Author response:
The following is the authors’ response to the original reviews.
Public Reviews:
Reviewer #1 (Public Review):
First, all the experiments are performed in Jurkat T cells that may not recapitulate the regulation of polarization in primary T cells.
To extend our results in Jurkat cells forming IS to primary cells, we have now performed experiments using synapses established by Raji cells and either primary T cells (TCRmediated) or primary CAR T cells (CAR-mediated) (new Suppl. Fig. S7). These experiments clearly show the presence of FMNL1 at these two different IS classes (new Suppl. Fig. S7), similar to what was found in Jurkat-Raji synapses. In addition, since most of the experiments were performed in Jurkat cells, we have changed the title of our manuscript, to be faithful to the main body of our results. New sentences dealing with this important issue have been included in the Results and Discussion sections.
Moreover, all the experiments analyzing the role of PKCdelta are performed in one clone of wt or PKCdelta KO Jurkat cells. This is problematic since clonal variation has been reported in Jurkat T cells.
Referee is right, this is the reason why we have studied three different control clones (C3, C9, C7) and three PKCdelta-interfered clones (P5, P6 and S4) all derived from JE6.1 clone and the results have been previously published (Herranz et al 2019)(Bello-Gamboa et al 2020). All these clones expressed similar levels of the relevant cell surface molecules and formed synaptic conjugates with similar efficiency (Herranz et al 2019). The P5, P6 and S4 clones exhibited a similar defect in MVB/MTOC polarization when compared with the control clones (Herranz et al 2019)(Bello-Gamboa et al 2020). Experiments developed by other researchers using a different clone of Jurkat (JE6.1) and primary CD4+ and CD8+ lymphocytes interfered in FMNL1 (Gomez et al. 2007), showed a comparable defect in MTOC polarization to that found in our control clones when were transiently interfered in FMNL1 (Bello-Gamboa et al 2020, this manuscript). In this manuscript we have studied, instead of canonical JE6.1 clone, C3 and C9 control clones derived from JE6.1, since the puromycin-resistant control clones (containing a scramble shRNA) were isolated by limiting dilution together with the PKCdelta-interfered clones (Herranz et al. 2019), thus C3 and C9 clones are the best possible controls to compare with P5 and P6 clones. Please realize that microsatellite analyses, available upon request, supports the identity of our C3 clone with JE6.1. Moreover, when GFP-PKCdelta was transiently expressed in the three PKCdelta-interfered clones, MTOC/MVB polarization was recovered to control levels (Herranz et al. 2019). Therefore, the deficient MTOC/MVB polarization in all these clones is exclusively due to the reduction in PKCdelta expression (Herranz et al 2019), and thus clonal variation cannot underlie our results in stable clones. We have now included new sentences to address this important point and to mention the inability of FMNL1betaS1086D to revert the deficient MTOC polarization occurring in P6 PKCdelta-interfered clone, as occurred in P5 clone. Due to the fact we have now included more figures and panels to satisfy editor and referees’s comments, we have not included the dot plot data corresponding to C9 and P6 clones to avoid a too long and repetitive manuscript. Since all the FMNL1 interference and FMNL1 variants reexpression experiments were performed in transient assays (2-4 days after transfection), there was no chance for any clonal variation in these short-time experiments. Moreover, internal controls using untransfected cells or Raji cells unpulsed with SEE were carried out in all these transient experiments.
Finally, although convincing, the defect in the secretion of vesicles by T cells lacking phosphorylation of FMNL1beta on S1086 is preliminary. It would be interesting to analyze more precisely this defect. The expression of the CD63‑GFP in mutants by WB is not completely convincing. Are other markers of extracellular vesicles affected, e.g. CD3 positive?
