Dynamic proteomic and phosphoproteomic atlas of corticostriatal axons in neurodevelopment

  1. Vasin Dumrongprechachan
  2. Ryan B Salisbury
  3. Lindsey Butler
  4. Matthew L MacDonald
  5. Yevgenia Kozorovitskiy

  • Curated by eLife

    eLife logo

    Evaluation Summary:

    Knowledge of the protein composition of defined sub-cellular compartments is of key importance for the characterization of protein machines that mediate defined cellular functionalities. The current paper presents a novel mouse line that will serve as a helpful tool in this context - a Cre-inducible APEX2 reporter mouse line for acute ex-vivo proximity biotinylation. The paper documents the successful use of the novel reporter line to assess circuit-specific proteomes and phosphoproteomes in the corticostriatal system during development. The corresponding data largely align with the published record, but potentially new biological insights deduced from bioinformatic analyses of proteomic data were not followed up by experimental validation. In sum, the new APEX2 reporter mouse line will be of substantial interest to researchers in many fields of mammalian biology. The extent of 'new biology' provided is rather limited, but will be of interest to readers in neurodevelopment.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 agreed to share their name with the authors.)

Reviewed by

Read the full article See related articles

Listed in

Log in to save this article

Abstract

Mammalian axonal development begins in embryonic stages and continues postnatally. After birth, axonal proteomic landscape changes rapidly, coordinated by transcription, protein turnover, and post-translational modifications. Comprehensive profiling of axonal proteomes across neurodevelopment is limited, with most studies lacking cell-type and neural circuit specificity, resulting in substantial information loss. We create a Cre-dependent APEX2 reporter mouse line and map cell-type-specific proteome of corticostriatal projections across postnatal development. We synthesize analysis frameworks to define temporal patterns of axonal proteome and phosphoproteome, identifying co-regulated proteins and phosphorylations associated with genetic risk for human brain disorders. We discover proline-directed kinases as major developmental regulators. APEX2 transgenic reporter proximity labeling offers flexible strategies for subcellular proteomics with cell type specificity in early neurodevelopment, a critical period for neuropsychiatric disease.

Article activity feed

  1. Version published to 10.7554/elife.78847 on eLife
  2. eLife

    Evaluation Summary:

    Knowledge of the protein composition of defined sub-cellular compartments is of key importance for the characterization of protein machines that mediate defined cellular functionalities. The current paper presents a novel mouse line that will serve as a helpful tool in this context - a Cre-inducible APEX2 reporter mouse line for acute ex-vivo proximity biotinylation. The paper documents the successful use of the novel reporter line to assess circuit-specific proteomes and phosphoproteomes in the corticostriatal system during development. The corresponding data largely align with the published record, but potentially new biological insights deduced from bioinformatic analyses of proteomic data were not followed up by experimental validation. In sum, the new APEX2 reporter mouse line will be of substantial interest to researchers in many fields of mammalian biology. The extent of 'new biology' provided is rather limited, but will be of interest to readers in neurodevelopment.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #2 agreed to share their name with the authors.)

  3. eLife

    Reviewer #1 (Public Review):

    The aim of the present study was to develop and validate a novel mouse model that allows to determine the proteome of defined sub-cellular compartments, and to use this model in order to elucidate the molecular processes that govern the establishment of synaptic contacts between cortical and striatal neurons in the brain. Given that knowledge of the protein composition of defined sub-cellular compartments is of key importance for the characterisation of protein machines that mediate defined cellular functionalities, the establishment of corresponding mouse models to study such issues is of major general interest. The same is true for the development and function of cortico-striatal connectivity in the brain, which plays key roles in multiple major brain processes and is perturbed in many neuropsychiatric disorders.

    The major strength of the present paper is that it presents a novel mouse line that promises to serve as a very helpful tool in this context. The authors generated a KI mouse line that expresses APEX2 under the control of a Cre-activatable promoter from the ROSA26 locus, and they show convincingly that this new mouse line, upon crossing with corresponding Cre-expressing driver lines, allows the identification of cell-sub-compartment specific proteomes and phosphoproteomes - via APEX2-mediated proximity biotinylation, tissue dissection, protein affinity purification, and mass spectrometric analysis.

