59. (Methods in Molecular Biology). Available from: https://doi.org/10.1007/978-1-0716-0779-4_13
Kicheva A, Holtzer L, Wartlick O, Schmidt T, González-Gaitán M. Quantitative Imaging of Morphogen Gradients in Drosophila Imaginal Discs. Cold Spring Harbor protocols. 2013;2013.
Kobb AB, Zulueta-Coarasa T, Fernandez-Gonzalez R. Tension regulates myosin dynamics during Drosophila embryonic wound repair. Journal of cell science. 2017;130(4):689–96.
Lin L, Othmer HG. Improving Parameter Inference from FRAP Data: an Analysis Motivated by Pattern Formation in the Drosophila Wing Disc. Bull Math Biol. 2017;79(3):448–97.
Molinar-Inglis O, Oliver SL, Rudich P, Kunttas E, McCartney BM. APC2 associates with the actin cortex through a multipart mechanism to regulate cortical actin organization and dynamics in the Drosophila ovary. Cytoskeleton. 2018;75(7):323–35.
Müller P, Rogers KW, Yu SR, Brand M, Schier AF. Morphogen transport. Development. 2013;140(8):1621–38.
Snapp EL, Iida T, Frescas D, Lippincott-Schwartz J, Lilly MA. The Fusome Mediates Intercellular Endoplasmic Reticulum Connectivity in Drosophila Ovarian Cysts. MBoC. 2004;15(10):4512–21.
Zhou S, Lo WC, Suhalim JL, Digman MA, Gratton E, Nie Q, et al. Free Extracellular Diffusion Creates the Dpp Morphogen Gradient of the Drosophila Wing Disc. Current Biology. 2012;22(8):668–75.
Martinek N, Shahab J, Saathoff M, Ringuette M. Haemocyte-derived SPARC is required for collagen-IV-dependent stability of basal laminae in Drosophila embryos. Journal of Cell Science. 2008;121(10):1671–80.
Sánchez-Sánchez BJ, Urbano JM, Comber K, Dragu A, Wood W, Stramer B, et al. Drosophila Embryonic Hemocytes Produce Laminins to Strengthen Migratory Response. Cell Reports. 2017;21(6):1461–70.
Broadie K, Baumgartner S, Prokop A. Extracellular matrix and its receptors in Drosophila neural development. Dev Neurobiol. 2011;71(11):1102–30.
Pastor-Pareja JC, Xu T. Shaping cells and organs in Drosophila by opposing roles of fat body-secreted Collagen IV and perlecan. Dev Cell. 2011;21(2):245–56.
Zang Y, Wan M, Liu M, Ke H, Ma S, Liu LP, et al. Plasma membrane overgrowth causes fibrotic collagen accumulation and immune activation in Drosophila adipocytes. Lemaître B, editor. eLife. 2015;4:e07187.
Morin X, Daneman R, Zavortink M, Chia W. A protein trap strategy to detect GFP-tagged proteins expressed from their endogenous loci in Drosophila. PNAS. 2001;98(26):15050–5.
Day CA, Kraft LJ, Kang M, Kenworthy AK. Analysis of protein and lipid dynamics using confocal fluorescence recovery after photobleaching (FRAP). Curr Protoc Cytom. 2012;Chap. 2:Unit2.19.
Lippincott-Schwartz J, Snapp EL, Phair RD. The Development and Enhancement of FRAP as a Key Tool for Investigating Protein Dynamics. Biophysical Journal. 2018;115(7):1146–55.
Rotstein B, Paululat A. On the Morphology of the Drosophila Heart. Journal of Cardiovascular Development and Disease. 2016;3(2):15.
Hayashi S, Kondo T. Development and Function of the Drosophila Tracheal System. Genetics. 2018;209(2):367–80.
Öztürk-Çolak A, Moussian B, Araújo SJ, Casanova J. A feedback mechanism converts individual cell features into a supracellular ECM structure in Drosophila trachea. eLife. 5:e09373.