We acknowledge this comment. It is true that the mentioned results do not directly demonstrate the presence of exosomes at the synaptic cleft of the synapses, since the nanovesicles were harvested from the cell culture supernatants from synaptic conjugates and these nanovesicles could be produced by multi‑directional degranulation of MVBs. To address this important issue, we have performed STED super‑resolution imaging of the immune synapses made by control and FMNL1-interfered cells. Nanosized (100-150 nm) CD63+ vesicles can be found in the synaptic cleft between APC and control cells with polarized MVBs, whereas we could not detect these vesicles in the synaptic cleft from FMNL1-interfered cells that maintain unpolarized MVBs (New Fig. 10). New sentences have been included in the Results and Discussion dealing with this important point. Regarding the use of CD3 as a marker of extracellular vesicles, please realize that CD3 is neither an enriched nor a specific marker of exosomes, since it is also present in plasma membrane shedding vesicles, molting vesicles from microvilli, apoptotic bodies and small cell fragments, apart from exosomes, thus we have preferred to use the canonic exosome marker CD63 as a general exosome reporter readout, for WB and immunofluorescence (MVBs, exosomes), time-lapse of MVBs (suppl. Video 8) and super resolution experiments (Fig. 10).
Reviewer #2 (Public Review):
Summary:
The authors have addressed the role of S1086 in the FMNL1beta DAD domain in 4 F-actin dynamics, MVB polarization, and exosome secretion, and investigated the potential implication of PKCdelta, which they had previously shown to regulate these processes, in FMNL1beta S1086 phosphorylation. This is based on:
(1) the documented role of FMNL1 proteins in IS formation
(2) their ability to regulate F-actin dynamics
(3) the implication of PKCdelta in MVB polarization to the IS and FMNL1beta phosphorylation
(4) the homology of the C-terminal DAD domain of FMNL1beta with FMNL2, where a phosphorylatable serine residue regulating its auto-inhibitory function had been previously identified. They demonstrate that FMNL1beta is indeed phosphorylated on S1086 in a PKCdelta-dependent manner and that S1086-phosphorylated FMNL1beta acts downstream of PKCdelta to regulate centrosome and MVB polarization to the IS and exosome release. They provide evidence that FMNL1beta accumulates at the IS where it promotes F-actin clearance from the IS center, thus allowing for MVB secretion.
Strengths
The work is based on a solid rationale, which includes previous findings by the authors establishing a link between PKCdelta, FMNL1beta phosphorylation, synaptic F-actin clearance, and MVB polarization to the IS. The authors have thoroughly addressed the working hypotheses using robust tools. Among these, of particular value is an expression vector that allows for simultaneous RNAi-based knockdown of the endogenous protein of interest (here all FMNL1 isoforms) and expression of wild-‐‑type or mutated versions of the protein as YFP‐tagged proteins to facilitate imaging studies. The imaging analyses, which are the core of the manuscript, have been complemented by immunoblot and immunoprecipitation studies, as well as by the measurement of exosome release (using a transfected MVB/exosome reporter to discriminate exosomes secreted by T cells).
Weaknesses
The data on F-‐‑actin clearance in Jurkat T cells knocked down for FMNL1 and expressing wild-type FMNL1 or the non‑phosphorylatable or phosphomimetic mutants thereof would need to be further strengthened, as this is a key message of the manuscript. Also, the entire work has been carried out on Jurkat cells. Although this is an excellent model easily amenable to genetic manipulation and biochemical studies, the key finding should be validated on primary T cells
Referee’s global assessment is right. To extend our results in Jurkat cells forming IS, we have now performed experiments using synapses established by Raji cells and either primary T cells (TCR-mediated) or primary CAR T cells (CAR-mediated) (new Suppl. Fig. S7). These experiments clearly show the presence of FMNL1 at these two different IS classes (new Suppl. Fig. S7), similar to what was found in Jurkat-Raji synapses. In addition, since most of the experiments were performed in Jurkat cells, we have changed the title of our manuscript, to be faithful to the main body of our results. New sentences have been included in Results and Discussion to address these important points.
Recommendations for the authors:
Reviewer #1 (Recommendations For The Authors):
This study shows the role of the phosphorylation of FMNL1b on S1086 on the polarity of T lymphocytes in T lymphocytes, which is a new and interesting finding. It would be important to confirm some of the key results in primary T cells and to analyze in-depth the defect in actin remodeling (quantification of the images, analysis of some key actors of actin remodeling). The description of the defect in the secretion of extracellular vesicles would also benefit from a more accurate analysis of the content of vesicles.
