FLARIM v2.0, an improved method to quantify transcript-ribosome interactions in vivo in the adult Drosophila brain

Neural injury triggers striking immune reactions from glial cells, including significant transcriptional and morphological changes, but it is unclear how these events are coordinated to mount an effective immune response. Here, we present a new variant of the Fluorescence assay to detect ribosome interactions with mRNA (FLARIM), which we term FLARIM v2.0, to visualize single immune gene transcripts and association with ribosomes in glia responding to neurodegeneration. Specifically, using an in vivo axotomy assay in Drosophila, we show that matrix metalloproteinase-1 (Mmp-1) mRNAs and associated ribosomes are detected in distal processes of reactive glia where they are actively engulfing degenerating axonal material, suggesting that local translation is an important component of the glial immune response to axotomy. This work also validates our enhanced FLARIM assay as a promising tool to investigate mechanisms of mRNA transport and translation in a wide range of in vitro and in vivo paradigms.

mechanisms of mRNA transport and translation in a wide range of in vitro and in vivo paradigms.

23 24
Introduction 25 26 Glia are the resident immune cells of the brain and respond swiftly to neuronal trauma, 27 pathogenic insult, and degeneration(1-3). Following neuronal damage, activated glia undergo 28 distinct transcriptional, morphological, and functional changes(4-6). In many cases, reactive glia 29 are neuroprotective, releasing pro-survival factors and clearing damaged neurons through 30 phagocytic engulfment (7,8). Thus, understanding how glial immune responses are activated and 31 carried out will offer new insight into approaches that could delay the onset or progression of a 32 range of neurodegenerative disorders and injury conditions.

33
The fruit fly, Drosophila melanogaster, offers a powerful genetic in vivo model to explore 34 evolutionarily conserved glia-neuron signaling events after neural injury(9-11). For example, after  raises interesting questions about how key immune molecules are released within the neuropil at injury sites (12,24). Previous work in other model systems has indeed demonstrated that directed 48 mRNA transport and local translation are important for glia to carry out normal functions. For example, in oligodendrocytes, myelin basic protein transcripts are localized to oligodendrocyte 50 processes to adequately myelinate axons in an activity-dependent manner, while astrocytes have 51 been shown to influence interactions at tripartite synapses and the gliovascular interface through 52 a subset of discrete localized and locally translated transcripts (25)(26)(27)(28)(29).

53
In order to further explore the transcriptional and translational changes that are essential 54 for proper glial responses to damaged axons, our lab has utilized various single molecule 55 fluorescence in situ hybridization (smFISH) techniques for the detection of individual transcripts 56 and a new variation of Fluorescence assay to detect ribosome interactions with mRNA (FLARIM), 57 referred to as FLARIM v2.0, to detect ribosome association with mmp-1 transcripts in glial cells 58 following nerve injury(30). smFISH has been employed in glial cells and various Drosophila tissues 59 to detect individual mRNAs; however, previous studies have not yet employed mRNA and 60 ribosome detection together in the context of neuronal injury in vivo (31)(32)(33)(34)(35)(36)(37)(38). Here, we have utilized 61 single molecule inexpensive FISH (smiFISH), hybridization chain reaction (HCR), and FLARIM, in 62 order to visualize mmp-1 transcript localization and translation in reactive glia following 63 axotomy (30,(39)(40)(41). Our findings reveal that mmp-1 transcripts are localized to and associate 64 with ribosomes at distal glial processes at injury sites, suggesting that local translation of Mmp-65 1 may be an important mechanism by which glia access and phagocytically clear neuronal debris.

69
Ensheathing glia respond to olfactory receptor neuron injury in Drosophila

70
Axotomy in the adult Drosophila antennal system is a well-characterized model to

98
In order to accomplish this, we utilized smiFISH to detect the mmp-1 transcript following injury

102
As a result, we used this as a starting point for our smiFISH studies.

103
In accordance with our qPCR and RNAseq data, we found that smiFISH reveals a low level 104 of Mmp-1 transcript present in the uninjured brain, with the highest amount of transcript being 105 expressed in cells surrounding the antennal lobe, likely ensheathing glia (Fig. 1D). In response to 106 injury, we observe a significant increase in the amount of mmp-1 transcript detected at 4.5hpi 107 ( Fig. 1D''). Interestingly, when labeling ensheathing glial membranes with GFP, we also observed 108 the localization of the mmp-1 smiFISH signal to fine processes following injury (Fig. 1E).

