Quantitative determination of the spatial distribution of components in single cells with CellDetail

The dipole moment is the underlying concept of the algorithm of CellDetail

The polarity quantification algorithm of CellDetail is modeled on the molecular dipole moment P usually applied to molecules like H₂O (polar) or CO₂ (apolar) (Fig. 1c, left). The dipole moment is defined as the absolute value of charge δ (total positive, or respectively, negative charge) multiplied with the distance vector d between positive and negative charge center (R_δ+ blue cross, R_δ− red cross): P = |δ| · d. This concept was transferred to fluorescence images of biomolecules in cells (Fig. 1c, right). The distance vector between charge-weighted positive and negative charge centers, multiplied by the absolute value of charge, provides the value for the dipole moment which is considered as a measure for the polarity of the distribution of the component under investigation. Regions of low biomolecule density form the negative charge center R₋ (blue cross), regions with a high biomolecule density form the positive charge center R₊ (red cross). By multiplying the total amount of positive charges q with the vector d (= distance vector, green) between the charge centers, the resulting dipole moment P is a value for the polarity of the distribution of the fluorescence intensity (aka the biomolecule) within the cell. For 3D analyses, the charge centers as well as the positive charges are calculated across all layers together, not per single layer. See also Supplementary Information F, Algorithm.

CellDetail follows a sequential order of analysis steps (Fig. 1d): (i) cell detection is performed on either an image stack or single images. Single layer or z-stacked tiff images of single cells with a fluorescence signal are used as input. Cell detection is included in CellDetail. It is based on Otsu’s method for automatized thresholding. Then, voxels are separated into background voxels and cell voxels (cell detection). Subsequently, the average intensity of background voxels is calculated and subtracted from the cell voxel intensities to obtain background corrected voxels for each cell and the intensity of the voxels is normalized to the average cell intensity per voxel being 1. Next, the average intensity value of voxels is calculated and subtracted from voxel intensities leading to positive and negative voxel values, resulting in the charges q₊ and q₋, then the charge centers R₊ and R₋ and subsequently the dipole moment P is calculated in 3D and normalized (Fig. 1e and Supplementary Information F, Algorithm).

In chemical molecules like H₂O (polar) or CO₂ (apolar), both the distance between charge centers as well as the amount of charge are not variable within and among molecules (aka each molecule looks alike). In contrast, each cell is distinct in size and shape, so maximal distance of charge centers is also dependent on the cell size. The amount of the charge is further dependent on the overall amount of the fluorescence signal (total protein of interest content) and the microscope settings (bit-depth with which the image was acquired). Normalizations of the distance (Supplementary Information F Fig. S9) and charge and protein content (Supplementary Information F Fig. 10) are therefore required to allow to compare polarity values among cells. To this end, the absolute value of the dipole moment P is normalized (Supplementary Information F) to result in the normalized dipole moment P_n. Supplementary Information F Table S2 and Figs. S11–S15) provide detailed information on the normalization process. Decision trees on normalization options to guide choices of normalization option for distinct types of images/distributions are provided. CellDetail includes two options for distance normalization, five options for charge normalization and two options for protein content normalization. P_n is the central output parameter of CellDetail (Fig. 1e). CellDetail works in general in 3D, but can also be used for single layer images (2D).

CellDetail provides the following output parameters (Fig. 1f): (i) the normalized dipole moment of a a biomolecule, P_n, (ii) the amount/intensity of the biomolecule in the cell, (iii) the spatial volume of the biomolecule in the cell, (iv) the vectors between charge centers R₊, R₋ and the position M of the center point of the cell which are required for the determination of correlations of biomolecule 1 to biomolecule 2 position like angles and distances, (v) a distance describing the constriction of positive and negative charge center to a subcell region or across a cell to see whether polarity is constricted to a subcell structure and (vi) co-localization parameters like voxel intensity-based Pearson correlation coefficients across channels/molecules. CellDetail has a graphical user interface (GUI) (https://github.com/xyq91/CellDetail-TS for downloading CellDetail, Supplementary Information H Manual). Output files of CellDetail are.txt, .xls and .mat files. A full list and description of all output parameters is listed in the CellDetail manual (Supplementary Information H Manual). Robustness of CellDetail concerning brightness variation, SNR variation, oversaturation and shape differences was confirmed (Supplementary Information C) and the distribution function of P_n was mathematically derived (Supplementary Information I).