Referee is right. We have now performed experiments using synapses containing Raji cells and either primary T cells (TCR-mediated) or primary CAR T cells (CAR-mediated) (new Suppl. Fig. S7). These experiments clearly show the presence of FMNL1 at these two different IS classes, similar to what was found in Jurkat-‐‑Raji synapses. Moreover, since most of the experiments were performed in Jurkat cells, we have changed the title of our manuscript, to be faithful to the main body of our results. Regarding the use of CD63 instead of other markers such as for instance, CD3 (as stated by the other referee), please realize that CD3 is neither an enriched nor a specific marker of exosomes, since it is also present in plasma membrane shedding vesicles, molting vesicles from microvilli, apoptotic bodies and small cell fragments, apart from exosomes, thus we have preferred to use the accepted consensus, canonic extracellular vesicle marker CD63 (International Society of Extracellular Vesicles positioning, Thery et al 2018, doi: 10.1080/20013078.2018.1535750. eCollection 2018., Alonso et al. 2011) as a general exosome reporter readout, for both WB, immunofluorescence (MVBs, exosomes) and super-resolution experiments. Accordingly, GFP-‐‑CD63 reporter plasmid was used for exosome secretion in transient expression studies and living cell time-lapse experiments (Suppl. Video 8). Any other exosome marker will also be present in Raji cells and will not allow to analyse exclusively the secretion of exosomes by the effector Jurkat cells, since B lymphocytes produce a large quantity of exosomes upon MHC‑II stimulation by Th lymphocytes (Calvo et al, 2020, doi:10.3390/ijms21072631). To reinforce the exosome data in the context of the immune synapse, STED super-resolution imaging of the immune synapses made by control and FMNL1‑interfered cells was performed. Nanosized (100-150 nm) CD63+ vesicles can be found in the synaptic cleft of control cells with polarized MVBs, whereas we could no detect these vesicles in the synaptic cleft from FMNL1-interfered cells that maintain unpolarized MVBs (new Fig. 10).
Moreover, all the videos are not completely illustrative. For example, in video 2 it would be more appropriate show only the z plane corresponding to the IS to see more precisely the F-actin remodeling relative to CD63 labeling.
Referee is right. It is true that the upper rows in some videos may distract the reader of the main message contained in the lower row, that includes the 90º turn-generated, zx plane corresponding to the IS interface. Accordingly, we have maintained the still images of the whole synaptic conjugates in the first row from video 2; this will allow the reader to perceive a general view of the fluorochromes on the whole cell conjugates, as a reference, and to compare precisely the F-actin remodeling relative to CD63 labeling only at the zx interface (lower row). We have now processed the videos 1 and 5 following similar criteria
The quality of videos 3 and 4 are not good enough. For video 7, it seems that the labeling of phospho-‐‑Ser is very broad at the IS, which is expected since it should label all the proteins that are phosphorylated by PKCs. The resolution of microscopy (at the best 200 to 300 nm) does not allow us to conclude on the co-‐localization of FMNL1b with phospho-‐‑Ser and is thus not conclusive. Finally, the study would benefit from a more careful statistical analysis. The dot plots showing polarity are presented for one experiment. Yet, the distribution of the polarity is broad. Results of the 3 independent experiments should be shown and a statistical analysis performed on the independent experiments
Referee is right, we have amended video settings (brightness/contrast) in videos 3 and 4 to improve this issue. In addition, we would like to remark that the translocation of proteins to cellular substructures in living cells is not a trivial issue, since certain protein localizations are too dynamic to be properly imaged with enough spatial resolution. The equilibrium resulting from the association/dissociation of a certain protein to the membrane, in addition to the protein diffusion naturally occurring in living cells, as well as signal intensity fluctuations inherent to the stochastic nature of fluorescence emission often provide barriers for image quality (Shroff et al, 2024). Thus, additional image blurring is expected when compared with that observed in fixed samples. However, we think it is important to provide the potential readers with a dynamic view of FMNL1 localization, which can only be achieved through real-time videos, in addition to the still frames from the same videos provided in Fig. 6A (the referee did not argue against the inclusion of these frames), together with images from fixed cells in Fig 6B, for comparison. This is the reason why we have preferred to maintain the improved videos to complement the results of some spare frames from the videos, together with images from fixed cells in the same figure (Fig. 6).