Beckstead RB, Lam G, Thummel CS. The genomic response to 20-hydroxyecdysone at the onset of Drosophila metamorphosis. Genome Biol. 2005;6(12):R99.
de Oliveira AL, Calcino A, Wanninger A. Ancient origins of arthropod moulting pathway components. VijayRaghavan K, Martinez P, editors. eLife. 2019;8:e46113.
Bonnans C, Chou J, Werb Z. Remodelling the extracellular matrix in development and disease. Nat Rev Mol Cell Biol. 2014;15(12):786–801.
Hughes C. The Larval Requirement for Matrix Metalloproteinase-Mediated Remodelling of the Cardiac Extracellular Matrix in Drosophila melanogaster [Internet] [Thesis]. 2018 [cited 2019 Sep 18]. Available from: https://macsphere.mcmaster.ca/handle/11375/22878
Hughes CJR, Turner S, Andrews RM, Vitkin A, Jacobs JR. Matrix metalloproteinases regulate ECM accumulation but not larval heart growth in Drosophila melanogaster. J Mol Cell Cardiol. 2020;140:42–55.
Page-McCaw A, Ewald AJ, Werb Z. Matrix metalloproteinases and the regulation of tissue remodelling. Nature Reviews Molecular Cell Biology. 2007;8(3):221–33.
Glasheen BM, Robbins RM, Piette C, Beitel GJ, Page-McCaw A. A matrix metalloproteinase mediates airway remodeling in Drosophila. Developmental Biology. 2010;344(2):772–83.
Knorr E, Schmidtberg H, Vilcinskas A, Altincicek B. MMPs Regulate both Development and Immunity in the Tribolium Model Insect. PLOS ONE. 2009;4(3):e4751.
Page-McCaw A, Serano J, Santé JM, Rubin GM. Drosophila Matrix Metalloproteinases Are Required for Tissue Remodeling, but Not Embryonic Development. Developmental Cell. 2003;4(1):95–106.
Vishnuvardhan S, Ahsan R, Jackson K, Iwanicki R, Boe J, Haring J, et al. Identification of a Novel Metalloproteinase and Its Role in Juvenile Development of the Tobacco Hornworm, Manduca sexta (Linnaeus). Journal of Experimental Zoology Part B: Molecular and Developmental Evolution. 2013;320(2):105–17.
Birkedal-Hansen H. Proteolytic remodeling of extracellular matrix. Curr Opin Cell Biol. 1995;7(5):728–35.
Stamenkovic I. Extracellular matrix remodelling: the role of matrix metalloproteinases. The Journal of Pathology. 2003;200(4):448–64.
Wang Z, Li W, Chen S, Tang XX. Role of ADAM and ADAMTS proteases in pathological tissue remodeling. Cell Death Discov. 2023;9(1):1–6.
Arpino V, Brock M, Gill SE. The role of TIMPs in regulation of extracellular matrix proteolysis. Matrix Biology. 2015;44–46:247–54.
Boumiza S, Chahed K, Tabka Z, Jacob MP, Norel X, Ozen G. MMPs and TIMPs levels are correlated with anthropometric parameters, blood pressure, and endothelial function in obesity. Sci Rep. 2021;11(1):20052.
Van De Bor V, Loreau V, Malbouyres M, Cerezo D, Placenti A, Ruggiero F, et al. A dynamic and mosaic basement membrane controls cell intercalation in Drosophila ovaries. Development. 2021;148(4):dev195511.
Cevik D, Acker M, Michalski C, Jacobs JR. Pericardin, a Drosophila collagen, facilitates accumulation of hemocytes at the heart. Dev Biol. 2019;454(1):52–65.
Dobi KC, Schulman VK, Baylies MK. Specification of the somatic musculature in Drosophila. Wiley Interdiscip Rev Dev Biol. 2015;4(4):357–75.