CellDetail correctly separated distinct HSCs based on 3D polarity/apolarity that were visually identified to be polar and apolar for Cdc42 and used as the reference data set (Fig. 1g, h). More importantly, in-depth benchmarking analyses that involved other currently available polarity analysis tools or algorithms demonstrated that CellDetail performed in general with a higher level of accuracy in separating image layers of HSCs visually scored for being polar or apolar for Cdc42, while being similar to the accuracy of the barycenter method (Supplementary Information B, Fig. S2 and Table S1).

Polarity of Cdc42 and Tubulin in young, old and old rejuvenated HSCs

HSCs from young mice show a high frequency of cells polar for the polarity protein Cdc42 or for Tubulin^2,34. Polarity in HSCs is tightly linked to an asymmetric division of HSCs. An asymmetric division ensures proper stem cell homeostasis¹⁹. Upon aging, the frequency of HSCs with a polar distribution of these proteins decreases, and aged HSCs are primarily apolar. Apolarity is a critical hallmark of aged stem cells^2,35,36,37.

The identification of polarity in HSCs has been so far primarily performed by visual examination of 2D or 3D images (Fig. 2a, Supplementary Movie 1 and Supplementary Movie 2)². We set out to quantify the change in the polarity values P_n of Cdc42 and Tubulin in 3D (stained via established IF protocols²) in individual young and old HSCs. This data might also serve as additional validation test for CellDetail. We analyzed z-stacked confocal microscopy images as well as z-stacked widefield microscopy images of individual young or old HSCs (Fig. 2a, b left, one layer shown). For confocal imaging, the medians of P_n of Cdc42 or Tubulin of individual HSCs were elevated in young compared to old HSCs (Fig. 2a right), with a significant difference of the median of P_n of Cdc42 between young and old HSCs. For widefield imaging, the medians of P_n of Cdc42 or Tubulin of individual HSCs were also elevated in young compared to old HSCs (Fig. 2a right), with a significant difference of the median of P_n of Cdc42 and Tubulin between young and old HSCs. Overall, significant differences showed an effect size in the range from a medium to a small effect. These data recapitulate a reduced overall polar distribution of especially Cdc42 in old compared to young HSCs. P_n values of 3D IF images of HSCs (here all z-stacks of the IF images, Fig. 2c) successfully align distinct levels of P_n values with visually scored types of polarity. As semi-automatic acquisition of widefield imaging was established in our laboratory, we focused in all subsequent experiments on widefield IF images.