109
To determine if the Mmp-1 probeset is specific to the mmp-1 transcript, we performed an

117
The FLARIM v2.0 method can be used to detect mRNAs and associated ribosomes

118
While the smiFISH approach allowed us to localize the mmp-1 transcript to distal glial 119 processes following neuronal injury, it does not give us insight into whether there may be local    1) an 18S ribosomespecific probe with the first half of a split B3 initiator and an mRNA-specific probe with the second half of the split B3 initiator and 2) a full B2 initiator on the 5' end of the gene specific probe. This B2 initiator allows for transcript localization and aids in quantification by creating a normalization channel for the FLARIM signal Hybridization with these probesets and amplification using metastable hairpins generates a separate fluorescent signal for mRNA and mRNA-ribosome association detection in vivo. Created using Biorender.com.
probes and the gene-specific probes are reconstituted, a second set of hairpins are opened, 150 creating a distinct fluorescent signal to allow for mRNA-ribosome association detection (Fig. 2B).

151
The dual detection method produces diffraction limited spots, if amplification is carried ribosome-association signals significantly increased within antennal lobes 20hpi (Fig. 3B).

175
A series of control experiments were also conducted to ascertain the specificity of the 176 probesets and our signal. In the absence of hybridized probes, the hairpins amplified alone do not 177 generate a fluorescent signal (Fig. 3A, B). Additionally, hybridization of only the 18S FLARIM 178 probes does not result in a signal, as only half of the initiator sequence is present, and hairpins

179
are unable to open and generate a fluorescent signal (Fig. 3A, B). We also determined that the 180 mmp-1 probeset was specific to our transcript of interest. Upon Mmp-1 RNAi knockdown in glial 181 cells, mmp-1 mRNA and ribosome-association signals were significantly diminished in the injured 182 condition (Fig. 4A, B, C).

184
Translation machinery is present in ensheathing glial processes

185
We observe that mmp-1 transcripts and associated ribosomes are localized to 186 ensheathing glial processes following ORN injury. As a result, we hypothesize that transcript

212
Thus, we propose that this novel methodology to visualize and quantify transcripts and, notably, 213 ribosome association in whole tissue will be broadly valuable.

227
Here, we show that that our mmp-1 FLARIM v2.0 approach can reliably detect significant 228 mmp-1 upregulation in glia following olfactory nerve injury and that this strategy allows us to 229 monitor changes in the spatial distribution of mmp-1 transcripts, as well as association with 230 ribosomes within one day after axotomy. We propose that our detection of transcript/ribosome 231 association indicates that local translation of secreted Mmp-1 protein occurs at the distal 232 process of glial cells as they invade injury sites (Fig. 3). We observe markers for both ER and GA 233 in fine distal processes of ensheathing glia under both basal and injury conditions, suggesting 234 that this glial subtype is equipped with organelles to locally translate and secrete a released 235 factor such as Mmp-1 (Fig. 5). Given the ramified morphology of ensheathing glia, and the well-     The smiFISH reagents, flap hybridization, and protocol have been described previously (37,40).

292
The smiFISH protocol has been adapted as follows: After dissection, fixation, and

300
The HCR v3.0 hybridization and amplification protocol has been described previously for whole-301 mount fruit fly embryos(41, 67). The protocol was utilized in adult fruit fly brains as follows: After 302 permeabilization, brains were pre-hybridized for 20min at 37˚C in probe hybridization buffer 303 (Molecular Instruments), which had already been heated to 37˚C. Brains were then hybridized 304 overnight (16-18hrs) with the gene-specific and 18S FLARIM probes in a thermal cycler at 37˚C.

305
The following day, brains were washed in probe wash buffer (