Article Figures & Data
Figures
- Figure 1. Proteomic analysis of WT and AROM+ testes.
(A) The upper panel shows HE stained tissue sections of WT and AROM+ testis. Scale bar—100 μm. Venn diagram in the lower panel shows that 3,227 proteins were identified in both WT and AROM+. In addition, 476 and 624 proteins were detected exclusively in WT and AROM+, respectively. (B) Volcano plot generated from relative abundance of the shared 3,227 proteins after statistical analysis by LIMMA moderated t test. The filled brown circles represent the significantly enriched proteins in both WT (negative fold change) and AROM+ (positive fold change). (C) Statistically significant molecular pathways generated by enrichment analysis using hypergeometric distribution model represented by the WT enriched proteins (Table S2). (D) Statistically significant molecular pathways using enrichment analysis (pathways having the lowest 25 P-values generated using hypergeometric distribution model are displayed here) represented by the AROM+ enriched proteins (Table S3) (data from three mice for each group WT and AROM+ was considered for the statistical analyses). The colour code is representative of the statistical significance and is displayed according to the adjusted P-values generated using hypergeometric distribution model (as indicated by the magenta-white gradient bar). The size of the nodes in the network (scale represented by numbered black solid circles) are representative of the number of genes involved in a particular pathway within the network.
- Figure 2. Pathways exclusive to WT and AROM+ testes.
(A) Significant molecular pathways associated with early phases of transcription, detected exclusively in WT. (B) Pathways associated with later stages of transcription that are detected only in the testes of healthy animals. (C) Mitochondrial pathways observed solely in WT. (D) Inflammatory pathways observed only in 11 mo old AROM+ mice, which are otherwise not observed in healthy conditions (n = 3 mice for each group WT and AROM+). The colour code is representative of the statistical significance and is displayed according to the adjusted P-values generated using hypergeometric distribution model (as indicated by the magenta-white gradient bar). The size of the nodes (scale represented by numbered black solid circles) is representative of the number of genes involved in a particular pathway within the network.
- Figure S1. Unbiased hierarchical clustering shows consistent segregation of WT and AROM+ testes into tubular and interstitial clusters.
The left and the right panels are tissues from two different animals of the same age, respectively. Scale bars—1.8 mm.
- Figure 3. Proteomic map of WT and AROM+ testes.
(A, B) Initial unbiased hierarchical clustering of IMS peptide data from generated from whole tissue sections of (A) WT and (B) AROM+ testis (dendrograms: Fig S2) at 25 μm spatial resolution. (C, D) Further clustering results for the (C) WT and (D) AROM+ tissues (Dendrograms: Fig S3). In case of WT, four different sets of tubules are represented by the red, brown, green and deep blue clusters, respectively. The orange cluster represents peritubular regions and the royal blue cluster characterizes the interstitial spaces. (E, F) A further level of clustering of (E) WT and (F) AROM+ tissues (dendrogram: Fig S4) shows an exclusive cluster (yellow-green) that is localized to the interstitial regions of AROM+ testis. Scale bar—2 mm.
- Figure S2. Dendrogram for clustering shown in Fig 3A and B in main text.
- Figure S3. Dendrogram for clustering shown in Fig 3C and D in main text.
- Figure S4. Dendrogram for clustering shown in Fig 3E and F in main text.
- Figure 4. Protein identification in situ with maximum likelihood.
(A) Imaging mass spectrometry (IMS) measurements lead to an overlap of isotope envelopes of different peptides in the same mass range (Panel: MALDI-IMS). This is indicated by different distribution patterns of the concerned m/z peaks. Therefore, to segregate the spectra, we have used unbiased hierarchical clustering (SCiLS Lab 2016b) of IMS data from the entire tissue (Panel: Hierarchical clustering) that led to efficient deisotoping of the IMS spectra and create a mono-isotopic list (Panel: De-isotoping). The grey rectangles represent the mono-isotopic mass list obtained from each cluster. (B) LC–MS/MS measurements were performed on tissue following the section used for IMS and peptides/proteins were identified following spectra analysis and database search (MaxQuant 1.6.0.16). The grey rectangles represent the identified peptide dataset. (C) Upon mass comparison, each m/z value (m(1), m(2)…m(n)) from the IMS mono-isotopic list was, in many cases, annotated to multiple peptides identified in LC–MS/MS measurements (P(1), P(2)…P(k)) because of moderate mass accuracy (0.1 Da) of the current IMS setup. To clear the aforesaid ambiguity, we developed a scoring system (MLP scoring) that enabled to identify the parent proteins with maximum likelihood. (D) Distribution of identified proteins (proteins having highest MLP scores) in the WT and AROM+ testis. Proteins involved in late stage spermatogenesis (that were among the significantly altered proteins in LC–MS measurements also) were identified in the seminiferous tubules of WT (Table S7), whereas proteins involved in ECM regulation were found in the interstitia of AROM+ testis (Table S8).