a Left: Confocal IF images (example from z-stacks) of Cdc42, Tubulin and DAPI (scale bar 5 µm). Right: Absolute value of P_n, with median as central mark, edges indicating 25th and 75th percentiles, whiskers extending to most extreme data points not considered outliers, outliers individually plotted. Median Cdc42: P_n,_y = 1.0e-6, P_n,_o = 0.7e-6, p = 2.3e-24, common language effect size (CLES) D_Cdc42 = 0.25, median Tubulin: P_n,y = 1.1e-6, P_n,o = 1.0e-6, p = 0.8, CLES D_Tubulin = 0.49, n_y = 217, n_o = 372 b Left: Widefield IF images (example from z-stacks) of Cdc42, Tubulin and DAPI, scale bar 5 µm. Right: Absolute value of P_n, with median as central mark, edges indicating 25th and 75th percentiles, whiskers extending to most extreme data points not considered outliers, outliers individually plotted. Median Cdc42: P_n,_y = 0.89e-6, P_n,o = 0.68e-6, p = 1e-12, CLES D_Cdc42 = 0.40, median Tubulin: P_n,y = 0.60e-6, P_n,o = 0.55e-6, p = 0.009, CLES D_Tubulin = 0.46, n_y = 630, n_o = 1370. c Z-layers of individual cells (3D) and their P_n values of Cdc42 and Tubulin. d Percentage of cells over their P_n for young, old, and old HSCs treated with CASIN (n_young = 92, n_old = 98, n_oldCASIN = 96, median P_n,young = 0.81e-6, P_n,old = 0.64e-6, P_n,oldCASIN = 0.84e-6, p_young,old = 0.0032, p_{young,oldCASIN} = 8.0e-4, p_{young,oldCASIN} = 0.9, CLES D_young,old = 0.38, D_{old,old treated} = 0.36) (a–d) all two-sided Wilcoxon-Ranksum test. e Cumulative percentage of HSCs in dependence on P_n value distribution (data from (d)). f Scheme for determination of d_nucleus via |MR₋| (green arrow) and the maximal diameter d_max (orange arrow). g Pearson correlation coefficients R_Pearson of the P_n of DAPI, Cdc42 and Tubulin vs. d_nucleus, with p < 0.05 for all correlations shown (two-sided). p_{old,dnucleus,Cdc42} = 2.8e-129_, p_{old,dnucleus,Tubulin} = 0.0024, p_{old,dnucleus,DAPI} = 7.3e-157_; p_{young,dnucleus,Cdc42} = 1.0e-30, p_{young,dnucleus,Tubulin} = 0.001, p_{young,dnucleus,DAPI} = 2.0e-46. Source data are provided as a Source Data file.

Full size image

Cdc42 is a small RhoGTPase³⁸. Cdc42 is not only a polarity marker protein, but via its GTPase function, also a regulator of polarity^10,39. Upon aging, there is elevated activity of Cdc42 in HSCs, which results in the higher frequency of old HSCs being apolar for Cdc42 itself^2,40. Pharmacological attenuation of the elevated Cdc42 activity in old HSCs to the level of the activity in young HSCs by the drug CASIN (at 5 µM) is known to repolarize old HSCs (by visual determination)^2,17. We determined therefore the extent of repolarization of old HSCs by CASIN in 3D by CellDetail in comparison to young and old HSCs. Displaying the data as the percentage of HSCs relative to their P_n value for Cdc42 (Fig. 2d) identified that there are more old HSCs with a lower value for P_n and there is a higher percentage of young and old+CASIN HSCs with higher P_n values, implying shifts in polarity between young and aged HSCs along the whole range of values for polarity, while P_n distribution after re-polarization of old HSCs by CASIN aligns more to young HSCs along the range of P_n. Second, to be able to compare polarity identified by CellDetail to previously published binary polarity data on young, old and old HSC rejuvenated by CASIN, we calculated the cumulative percentage of cells along P_n values (Fig. 2e) of the data presented in Fig. 2d. We previously reported that on average 60–65% of young HSCs are polar for Cdc42 distribution (visual examinations,²). Using this frequency as the threshold for a binary scoring of polarity like in visual examinations in our analyses (black vertical line in Fig. 2e, which falls onto a P_n of about 0.6 × 10^-6, which is also the level of P_n that separated polar from non-polar HSCs in Fig. 1h) 49% of old HSCs are listed as polar, while now 68% of old HSCs treated with CASIN are listed as polarized. These frequencies of polar cells of rejuvenated HSCs are in accordance with published frequencies of rejuvenation, while the overall frequency of polar cells among old HSCs, while still significantly reduced compared to young HSCs, is somewhat higher than previously identified by visual examination². The data further show that CASIN treated aged HSCs follow closely the cumulative frequency curve of young HSCs along the range of P_n values, which again implies that all types/levels/states of polarity of aged HSCs respond with an increase in polarization upon CASIN treatment.