Müller D, Jagla T, Bodart LM, Jährling N, Dodt HU, Jagla K, et al. Regulation and Functions of the lms Homeobox Gene during Development of Embryonic Lateral Transverse Muscles and Direct Flight Muscles in Drosophila. PLOS ONE. 2010;5(12):e14323.
Karunendiran A, Nguyen CT, Barzda V, Stewart BA. Disruption of Drosophila larval muscle structure and function by UNC45 knockdown. BMC Molecular and Cell Biology. 2021;22(1):38.
Cabral-Pacheco GA, Garza-Veloz I, Castruita-De la Rosa C, Ramirez-Acuña JM, Perez-Romero BA, Guerrero-Rodriguez JF, et al. The Roles of Matrix Metalloproteinases and Their Inhibitors in Human Diseases. Int J Mol Sci. 2020;21(24):9739.
Huang X. Single-cell profiling of the developing embryonic heart in Drosophila. 2023;
Mouw JK, Ou G, Weaver VM. Extracellular matrix assembly: a multiscale deconstruction. Nat Rev Mol Cell Biol. 2014;15(12):771–85.
Shi Y, Liu TY, Jiang HB, Liu XQ, Dou W, Park Y, et al. The Ecdysis Triggering Hormone System, via ETH/ETHR-B, Is Essential for Successful Reproduction of a Major Pest Insect, Bactrocera dorsalis (Hendel). Frontiers in Physiology [Internet]. 2019 [cited 2023 Jul 20];10. Available from: https://www.frontiersin.org/articles/10.3389/fphys.2019.00151
Davis JR, Ainslie AP, Williamson JJ, Ferreira A, Torres-Sánchez A, Hoppe A, et al. ECM degradation in the Drosophila abdominal epidermis initiates tissue growth that ceases with rapid cell-cycle exit. Current Biology. 2022;32(6):1285–1300.e4.
Matsubayashi Y, Sánchez-Sánchez BJ, Marcotti S, Serna-Morales E, Dragu A, Díaz-de-la-Loza M del C, et al. Rapid Homeostatic Turnover of Embryonic ECM during Tissue Morphogenesis. Dev Cell. 2020;54(1):33–42.e9.
Chinta R, Tan JH, Wasser M. The study of muscle remodeling in Drosophila metamorphosis using in vivo microscopy and bioimage informatics. BMC Bioinformatics. 2012;13(17):S14.
Maartens AP, Brown NH. The many faces of cell adhesion during Drosophila muscle development. Developmental Biology. 2015;401(1):62–74.
Jiang L. Cell-Mediated Branch Fusion in the Drosophila Trachea. Results Probl Cell Differ. 2024;71:91–100.
Scholl A, Ndoja I, Jiang L. Drosophila Trachea as a Novel Model of COPD. Int J Mol Sci. 2021;22(23):12730.
Shindo M, Wada H, Kaido M, Tateno M, Aigaki T, Tsuda L, et al. Dual function of Src in the maintenance of adherens junctions during tracheal epithelial morphogenesis. Development. 2008;135(7):1355–64.
Chioran A, Duncan S, Catalano A, Brown TJ, Ringuette MJ. Collagen IV trafficking: The inside-out and beyond story. Developmental Biology. 2017;431(2):124–33.
Sirokmány G, Kovács HA, Lázár E, Kónya K, Donkó Á, Enyedi B, et al. Peroxidasin-mediated crosslinking of collagen IV is independent of NADPH oxidases. Redox Biology. 2018;16:314–21.
Wilmes AC, Klinke N, Rotstein B, Meyer H, Paululat A. Biosynthesis and assembly of the Collagen IV-like protein Pericardin in Drosophila melanogaster. Biol Open. 2018;7(4).
Bogatan S, Cevik D, Demidov V, Vanderploeg J, Panchbhaya A, Vitkin A, et al. Talin Is Required Continuously for Cardiomyocyte Remodeling during Heart Growth in Drosophila. PLoS ONE. 2015;10(6):e0131238.
Matsubayashi Y. Dynamic movement and turnover of extracellular matrices during tissue development and maintenance. Fly (Austin). 2022;16(1):248–74.