- Figure S5.
(A, B) Isotope envelopes of (A) m/z = 936.55 and (B) m/z = 1,196.8 for WT testis. The upper panel shows the accumulated mass spectrum from the whole tissue with the isotope peaks. The lower panel shows the ion image alone and merged with HE stained tissue. Scale bars—1 mm.
- Figure S6.
(A, B, C, D, E, F, G, H, I, J) Accuracy of MALDI-IMS measurements for WT (A, B, C, D, E) and AROM+ (F, G, H, I, J). For each figure the left part shows peak for key peptides with an interval width ± 0.1 Da (colored bar) along with the corresponding ion map (m/z image) of the peptide on the right. Upper panel: (A) m/z = 784.5 (identified as Cct6a) (B) m/z = 845.5 (identified as Smcp) (C) m/z = 1,050.6 (identified as Hspa2) (D) m/z = 1,389.8 (identified as Tdrd6) (E) m/z = 2,343.3 (identified as Spesp1) Lower panel: (F) m/z = 763.5 (identified as Ogn) (G) m/z = 881.5 (identified as Col6a2) (H) m/z = 929.5 (identified as Cola1) (I) m/z = 1848.1 (identified as Ckb) (J) m/z = 2045.16 (identified as Pfas). For an estimation of the variation of these centroid masses over the measurement area, refer to Table S12.
- Figure 5. Inter-tubular heterogeneity and associated processes.
(A) WT testis showing inter-tubular heterogeneity (red, brown, green, and deep blue) based on their protein composition. AROM+ testis (on the right side) does not exhibit this proteomic heterogeneity. (B) Gene ontology (GO) term analysis of first set of tubules (red, tubuli A) based on the proteins identified there by MLP scoring. (C) GO term analysis of the second set of tubules (brown, tubuli B) based on the proteins identified there by MLP scoring. (D) GO term analysis of the third set of tubules (blue, tubuli C) based on the proteins identified there by MLP scoring. Identification of proteins defining different regions (clusters) was performed by MLP scoring based comparison with bulk proteomics. The Fig in the inset of each panel shows part of volcano plot that marks respective positions of the identified proteins in the enrichment landscape. Scale bars—1 mm. GO term analysis was done based on Over-representation test that determines whether genes belonging to a specific GO term are present more than that would be expected (over-represented) in a subset of the experimental data. The yellow nodes in the network indicate the biological processes (named within the grey coloured ovals). The size of these nodes in the GO term maps (scale represented by the numbered black solid circles) are representative of the number of genes involved in each of them. The grey nodes along with their labels show the genes that are involved in the aforesaid pathways.
- Figure 6. Validation of MLP score based identification.
Distribution pattern of some key proteins in imaging mass spectrometry (IMS) measurements (as identified by MLP scoring) and their corresponding corroboration by immunohistochemical staining (IHC). (A) Interstitial distribution of Mimecan (Ogn) (upper panel: ion image of m/z = 763.5 as measured in IMS; lower panel: IHC, interstitial staining indicated by *). In both IMS and IHC, the protein is observed to be present only in AROM+. (B) Interstitial distribution of Collagen I (Cola1) (upper panel: ion image of m/z = 929.5 as measured in IMS; lower panel: IHC, interstitial staining indicated by *). In both IMS and IHC, the protein is observed to be highly up-regulated in AROM+. (C) Tubular distribution of T-complex protein 1 subunit zeta (Cct6a) (upper panel: ion image of m/z = 784.5 as measured in IMS; lower panel: IHC, tubular staining indicated by *). In both IMS and IHC, the protein is observed to be present only in WT.
- Figure S7. Negative controls of IHC stainings.
(A, B) WT and (B) AROM+ tissue sections without primary antibody. They were incubated either with rabbit IgG or non-immune serum instead of the primary antiserum. Scale bar—50 μm.
Supplementary Materials
In this Issue
Subjects
Related Articles
Cited By...
- No citing articles found.