In HSCs, nuclei occupy a large volume of the cell (Fig. 2a, f). It therefore remains a possibility that polarity of proteins in HSCs is simply a consequence of the position of the cytoplasmic space driven by the position of the nucleus. To address this question, we determined the correlation between a parameter that identifies the position of the nucleus (d_nucleus) and the level of protein polarity in the cytoplasm (Fig. 2f, g). The position of the nucleus within the cell was calculated based on the negative charge-weighted center R₋ of the DAPI channel in the cytoplasm (Fig. 2f), as the position of the positive DAPI charge center within the nucleus is strongly affected by, due to their difference in DAPI intensity, the distribution of heterochromatin and euchromatin, and will thus deviate from the real barycenter of the nucleus. The parameter d_nucleus was calculated as percentage of the maximal diameter d_max (orange arrow): d_nucleus =|MR₋ | / d_max · 100, with |MR₋| (green arrow) being the distance between M (the middle of the cell) and R₋ in the DAPI channel (Fig. 2f). We expected a linear relationship of polarity with position of the nucleus and thus used Pearson correlation for calculations (R_Pearson). There were intermediate strong correlation coefficients of P_n of DAPI and d_nucleus (R_young = 0.53, R_old = 0.64), as the position of the nucleus and cytoplasm are indeed related to each other in HSCs (Fig. 2g). The low R_Pearson values for polarity of Tubulin and nucleus positioning (R_young = 0.13, R_old = 0.08) imply that the position of the nucleus within the cell does not simply determine in general the level of polarity of a polarity protein like Tubulin. The intermediate levels of correlation of polarity of Cdc42 and position of the nucleus (R_young = 0.44, R_old = 0.59) though could imply a role of the position of Cdc42 in positioning of the nucleus also in HSCs, as has been previously implied in other cell types^41,42,43.

CellDetail can be readily applied to other types of quantifications of cellular components. We previously investigated changes in the spatial distribution of chromosomes upon aging⁴⁴. Quantification of changes in the position of chromosomes remained challenging, and we finally used homolog distance as an indicator for changes in chromosome distribution (Supplementary Information A, Fig. S1a). Re-analysis of the original image data with CellDetail allows for a quantification of changes of the position of chromosome 11 homologs in the nucleus. CellDetail quantified the distinct distribution of chromosome 11 in the nucleus of young or aged HSCs, and confirmed that the attenuation of Cdc42 activity with CASIN restored a youthful distribution of chromosome 11 in the nucleus of chronologically aged murine HSCs (Supplementary Information A, Fig. S1b).

The network of the cytoskeletal Septin proteins is less organized in old HSCs

How changes in Cdc42 activity affect polarity within HSCs is not known. We recently demonstrated that the level of activity of Cdc42 determines the localization/polarity of Septin7 within HSCs³¹. Septins interact with Actin and Tubulin. They compartmentalize cells by forming hetero-oligomeric complexes and filaments which serve as scaffolds for protein recruitment or as diffusion barriers^30,33,45. There are 13 Septins in mammals, which are classified into four SEPT groups based on homology.