Töpfer U, Guerra Santillán KY, Fischer-Friedrich E, Dahmann C. Distinct contributions of ECM proteins to basement membrane mechanical properties in Drosophila. Development. 2022;149(10):dev200456.
Urbano JM, Torgler CN, Molnar C, Tepass U, López-Varea A, Brown NH, et al. Drosophila laminins act as key regulators of basement membrane assembly and morphogenesis. Development. 2009;136(24):4165–76.
Stevens LJ, Page-McCaw A. A secreted MMP is required for reepithelialization during wound healing. Mol Biol Cell. 2012;23(6):1068–79.
Wei S, Xie Z, Filenova E, Brew K. Drosophila TIMP Is a Potent Inhibitor of MMPs and TACE: Similarities in Structure and Function to TIMP-3. Biochemistry. 2003;42(42):12200–7.
Pearson JR, Zurita F, Tomás-Gallardo L, Díaz-Torres A, Loza M del CD de la, Franze K, et al. ECM-Regulator timp Is Required for Stem Cell Niche Organization and Cyst Production in the Drosophila Ovary. PLOS Genetics. 2016;12(1):e1005763.
Deichsel H, Friedel S, Detterbeck A, Coyne C, Hamker U, MacWilliams HK. Green fluorescent proteins with short half-lives as reporters in Dictyostelium discoideum. Dev Genes Evol. 1999;209(1):63–8.
Lemke SB, Weidemann T, Cost AL, Grashoff C, Schnorrer F. A small proportion of Talin molecules transmit forces at developing muscle attachments in vivo. PLOS Biology. 2019;17(3):e3000057.
Vogler G, Ocorr K. Visualizing the beating heart in Drosophila. J Vis Exp. 2009;(31):1425.
Fluorescent Labeling Of Drosophila Heart Structures - Video [Internet]. [cited 2023 Jan 27]. Available from: https://www.jove.com/t/1423
Chang J, Garva R, Pickard A, Yeung CYC, Mallikarjun V, Swift J, et al. Circadian control of the secretory pathway maintains collagen homeostasis. Nat Cell Biol. 2020;22(1):74–86.
Cevik D, Acker M, Arefi P, Ghaemi R, Zhang J, Ravi Selvaganapathy P, et al. Chloroform and Desflurane Immobilization with Recovery of Viable Drosophila Larvae for Confocal Imaging. Journal of Insect Physiology. 2019;103900.
Wershof E, Park D, Barry DJ, Jenkins RP, Rullan A, Wilkins A, et al. A FIJI macro for quantifying pattern in extracellular matrix. Life Science Alliance [Internet]. 2021 Mar 1 [cited 2023 May 25];4(3). Available from: https://www.life-science-alliance.org/content/4/3/e202000880
Yost RT, Robinson JW, Baxter CM, Scott AM, Brown LP, Aletta MS, et al. Abnormal Social Interactions in a Drosophila Mutant of an Autism Candidate Gene: Neuroligin 3. International Journal of Molecular Sciences. 2020;21(13):4601.
Lenth RV, Bolker B, Buerkner P, Giné-Vázquez I, Herve M, Jung M, et al. emmeans: Estimated Marginal Means, aka Least-Squares Means [Internet]. 2023 [cited 2023 Apr 25]. Available from: https://cran.r-project.org/web/packages/emmeans/index.html
Bates D, Mächler M, Bolker B, Walker S. Fitting Linear Mixed-Effects Models Using lme4. Journal of Statistical Software. 2015;67:1–48.
Clarke E, Sherrill-Mix S, Dawson C. ggbeeswarm: Categorical Scatter (Violin Point) Plots [Internet]. 2022 [cited 2023 Apr 25]. Available from: https://cran.r-project.org/web/packages/ggbeeswarm/index.html
Wickham H. ggplot2: Elegant Graphics for Data Analysis [Internet]. New York, NY: Springer; 2009 [cited 2023 Apr 25]. Available from: https://link.springer.com/10.1007/978-0-387-98141-3

