Development and characterization of an inducible Tensin1 deficient transgenic murine model
Ksenija Bernau, Kelsey Holbert, Ivy S. McDermott, Babita Rahar, Kade W. Coulter, Madalyn A. Osterhaus, Thomas J. Harr, Carson Gehl, Angie Tebon Oler, Christopher Huffman, Beth Gray, Trevor S. Langel, Kenneth M. Chan, Amir Forati, Sandra Splinter BonDurant, Mark E. Berres
TL;DR
This study creates a new mouse model to study Tensin1 and finds changes in gene expression and immune cell behavior when Tensin1 is removed.
Contribution
A novel inducible TNS1-deficient transgenic mouse model was developed for lineage-specific and temporally controlled knockout.
Findings
TNS1 knockout had no effect on mouse health, breeding, or organ histopathology.
RNA sequencing revealed 171 differentially expressed genes linked to immune and osteoclast pathways.
Loss of TNS1 impaired migration and phagocytosis in THP-1 cells and increased lymphocyte detection in bronchoalveolar lavage.
Abstract
Tensin1 (TNS1) is a key component of focal and fibrillar adhesions, mediating fibrillogenesis, as well as the transduction of mechanical cues and adhesive signaling. To enable further TNS1 characterization, we have developed a novel transgenic mouse that allows for temporally controlled and lineage specific knockout of TNS1. We found no differentially observed effects of TNS1 knockout on mouse health, breeding capacity, or vital organ histopathology. In contrast, RNA sequencing analysis identified 171 differentially expressed genes with Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis demonstrating differential expression in osteoclast differentiation and a number of pathways involved in host immune response. Furthermore, STRING pathway analysis showed differential regulation of genes involved in B-cell and myeloid-related signaling pathways, a number of which were validated by…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Figure 1
Figure 2
Figure 3
Figure 4
Figure 5
Figure 6
Figure 7- —https://doi.org/10.13039/100000002National Institutes of Health
- —https://doi.org/10.13039/100000968American Heart Association
- —https://doi.org/10.13039/100003284Pulmonary Fibrosis Foundation
- —https://doi.org/10.13039/100000090Congressionally Directed Medical Research Programs
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsCell Adhesion Molecules Research · Connective tissue disorders research · Cellular Mechanics and Interactions
Introduction
Tensin 1 (TNS1) is a 220 kDa multidomain protein that is a component of focal and fibrillar adhesive structures of the cell. TNS1 links the actin cytoskeleton with the cytoplasmic tail of the extracellular matrix (ECM)-binding integrins and serves as a scaffold for adhesion-related signaling via several protein domains, including a phosphotyrosine-binding (PTB) domain, Src homology 2 (SH2) domain, and protein tyrosine phosphatase (PTP) domain^1^. TNS1 is a key component of the specialized fibrillar adhesions that bind fibronectin and, via actomyosin-mediated translocation, facilitates fibronectin fibrillogenesis^2,3^. In addition to playing a key role in mediating fibronectin fibrillogenesis, TNS1 binds to several adhesion-related kinases, enabling transduction of adhesive signaling^4^. Finally, given its localization to mature and enlarging focal adhesions, TNS1 is also involved in mechanotransduction in response to biophysical matrix cues, such as matrix viscoelasticity^5^. Several studies have found that Tns1 expression or single-nucleotide polymorphisms (SNPs) within the Tns1 gene locus are associated with a number of diseases, including cancers, obstructive lung disease, and lung function decline^6–9^. We have previously shown that TNS1 plays a key role in myofibroblast differentiation and upregulation of fibronectin assembly^10^. However, germline deletion of TNS1 results in degenerative cystic kidney disease associated with hyperproliferative Madin-Darby canine kidney (MDCK) cells, while no obvious pathologic alterations in other organ systems were detected^11,12^.
Due to the putative role of TNS1 in modifying adhesive signals and fibronectin fibrillogenesis, we sought to develop a transgenic mouse that enabled temporally controlled knockout of TNS1, thereby potentially avoiding the deleterious consequences of germline deletion on renal function. In this study, we describe the development of a novel transgenic mouse and characterize the gene expression changes that occur under global loss of TNS1 in adult mice. Through this work, we have unexpectedly identified differentially expressed genes involved in immune cell signaling by RNA sequencing analysis, which may indicate a role for TNS1 in immune cell function.