    The biological context of the present study is less convincingly established. Focussing on neuronal connections between the cerebral cortex and the striatum, bioinformatic analyses of corresponding datasets pinpoint a selection of axon guidance systems and protein kinase cascades to play roles in the development of cortico-striatal connectivity. The corresponding data partially align with the published record, but potentially new biological insights deduced from bioinformatic analyses of proteomic data were not followed up by experimental validation.

    In sum, the new APEX2 reporter mouse line reported in the present paper will likely be of substantial interest to researchers in many fields of mammalian biology, but the extent of 'new biology' provided in the present study is very limited.

  4. eLife

    Reviewer #2 (Public Review):

    This work by Dumrongprechachan and colleagues establishes a new method to isolate and quantify projection-specific axon proteomes using a genetically-targeted biotin ligase approach in the mouse brain. The authors developed a new mouse line that expresses CRE-dependent APEX2 to genetically target proteomes in specific cell types for acute ex vivo biotin labeling. They use this method to interrogate axon proteomes and phosphoproteomes during the development of corticostriatal projections in mice, providing rich datasets that can be mined in the context of future studies into brain circuit development.

    This work is particularly useful for the new Cre-dependent biotin ligase mouse and projection proteomics methodological pipeline. The flexed Rosa26 Apex2 mouse line is robustly characterized and will no doubt find many applications in circuit-specific proteomics.

    The authors use their new mouse line to express APEX2 in layer 4/5 corticostriatal projection neurons. They demonstrate biotinylation in the striatum from cortical axons at P5, P11, P20, and P50 to investigate axonal proteins in that developmental time window. They highlight several known developmentally regulated proteins, and find some of these projection proteins to be known risk genes in neurodevelopmental and neuropsychiatric disorders such as autism spectrum disorder, bipolar disorder, and epilepsy. They give examples on how to use the data they produce, showing that the Netrin1-DCC and mTOR pathways are differentially regulated at different ages. They also characterize the phosphoproteome of corticostriatal axons, showing that not only protein abundance, but protein activity changes over development, with Tsc2 as an example. These experiments establish a new technique to interrogate axon proteomics in the developing brain.

    Major Strengths:
    The conclusions from these data are well supported by the results, and the approaches used are well characterized, including the APEX2 mouse reporter line created for this study. The data support the claims that this animal model can be used to biotinylate proteomes with high cell-type and temporal specificity.

    Figures are well made and easy to interpret. There are clear images showing cell-type specificity of the reporter line, and western blots are conclusive. The authors investigated off-target effects of Cas9 activity sufficiently.

    Looking at the projection phosphoproteome is novel and offers mechanistic implications for axon development. The correlation analysis of phosphopeptide and protein abundance is very interesting. Adding decorrelated phosphosites as a new biologically-relevant variable is innovative for big data approaches. Workflows established in this manuscript can be used in the analysis of other circuits in the developing brain and the datasets provided can be mined for future studies.

    There are implications from the highlighted data regarding the presence and activity of mTOR in axons that are particularly interesting in the context of projection neuron development and regeneration. The authors may comment on how their data can be interpreted regarding mTOR in developing axon projections and mature presynaptic terminals.

    Minor Weaknesses:
    Although both male and female animals are used in the study, the authors do not note any sex differences in the proteome or phosphoproteome during development. If such data are available, seeing changes in proteomics and phosphoproteomics across sexes during development would be very interesting. If there are no observed differences, adding a sentence to clarify that would be helpful.

    When establishing proteins that are enriched in somata vs. axons, a soma prep from the same APEX mouse line would have been more appropriate as a control compared to the virally overexpressed Histone 2B-APEX control used, which would label proteins sequestered only to the nucleus. The authors should qualify in the text that proteome differences they see in these two datasets do not arise solely from somatic versus axonal enrichment, but also from the confounding differences of viral versus Rosa26-locus expression levels, and cytosolic versus histone-fused Apex.