Regarding video 7, we agree that colocalization is limited by the spatial resolution of confocal microscopy, and this fact does not allow us to infer that FMNL1beta is phosphorylated at the IS. However, please realize we have never concluded this in our manuscript. Instead, we claimed that “colocalization of endogenous FMNL1 and YFP‑FMNL1βWT with anti‑phospho‑Ser …is compatible with the idea that both endogenous FMNL1 and YFP‑FMNL1βWT are specifically phosphorylated at the cIS”. Moreover, we have now performed colocalization in super‑resolved STED microscopy images, that reduces the XY resolution down to 30-40 nm (Suppl. Fig. S12), and the results also support colocalization of endogenous FMNL1 with anti-phospho‑Ser PKC at the IS within a 30 nm resolution limit. We have now somewhat softened our conclusion: “Although all these data did not allow us to infer that FMNL1β is phosphorylated at the IS due to the resolution limit of confocal and STED microscopes, the results are compatible with the idea that both endogenous FMNL1 and YFP-FMNL1βWT are specifically phosphorylated at the cIS”.
Regarding statistical analyses we agree the dot distribution in the polarity experiments is quite broad, but this is consistent with the end point strategy used by a myriad of research groups (including ourselves) to image an intrinsically stochastic, rapid and asynchronous processes such as immune synapse formation and to score MTOC/MVB polarization (Calvo et al 2018, https://doi.org/10.3389/fimmu.2018.00684). Despite this fact, ANOVA analyses have underscored the statistical significance of all the experiments represented by dot plot experiments. We cannot average or perform meta statistical analyses by combining the equivalent cohort results from independent experiments, since we have observed that small variations of certain variables (SEE concentration, cell recovery, time after transfection, etc.) affect synapse formation and PI values among experiments without altering the final outcome in each case. Please, note that our manuscript includes now 10 multi‑panel figures, 12 multi‑panel supplementary figures and 8 videos, and it is already quite large. Thus, we feel the inclusion of redundant, triplicate dot plot figures will dilute and distract to any potential reader from the main message of our already comprehensive contribution. We have now included new sentences at the figure legends to remark ANOVA analyses were executed separately in all the 3 independent experiments.
Reviewer #2 (Recommendations For The Authors):
(1) The key findings should be validated on primary CD4+ T cells (of which Jurkat is a transformed model).
Referee is right. However, as commented by the other referee, the data from activating surfaces clearly shows that the synaptic actin architecture of the immune synapse from primary CD8+ T cells is essentially indistinguishable and thus unbiased from that of Jurkat T cells, but different to that of primary CD4+ cells (Murugesan, 2016). Thus, our data in Jurkat T cells are directly applicable to the synaptic architecture of primary CD8+ cells. In addition, to definitely extend our results in Jurkat cells forming IS, we have performed experiments using synapses established by Raji cells and either primary T cells (TCR-mediated) or primary CAR T cells (CAR-mediated) (new Suppl. Fig. S7) challenged by Raji cells. We have preferred to work with mixed CD4+ and CD8+ cells in order to maintain potential interactions in trans between these subpopulations that may affect or influence IS formation. These experiments clearly show the presence of FMNL1 at these two different IS classes (new Suppl. Fig. S7), similar to what was found in JurkatRaji synapses. Moreover, since most of the experiments were performed in Jurkat cells as stated by the referee, we have changed the title of our manuscript, to circumscribe our results to the model we have used and to be faithful to the main body of our results.
(2) The image of wt YFP-FMNL1beta in Figure 4A displays a weak CD63 signal and shows an asymmetric polarization of both the centrosome and MVBs. It should be replaced with a more representative one.