We hypothesized that the localization or distribution of Septins and their network structure in old HSCs is changed, not only for Septin7. A quantitative analysis of changes in the organization of a complex protein-protein network in adult stem cells has been so far elusive due to lack of proper image analysis tools. We determined here the distribution of Septin1, Septin2, Septin6, Septin7, Septin9 and Septin11 in individual HSCs with a newly developed multi-color staining panel (Supplementary Information D, Fig. 3a). The median of P_n for each of the Septins was reduced in old compared to young HSCs (Fig. 3b), except for Septin 1, with effect sizes in the small to medium range. Septins showed a more apolar distribution in aged HSCs, which implies additional changes in the overall structure of the network. We therefore determined the positions of Septin proteins relative to each other in young and aged HSCs. To this end (i) the distances between positive charge-weighted protein centers R₊ (Fig. 3c) as well as (ii) the angles between the positive charge-weighted centers R₊ (by determining the angle between the lines of the middle of the cell M to the individual positive protein charge centers) were calculated (Fig. 3e). The distance provides a measure of the similarity between distributions (Fig. 3d). A small distance can be due to two polar distributions close together or due to two apolar distributions, while a large distance is the result of two polar distributions opposite each other in the cell volume. Whether changes in these distances are associated with large changes of the overall location of proteins or with more loose inter-connections between proteins was determined via analysis of the angle value. Small angle differences with a single peak indicate more loose inter-connections, while large angle differences with multimodal distributions indicate larger units changes (Fig. 3f). The angle analysis is thus critical for further classifying inter-protein relationships. Aging affected the median distances between the positive charge centers of Septin7 and Septin6 (increase) and between the positive charge centers of Septin7 and Septin9 and between Septin7 and Septin11 (decrease). Thus, the distributions of Septin7 and Septin6 become less similar upon aging, while Septin7-Septin9 and Septin7-Septin11 distributions become more similar. Comparing the angles between positive charge centers of Septins showed an overall significant increase of the median angle for any combination of Septins upon aging (angles between Septin 6 and Septin 7 and between Septin 6 and Septin 11 are shown in Fig. 3e). These data imply that upon aging, there is indeed reduced inter-connectivity among Septins in HSCs.

a IF images (example from z-stacks) of a young or old HSC (scale bar 5 µm) of Septins 1,2,6,7,9,11. b Absolute value of P_n, with median as central mark, edges indicating 25th and 75th percentiles, whiskers extending to most extreme data points not considered outliers, outliers individually plotted. Common language effect size (CLES) D_Sept2 = 0.35, D_Sept6 = 0.36, D_Sept7 = 0.28, D_Sept9 = 0.30, D_Sept11 = 0.34; p_Sept1 = 0.27, p_Sept2 = 2.0e-12, p_Sept6 = 5.6e-11, p_Sept7 = 1.5e-24, p_Sept9 = 7.1e-22, p_Sept11 = 1.2e-13, n_y = 330 and n_o = 440. c Distance d |between positive charge centers (box plots, median as central mark, edges indicating 25th and 75th percentiles, whiskers extending to most extreme data points not considered outliers, outliers individually plotted), n_y = 330 and n_o = 440. CLES D_Sept6-Sept7 = 0.45, D_Sept7-Sept9 = 0.41, D_Sept7-Sept11 = 0.45. p_Sept6-Sept7 = 0.03, p_Sept7-Sept9 = 2.9e-5, p_Sept7-Sept11 = 0.03. d Graphical representations of examples of the distance d between charge centers. e Angle α between different positive charge centers of distinct Septins in young or old HSCs, median, n_y = 330 and n_o = 440. CLES D_Sept6-Sept7 = 0.38, D_Sept6-Sept11 = 0.40, p_Sept6-Sept7 = 6.3e-9, p_Sept6-Sept11 = 1.4e-6. f Graphical representations of examples of the angle α between charge centers. g Pearson correlation coefficient values (R_{Pearson,image}, output parameter of CellDetail) of the intensity channels for distinct combinations of Septins (box plots, median as central mark, edges indicating 25th and 75th percentiles, whiskers extending to most extreme data points not considered outliers, outliers individually plotted), p_Sept2-Sept6 = 2.2e-4, p_Sept2-Sept9 = 0.02, p_Sept2-Sept11 = 0.02, p_Sept6-Sept9 = 0.01, p_Sept7-Sept9 = 7.2e-6, p_Sept9-Sept11 = 0.006, n_y = 330 and n_o = 440. CLES for distributions with significant different medians are: D_2-6 = 0.42, D_2-9 = 0.45, D_2-11 = 0.45, D_6-9 = 0.45, D_7-9 = 0.41, D_9-11 = 0.44. b, c, e, g two-sided Wilcoxon Ranksum test. Source data are provided as a Source Data file.