Results
Development of conditional and temporally controlled TNS1-null mouse
Previous studies utilizing a murine model with constitutive deletion of amino acids 110–180 of TNS1 in mice showed the presence of progressive cystic kidney disease and loss of focal adhesive structures in the proximal tubule, along with small litter sizes^11^. To enable temporal control of TNS1 deletion and avoid the health-related issues of germline knockout, we set out to generate a novel, conditional TNS1 knockout mouse model. To accomplish this, we utilized the clustered regularly interspaced short palindromic repeats (CRISPR)/CRISPR-associated protein 9 (Cas9) approach, enabling Cre-Lox recombination-mediated knockout of TNS1.
We created B6.Tns1^lox, lox^ mice by floxing exon 8 of Tns1 (ENSMUST00000169786.8), as it is the most 5’ exon present in all full-length Tns1 isoforms and its loss results in a frameshift (Fig. 1A, B, **Supplementary Fig. **S1), thereby disrupting the reading frame and eliminating protein translation. To enable whole-mouse conditional and temporally controlled TNS1 knockout, we bred the B6.Tns1^lox, lox^ mouse line to the tamoxifen-inducible ubiquitous Cre-reporter strain (B6.129-Gt(ROSA)26Sor^tm1(cre/ERT2)Tyj/J^, Jackson Laboratories)^13^. The resulting mouse colony was termed Rosa^Cre/+^;Tns1^lox/lox^. Mice that were either homozygous negative for the Rosa driven Cre alleles (B6.Tns1^lox/lox^) or did not carry floxed Tns1 (Rosa^Cre/+^;Tns1^+/+^) served as negative controls. In subsequent experiments, when Rosa^Cre/+^;Tns1^lox/lox^ mice are treated with tamoxifen to induce TNS1 recombination, they are referred to as Tns1-/-. Tamoxifen-treated B6.Tns1^lox/lox^ and tamoxifen treated Rosa^Cre/+^;Tns1^+/+^ mice served as controls and are referred to as Tns1+/+ in our experiments.
Fig. 1. Generation of Tns1^lox, lox^ mouse for conditional TNS1 knockout. (A) Schematic of Tns1 gene locus on Chromosome 1. The targeted exon is expressed in five commonly translated variants. (B) Schematic of targeted exon with approximate location of LoxP sites, surrounding exons, and polymerase chain reaction (PCR) primers. C-H. Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) or TNS1-expressing wild type controls (Tns1+/+) from either B6.Tns1^lox, lox^ or Rosa^Cre/+^;Tns1^+/+^ were treated with daily tamoxifen. (C) Gel electrophoresis of tail DNA PCR product confirms excised exon. Real time quantitative PCR of Tns1 (D), Tns2 (E), and Tns3 (F) mRNA expression in lung tissue. Western blot (G) and associated densitometry (H) in lung tissue (n = 3 biological repeats). Unpaired Student’s t-test was utilized for all statistical assessments (*p < 0.05). Data is presented as mean ± SD.
As seen in Fig. 1C and Supplementary Fig. 2, the single 1516 bp PCR band from the genomic DNA indicates the intact floxed 8th exon in the tail samples of tamoxifen-treated B6.Tns1^lox/lox^ (Tns1+/+). In contrast, DNA samples from the tails of tamoxifen-treated Rosa^Cre/+^;Tns1^lox/lox^ mouse line (Tns1-/-) showed the 648 bp PCR band indicating successful recombination and loss of exon eight in the newly generated transgenic mouse colony. We found that this recombination continued to be evident up to 12 months post-tamoxifen treatment (**Supplementary Fig. **S1). Whole lung mRNA analysis revealed a 95% loss of Tns1 mRNA expression in these mice (Fig. 1D). Importantly, we did not find substantial changes in mRNA expression of Tns2 and Tns3 (Fig. 1E, F), demonstrating specificity of Tns1 targeting. Furthermore, we detected a dramatic reduction in TNS1 protein expression in the lungs of Tns1-/- mice compared to Tns1+/+ controls (Fig. 1G, H and **Supplementary Fig. **S2). Taken together, we show that our novel transgenic Tns1-/- mouse line results in recombination of loxP sites under control of tamoxifen-inducible Cre-recombinase, allowing for successful disruption of transcription and subsequent protein expression of TNS1.