    While there are a series of novelties in this work, overt "first ever" statements in the text (e.g. lines 94-96, 348-349) are redundant and inherently ambiguous in their factuality. The work's novelty is better served to speak for itself.

  5. eLife

    Reviewer #3 (Public Review):

    In this work, Dumrongprechachan et al. impressively expanded their earlier work on the identification of cell type-specific subcellular proteomes from mouse brain by APEX2 proximity labeling. Instead of using viral expression of APEX2, the authors now created a Cre-dependent APEX2 reporter mouse line using CRISPR knock-in, which can be combined with multiple Cre-driver lines for proteomic applications. Using this novel tool in combination with sophisticated mass spectrometry and elegant bioinformatics, they mapped the temporal dynamics of the axonal proteome in corticostriatal projections (instead of only identifying a static cell type- and compartment-specific proteome) together with its phosphorylation status (instead of only looking at protein abundance). The data will provide a valuable resource on developmental trajectories at the proteomic and phosphoproteomic level, and will allow for pathway- and phosphosite-centric systems-level analyses as exemplified by the identification of proline-directed protein kinases as major regulators of corticostriatal projection development.

    Strengths:
    The key tool developed in this work is the APEX2 reporter mouse line as it enables capturing of early postnatal time points, which was not possible before due to the time window of 2-4 weeks required for viral APEX expression. Thus, this tool puts the authors into position to access the temporal dynamics of the developing axon at time points spanning from neonate (as early as P5) to young adult (P50). Within this complex experimental design, the authors even managed to introduce a crucial compartment control at least for the time point P18, in which APEX expression is restricted to nucleus and soma upon viral expression. The resulting resource will be of high value as the data are derived from advanced mass spectrometric methods and stringent data handling. Examples of this high level of scrutiny include the use of MS3 methodology for the acquisition of TMT data to address the ratio distortion issues typically seen with isobaric labeling and thereby increase the quantification accuracy and the limitation to proteins quantified in all biological replicates.

    Weaknesses:
    As to sample preparation for mass spectrometry, the authors follow the interesting concept of first enriching the phosphopeptides from the pool of TMT-labeled tryptic peptides and then using the unbound fraction from that step for further peptide fractionation, followed by mass spectrometric protein quantification. While this strategy sounds very straightforward in principle, one would expect that the phosphopeptide enrichment comes with an unspecific loss of other peptides in general, and with a semi-specific loss of acidic peptides in particular. Was this potential issue investigated by comparison with samples that were fractionated directly without prior phosphopeptide enrichment? Or with other words: the rationale for this sequential procedure is compelling - quantification of both protein and phosphopeptide abundance from the same (limited) sample, but what is the price for it as to peptide loss?

    The APEX2 reporter mouse line is a novel tool with broad applicability for proximity labeling approaches and, understandably, the authors advertise its advantages, mainly via the suitability for short temporal windows. However, the discussion on the limitations of the approach falls short. The authors should make clear that the APEX method in general is limited to ex vivo approaches such as the acute brain slices used here due to the limitation that potentially toxic reagents (i.e. low membrane-permeable biotin-phenol and H2O2) have to be delivered to the target tissue. Although treatment with H2O2 is rather short, undesired oxidative stress signaling may have to be taken into account, particularly when protein phosphorylation rather than protein abundance is assessed. It would also be interesting to discuss the pros and cons of perfusing the mice prior to preparation of brain slices; e.g., in the context of removal of catalases/endogenous peroxidases or potential for substrate delivery (like recently shown in heart, doi: 10.1038/s41586-020-1947-z). Another issue with the Discussion is that the authors do not properly reflect the involvement of proline-directed kinases in the development of corticostriatal projections, which stands in contrast to the fact that they sell this as one of their major findings throughout the manuscript, including the Abstract.

  6. Version published to 10.1101/2022.03.21.485234v1 on bioRxiv