Full size image

For another type of determination of the extent of changes in the Septin network in HSCs upon aging, we calculated Pearson correlation coefficients (R_Pearson) for intensity values for every pair of intensity channel pair combination of the Septins analyzed (implemented in CellDetail). R_Pearson describes in this case the amount of co-localization of two Septins with R = 1 being the perfect match, R = 0 being random and R = -1 being the perfect mismatch. Overall, distinct Septin proteins showed significant, but minor changes between young and old HSCs (Fig. 3g), with interactions that include Septin2 or Septin7 being frequently affected by aging. These findings imply that the overall composition of the Septin filaments with respect to individual Septins and their position within the network is indeed altered upon aging, while there is no major reorganization of the composition and order of the basic filament network units upon aging.

Cdc42 activity determines the organization of the Septin network in HSCs

We also determined the extent to which attenuation of Cdc42 activity in old HSCs by CASIN results, besides in repolarization of Cdc42 (Fig. 2d, e), also in repolarization of Septins and whether exposure to CASIN affects the Septin overall network structure (Fig. 4a). Old HSCs treated with CASIN presented with elevated median values of P_n for Cdc42 itself, Septin6, Septin7 and Septin9 (Fig. 4b) compared to old HSCs. More importantly, the elevated levels of Spearman correlation values R_Spearman (due to positive correlations expected, but not necessarily linear ones) for the polarity of Septin6 and Septin7 with respect to polarization of Cdc42 strongly imply that rejuvenated old HSCs (aka CASIN treated HSCs,^2,17) restore also in part their Septin network (Fig. 4c). Interestingly, the correlation of Septin9 polarity with Cdc42 polarity was not affected by the treatment though its polarity was increased. This might imply a special role for Septin9 in the Septin network. Whether Cdc42 attenuation directly affects the distribution of all Septins, or the distribution of a single Septin like Septin7 that affect the polarity of other Septins will require further investigations.

a IF images (examples) of one z-stack plane of old or old HSC treated with CASIN (scale bar 5 µm) of Cdc42 and Septins 6,7,9. b P_n of Cdc42 and Septins 6,7, 9 (box plots, with median as central mark, bottom and top edges of box indicating 25th and 75th percentiles and whiskers extending to most extreme data points not considered outliers, and outliers individually plotted). n_old = 57, n_old,CASIN = 60 HSCs the same old mice. Median, two-sided Wilcoxon-Ranksum test, p_Cdc42 = 0.01, p_Sept6 = 0.0036, p_Sept7 = 1.5e-4, p_Sept9 = 3.7e-4. Common language effect size (CLES): D_Cdc42 = 0.36, D_Septin6 = 0.34, D_Septin7 = 0.30, D_Septin9 = 0.31. c R_Spearman of P_n of Cdc42 and P_n of Septins of old and old+CASIN HSCs. n_old = 57, n_old,CASIN = 60. p_{old,Cdc42Sept6} = 0.22, p_{old,Cdc42Sept7} = 0.19, p_{old,Cdc42Sept9} = 0.01, p_{oldCASIN,Cdc42Sept6} = 0.048, p_{oldCASIN,Cdc42Sept7} = 0.03, p_{oldCASIN,Cdc42Sept9} = 0.03. Source data are provided as a Source Data file.