RosaCre/+;Tns1lox/lox mice have normal litter sizes and appear healthy
Constitutive knockout of TNS1 has previously resulted in small litter sizes^11^. Since we seek to knockout TNS1 in adult mice, we analyzed pup count and health in TNS1-expressing Rosa^Cre/+^;Tns1^lox/lox^ mice prior to tamoxifen treatment. We demonstrate no difference in the number of litters per mating pair nor the number of pups per litter between Rosa^Cre/+^;Tns1^lox/lox^ and control mouse colonies, C57Bl/6J or B6.Tns1^lox/lox^ (Fig. 2A, B), suggesting that the genetically modified lines did not have abnormal fertility, fecundity, or pup survival.
Fig. 2. Mice carrying the Tns1 floxed alleles exhibit normal reproductive fitness and subsequent conditional TNS1 knockout in adult mice does not impact body weight. Number of litters per mating pair (A) and number of pups per litter (B) were analyzed in n = 5 breeding pairs in indicated colonies. The comparison includes three groups of TNS1-expressing mice: in-house bred WT mice (C57BL6/J), mice carrying the Tns1 floxed alleles on C57BL6/J background without a Cre-driver (B6.Tns1^lox, lox^) and with a Cre-driver (Rosa^Cre/+^; Tns1^lox, lox^). One-way ANOVA was utilized for statistical assessments. (C) Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) and TNS1 expressing controls (Tns1+/+) weights were collected for 14 days thereafter (n = 14 for Tns1+/+ and 7 for Tns1 -/-). Two-way ANOVA was used for statistical assessments. p < 0.05 was considered statistically significant and was not met in any of the comparisons. Data is presented as mean ± SD.
Next, we sought to investigate the broad phenotypic consequences resulting from the global knockout of TNS1, given its importance in focal and fibrillar adhesion formation and the previously identified effect on kidney morphology and function^11^. To induce global TNS1 knockout, we treated adult mice with intraperitoneal (IP) tamoxifen injections for seven consecutive days. We monitored mouse body weights daily starting at the point of tamoxifen administration until the endpoint three weeks later. We detected up to 10% weight-loss in both Tns1+/+ and the Tns1-/- groups (Fig. 2C), likely resulting from tamoxifen-administration and consistent with prior studies^14^. However, we found no change between the overall weights of the tamoxifen-treated Rosa^Cre/+^;Tns1^lox/lox^ mice and controls, indicating that disruption of TNS1 expression did not have an impact on overall mouse weights (Fig. 2C).
We then analyzed potential pathologic changes resulting from disruption of TNS1 expression. For this, we treated 12–14-week-old Rosa^Cre/+^;Tns1^lox/lox^ mice with tamoxifen for seven days to induce knockout of TNS1 (Tns1-/-), while tamoxifen-treated B6.Tns1^lox, lox^ (Tns1+/+ Tamoxifen) served as controls. As an additional control for tamoxifen-induced effects, we included tissue samples from one mouse not treated with tamoxifen (Tns1+/+ No treatment). On day 21, 14 days after termination of tamoxifen administration (or no treatment), mice were euthanized and vital mouse organs, including lungs, hearts, middle lobe of livers, spleens, and kidneys were fixed and submitted to a board-certified veterinary pathologist for Hematoxylin and Eosin (H&E) staining, and subsequent histopathologic analysis. Heart, spleen, and liver tissues remained within normal limits in all the mice (Supplementary Fig. S3). Unlike the kidneys from untreated animals (Fig. 3A), kidneys from tamoxifen-treated animals, regardless of the genotype, demonstrated variable minimal to moderate peritonitis and retroperitonitis (Fig. 3B, C). In addition, we detected the presence of plasma cells, lymphocytes, and minimal neutrophils, indicative of minimal subacute, multifocal inflammation in the retroperitoneum adjacent to the kidneys and the renal capsules, which may be a result of tamoxifen injections. In contrast to the untreated lungs (Fig. 3D), the lungs of tamoxifen-treated animals showed a few multifocal areas with minimal to mild interstitial infiltration of lymphocytes and plasma cells, histiocytosis, as well as perivascular cell infiltrates (Fig. 3E, F). In one mouse lacking TNS1 (out of n = 3 analyzed), we detected multiple variable sized areas compatible with microvacuolated and macrovacuolated macrophages, engulfing lipid-like material, mixed with mild numbers of plasma cells and lymphocytes (Fig. 3F). Importantly, this finding was not evident in any other mouse and was consistent with drug/vehicle-lipid artifact.