Supplementary Figure 1 to 5 are not available with this version.

Supplementary figure 1: Body wall muscles in the Drosophila third instar larva. Drosophila body wall muscles are segmentally repeated. Transverse body wall muscles 21, 22, and 23 within body segment A7 were imaged in FRAP experiments. Generated using BioRender.com.

Supplementary figure 2: Nested ROI measurements for more accurate assessment of Vkg-GFP fluorescence recovery. During the photobleaching period, the shape of the predetermined ROI (66X33 μm) becomes distorted by occasional movement of the animal and lateral movement of Vkg-GFP. To control for this, FRAP values are measured using a smaller ROI (33 x 33 mm), nested within the original 66 x 33 mm ROI. Average values for the immediately post-bleach and 24h-post bleach time points appear far lower when measured in the nested ROI relative to the larger original ROI. A one-tailed t-test was performed to determine the difference in fluorescence level between ROIs. The higher post-bleach values measured from the large ROI are likely due movement of tissue during photobleaching. As a result, some of the tissue measured within the ROI dimensions is unbleached. Because the smaller ROI of 33x33 mm is within the larger ROI and only covers bleached tissue, it is a more accurate measurement.

Supplementary figure 3: Collagen (ColIVa2) mRNA transcript levels throughout larval development and assessed via qPCR. Reaction norm plot of mRNA transcript levels at late second (LL2), early third (EL3) and mid-third (ML3) instar. Samples were pooled as follows: 10 larvae per LL2 sample; 5 larvae per EL3 and ML3 sample. For each gene and developmental stage, there were three biological replicates and three technical replicates of each biological replicate (N=3). qPCR was performed on a ThermoFisher Scientific StepOnePlus^TMinstrument. Analysis performed using R version 4.0.2 and libraries “lme4” and “emmeans”.

Supplementary figure 4: Driverless expression of UAS-timp decreases normalized fluorescence change relative to controls. Vkg-GFP fluorescence recovery is intermediate between control (Hand-GAL4,Vkg-GFP/yw) and timp OE (D). A one-way ANOVA with Tukey’s multiple comparisons test was used to evaluate differences in fluorescence change. The Kruskal-Wallis with Dunn’s multiple comparisons test was performed to determine differences in loss and gain. Scale bar in (A): 100 μm. N≥10.

Supplementary figure 5: Vkg-GFP fluorescence recovery is lower with homozygous vkg-GFP. Cardiac ECM in early to mid-third instar larvae of the genotypes yw; Hand-GAL4,Vkg-GFP/+ (heterozygous Vkg) (A-A’’), yw; Hand-GAL4,Vkg-GFP/Vkg-GFP (homozygous Vkg cross) (B-B’’), and yw; Vkg-GFP (homozygous Vkg) (C-C’’) pre, immediately post-bleach, and 24h post-bleach. Normalized change in Vkg-GFP fluorescence is not statistically different between larvae of the genotypes pictured (D). A one-way ANOVA paired with Tukey’s multiple comparisons test were used to determine differences in fluorescence change. The Kruskal-Wallis with Dunn’s multiple comparisons test was used to evaluate differences in loss and gain between genotypes. Scale bar in (A): 100 μm. Error bars represent SEM. N≥10.

No competing interests reported.

BMCBiology1MacDuffJacobs24SUPP.docx

Download PDF

Reviewers agreed at journal
11 Sep, 2024
Reviewers agreed at journal
05 Sep, 2024
Reviewers invited by journal
04 Sep, 2024
Editor assigned by journal
07 Aug, 2024
Submission checks completed at journal
07 Aug, 2024
First submitted to journal
06 Aug, 2024

You are reading this latest preprint version

A novel adaptation of FRAP quantifies the movement of Drosophila Basement Membrane Collagen in vivo

Status:

Version 1

Abstract

Figures

Introduction

Results

Novel FRAP does not alter organization of ECM proteins

Regulation of Vkg-GFP accretion is stage- and tissue-specific

Model for estimation of fluorescence loss and gain

Vkg-GFP turnover is stage- and tissue-specific

Altered matrix protease activity modifies Collagen turnover in the cardiac ECM

ECM and ECM-associated gene transcript levels vary little throughout larval development

Discussion

Vkg-GFP content of basement membranes varies with developmental time and tissue type

Vkg-GFP addition and loss are stage- and tissue- specific

ECM and ECM-associated gene transcript levels change throughout larval development

Altered matrix protease activity modifies Collagen turnover in the cardiac ECM

Conclusions

Experimental procedures

Fly stock maintenance and stocks

Larval collections

Larval freezing for qPCR

Larval dissections

Immunolabeling

Confocal imaging

Fluorescence recovery after photobleaching (FRAP) for live, intact larvae

Quantification of Pericardin fibre orientation

RNA extraction

cDNA synthesis

qPCR analysis

Abbreviations

Declarations

Author Contribution

Acknowledgement

References

Supplementary Figure

Additional Declarations

Supplementary Files

Status:

Version 1