Full size image

Changes in the organization of the Septin network in aged human fibroblasts

CellDetail has been developed for highly isotropic cells like HSCs, but allows analysis of other cell types, including fibroblastoid cells or in general cell line cells. Whether there are changes in the Septin network in senescent cells is not well known. We investigated the effect of aging on the distribution of the Septin network in FF95 human-derived primary fibroblasts to test whether aging also affects this Septin network, and if so, whether there are any similarities to changes of the network in murine HSCs. FF95 cells reach replicative senescence at cumulative population doublings (CPD) above 42 and are considered aged. FF95 cells with CPD 4.3 and CPD 59.0 were stained for Septin6, Septin7, Septin9, Septin11 and Tubulin. Overall, Septin6, Septin7, Septin9 and Septin11 were more clustered around the nucleus while Tubulin was distributed over the total cell body (Fig. 5a). P_n was significantly decreased for Septin6 and for Septin11 in senescent FF95 cells, but interestingly not for Septins7 and 9 and Tubulin (Fig. 5b). For determining the relationships of co-localizations of distinct Septins within individual cells, we first generated a Pearson correlation matrix of colocalizations among Septins and Tubulin for control and senescent FF95 cells, and then calculated the Pearson correlations of the Pearson correlation coefficients, which allows to investigate how the colocalizations among Septins correlate with each other and thus provides a broader view on the Septin network in control and senescent cells (Fig. 5c). The generally high correlation values for both control FF95 and senescent FF95 imply indeed a tight Septin network structure in both types of cells, with some important changes though upon senescence.

a IF images (examples) of one z-stack plane of young and senescent human FF95 dermal fibroblasts for Septins 6,7,11 and Tubulin (scale bar 50 µm). b Absolute value of P_n, with median as central mark, edges indicating 25th and 75th percentiles, whiskers extending to most extreme data points not considered outliers, outliers individually plotted. Common language effect size (CLES): D_Sept6 = 0.39, D_Sept11 = 0.39, two-sided Wilcoxon-Ranksum test (p_Sept6 = 0.04, p_Sept7 = 0.96, p_Sept9 = 0.4, p_Sept11 = 0.05, p_Tubulin = 0.6), n_CPD4.3 = 50, n_CPD59.0 = 68. c Pearson correlation coefficients (R_Pearson) of paired-channel-based Pearson correlation coefficients (R_Pearson, output parameter of CellDetail) for distinct combinations of Septins and Tubulin of young and senescent human FF95 dermal fibroblasts (n_CPD4.3 = 50, n_CPD59.0 = 68) and right, their difference in correlation coefficients. Except for 7-Tub with 7-9 in CPD 59.0 FF95 cells, all correlations were significant (p < 0.05, two-sided; provided within Source Data file). Source data are provided as a Source Data file.

Full size image

There is a strong decrease in the correlation of Septin7-Septin9 to Septin7-Tubulin upon senescence (Fig. 5c). There is further a decrease in correlations of Septin7-Septin9 to Septin6/Septin9/Septin11-Tubulin, and to a smaller, but still prominent extent also for correlations of Tubulin linked to Septin6-Septin9 and Septin9-Septin11 interactions. A further decrease is visible in correlations of Septin7-Septin9 to Septin6-Septin11 or Septin7-Septin11. Thus, in non-senescent cells, proteins are more present in the same region of the cell, while there is a change in the spatial appearance of especially Septin7 and Septin9 upon senescence, as correlations that contain Septin9 show more randomness in combination with other Pearson correlation coefficients, again, implying a special role of Septin9 upon aging/senescence. Interestingly, Septin9 has been implied to play a role in the senescence-associated secretory phenotype (SASP) of senescent fibroblasts via its promotion of secretion of matrix metalloproteinase^32,46,47. Our results are consistent with the possibility that changes in the position of Septins might contribute to the SASP phenotype of senescent fibroblasts. Thus, and at least in part similar to observations in aged HSCs, the Septin network is more disorganized in senescent human fibroblasts, while it likely maintains its basic composition of type of Septins that form the network.

• SEO Powered Content & PR Distribution. Get Amplified Today.
• PlatoData.Network Vertical Generative Ai. Empower Yourself. Access Here.
• PlatoAiStream. Web3 Intelligence. Knowledge Amplified. Access Here.
• PlatoESG. Carbon, CleanTech, Energy, Environment, Solar, Waste Management. Access Here.
• PlatoHealth. Biotech and Clinical Trials Intelligence. Access Here.
• Source: https://www.nature.com/articles/s41467-024-54638-8