Fig. 3. Conditional TNS1 knockout does not result in vital organ histopathologic abnormalities 14 days after completion of tamoxifen treatment. Kidney tissue from non-treated (A), tamoxifen-treated B6.Tns1^lox/lox^ (Tns1+/+) controls (B) and tamoxifen-treated Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) mice (C) are depicted. Similarly, lung tissue from tamoxifen-treated B6.Tns1^lox/lox^ (Tns1+/+) controls (E) and tamoxifen-treated Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) mice (F) depicted and demonstrate minimal to mild interstitial infiltration of lymphocytes and plasma cells, alveolar histiocytosis, as well as multifocal perivascular leukocytic infiltrates (*), compared to non-treated lungs (D) which did not demonstrate these findings. Blue arrow marks an area of histiocytic and lymphoplasmacytic pleuritis with intralesional lipid material (F). (A–C) Scale bars = 2 mm for large scale images, 200 μm for the first set of insets (top), and 50 μm for the second set of insets (bottom). (D–F) Scale bars = 1 mm for large scale images (left column), 200 μm for the first set of insets (left column, top image in D; middle column in E; middle column, top image in F), and 50 μm for the second set of insets (left column, bottom image in D; far right column in E; middle column bottom image and far right column in F).
In contrast to the lung pathologic findings 14 days after TNS1 knockout, mice with a conditional knockout of TNS1 at three months revealed the presence of mild to moderate alveolar histiocytosis with mild to moderate numbers of hemosiderin-laden macrophages in both Tns1+/+ and Tns1-/- mice (Table 1, Supplementary Fig. S4). Six months after knockout, two out of four Tns1-/- mice had evidence of either alveolar histiocytosis or perivascular lymphocytosis. These findings were resolved by 12 months (Table 1, Supplementary Fig. S4). We also found evidence of lipid pneumonia in one Tns1-/- mouse at the 6-month time point, which was likely a consequence of the tamoxifen administration (Supplementary Fig. S5). Kidney analysis revealed the presence of moderate hydronephrosis with mild and rare multifocal renal tubular cystic dilations in one Tns1-/- mouse three months post-tamoxifen treatment (Supplementary Fig. S5). However, no other mice (out of n = 19 total and 13 Tns1-/-) at three, six or 12 months had evidence of cystic kidney disease (Supplementary Fig. S6). Mice had evidence of peritoneal inflammation and fibrosis, which was consistent between Tns1+/+ and Tns1-/- animals. The fibrosis became more organized over time with marked reduction in inflammation by the 12 month time-point, indicating that this finding was likely related to the previous tamoxifen administration (Supplementary Fig. S7). At all three long-term timepoints, we found signs of histiocytic and lymphoplasmacytic peritonitis and steatitis in the peritoneum of most mice, regardless of the genotype (Supplementary Fig. S5). Furthermore, we detected endometrial thickening, including hyperplasia of the endometrial epithelium and stroma, endometrial cystic dilations, adenomyosis, as well as paraovarian cysts in both Tns1+/+ and Tns1-/- female mice throughout the course of our sample collection (Table 1, Supplementary Fig. S5). Overall, considering that tamoxifen-treated animals, regardless of the genotype, shared consistent changes, our data suggests that tamoxifen treatment may have an independent effect on vital mouse organ findings and is unrelated to the conditional TNS1 knockout.