Referee is right. Accordingly, we have modified the CD63 channel settings (brightness/contrast) in this panel to make it comparable to the other panels in the same figure. In addition, thanks to this referee´s comment, we have realized the position of the MTOC (yellow dot) in the diagram in the right side of the YFP-FMNL1betaWT panels row appeared mislocated, producing the mentioned apparent asymmetry with respect to MVBs’s center of mass (green dot) position. This mistake leads to an apparent segregation between the position of the center of mass of these organelles which certainly does not correspond with the real image. We have now amended the scheme and we apologize for this mistake.
(3) The images showing F-actin clearance at the IS (Figure 8, S4, S5) are not very convincing, also when looking at the MFI along the T cell-‐‑APC interface in the en-‐face views. Since the F-actin signal also includes some signal from the APC, transfecting T cells with an actin reporter to selectively image T cell actin could better clarify this key point.
Referee´s point is correct. However, we (83), and other researchers using the proposed actin reporter approach in the same Raji/Jurkat IS model (Fig. 4 in ref 84) have already excluded the possibility that actin cytoskeleton of Raji cells can also contribute to the measurements of synaptic F-actin. In Materials and Methods, page 37, lines 1048-1055 we included this related sentence: ¨It is important to remark that MHC-II-antigen triggering on the B cell side of the Th synapse does not induce noticeable F-actin changes along the synapse (i.e. F-actin clearing at the central IS), in contrast to TCR stimulation on T cell side (84) (85) (3). In addition, we have observed that majority of F‐‑actin changes along the IS belongs to the Jurkat cell (83). Thus, the contribution to the analyses of the residual, invariant F‐actin from the B cell is negligible using our protocol (83).
Thus, we can exclude this caveat may affect our results.
(4) A similar consideration applies to the MVB distribution in the en‑face images. For example, in Figure S5 the MVB profile, with some peripheral distribution, does not appear very different in cells expressing wt YFP‑tagged FMNL1beta versus the S1086A‑expressing cells.
The referee's assessment regarding Supp. Figure S5 is valid. Using only the plot profile, the outcomes obtained with YFP-FMNL1βWT may appear comparable to those derived from YFP-FMNL1βS1086A. Nonetheless, this resemblance is attributed to the plot profile's exclusive consideration of the MVBs signal in the interface from the immune synapse region (white rectangle). The upper images (second row), where the whole cell is displayed, illustrate that in YFP-FMNL1βWT, MVB are specifically accumulated within this specific region, in contrast to the scattered distribution observed in YFP-FMNL1βS1086A, where MVB are dispersed throughout the cell without distinction. While MVBs are evident in both instances within the synapse region, the reason behind this observation is different. The YFP-FMNL1βWT transfected cell (third column) shows a pronounced MVB concentration within the synaptic area (white rectangle), which leads to MVB PI=0.52, whereas the YFP-FMNL1βS1086A transfected cell (fourth column), as it presents a scattered distribution of MVB throughout the cell, also exhibits some MVB (but only a small proportion of the total cellular MVB) in the synaptic area, which yields MVB PI=-0.09. Please realise that the position of the center of mass of the distribution of MVB (MVBC) labelled in this figure (white squares) is an unbiased parameter that mirrors MVB center of mass polarization. A new sentence has been included in the figure legend to clarify this important point.
(5) The image in the first row in Figure 6B does not show a clear accumulation of FMNL1beta at the IS, possibly because the T cell is in contact with two APCs. This image should be replaced.
Referee is right Therefore, we have replaced the quoted example with a single cell:cell synapse that shows a clearer and more localized accumulation in the cIS, thereby avoiding the mentioned caveat.
(6) In Figure 2A the last row shows what appears to be a T:T cell conjugate (with one cell expressing the YFP-‐‑tagged protein). The image should be replaced with another showing a T cell-APC (blue) conjugate.
Referee is right, we have accordingly replaced the mentioned image with a T cell:APC conjugate.
(7) The Discussion is very long and dispersive. It would benefit from shortening it and making it more focused.
Referee is right, we have shortened and focused it, by eliminating the whole second and third paragraphs of the discussion. Moreover, a whole paragraph in page 24 has been also deleted.
We have also focussed the discussion towards the new data in primary T lymphocytes.