Table 1. Pathological findings from Tns1+/+ and Tns1−/− mice treated with tamoxifen three, six and twelve months prior to analysis.OrganPathologic finding3 months6 months12 monthsTns1+/+Tns1−/−Tns1+/+Tns1−/−Tns1+/+Tns1−/−LungAlveolar histiocytosis1/24/40/32/40/10/4Hemosiderosis1/23/40/31/40/10/4Lipid pneumonia0/20/40/31/40/10/4KidneyCysts0/10/50/30/40/10/4Hydronephrosis0/11/50/30/40/10/4Peritoneum-Retroperitoneum (serosa)Mild multifocal histiocytic and lymphoplasmacytic peritonitis-steatitis0/12/43/33/41/14/4Incipient fibrosis1/14/41/32/41/11/4EndometriumEndometrial hyperplasian/a0/3n/a1/2n/a2/3Uterine tube or endometrial cystsn/a2/3n/a1/2n/a1/3
We additionally reviewed the effect of TNS1 knockout on blood counts and plasma indices of renal and hepatic health. In this study, mice underwent a seven-day treatment with tamoxifen. On day 21, we collected blood samples from Tns1-/- and Tns1+/+ mice. We found no differences in white blood cell count, including monocytes and lymphocytes, platelet count or hemoglobin concentration observed (Fig. 4A-E). In addition, we obtained metabolic indices that showed no differences in blood urea nitrogen (BUN), creatinine (Cr) concentrations, total protein levels, albumin concentration or the levels of alanine aminotransferase (ALT), aspartate aminotransferase (AST) and alkaline phosphatase (Fig. 4F-L). We also collected blood from Tns1+/+ and Tns1-/- mice 3, 6 and 12 months post-tamoxifen. At the 3 and 6 month time-points, we found no differences between the groups. At the 12 month time-point, n = 1 Tns1+/+ prohibited statistical analyses but we found that most of the mice remained within normal limits for each parameter, corroborating our previous findings (Supplementary Fig. S8). Notably, one Tns1-/- mouse had elevated levels of ALT and AST, as well as moderate, focal extensive histiocytic and lymphoplasmacytic peritonitis with hepatic capsule to parietal diaphragm adhesion, as detected on histopathological examination. Overall, we found that conditional knockout of TNS1 globally results in normal mating potential and unchanged pathological and blood parameters.
Fig. 4. Conditional knockout of TNS1 does not result in significant hematologic or metabolic abnormalities. Select complete blood count measurements (A–E) and metabolic markers (F–L) were analyzed between tamoxifen treated Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) and TNS1 expressing controls (Tns1+/+). There were no noted hematologic or metabolic abnormalities in TNS1 knockout mice using unpaired Student’s t-test with Benjamini-Hochberg correction for false discovery rate for statistical analysis. p < 0.05 was considered statistically significant and was not met in any of the comparisons. Data is presented as mean ± SD.
With our initial experiments, we conclude that there are no considerable phenotypic changes in the vital organs of adult mice within three weeks following induced global loss of TNS1 expression. Subsequently and in contrast to previous studies that demonstrate that mice with a germline knockout of TNS1 exhibit progressive cystic kidney disease, we did not see evidence of these alterations intrinsic to the kidney histology out to 12 months after knockout in this TNS1 knockout model^11^.
RNA sequencing reveals alteration in the lung B-cell and myeloid populations upon TNS1 knockout in adult mice
Given our previous report of a role for TNS1 in lung fibroblast gene expression^10^ and our observed pathological findings, we sought to determine the transcriptional impact resulting from Tns1 knockout in the lung. For this, mice were treated with tamoxifen as above, RNA was isolated from five Tns1+/+ and three Tns1-/- mouse lungs and first subjected to Quality Control (QC) metrics. Between Tns1-/- and Tns1+/+, we identified 171 differentially expressed genes using the requirements of more than 1.5-fold difference with a False Discovery Rate (FDR) of less than 0.05 after correction for multiple comparisons. We subsequently analyzed these data through Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis^15–17^, a heatmap of the top 50 differentially expressed genes, a volcano plot, and the STRING database protein-protein interaction (PPI) network.
KEGG pathway analysis showed enrichment in genes associated with host immune response, including differentiation of immune system precursors (Hematopoietic cell lineage,* Osteoclast differentiation*), several infections (Leishmaniasis,* Tuberculosis*,* Chagas Disease*,* Malaria*) as well as immune signaling pathways (C-type lectin receptor signaling pathway,* Cytokine-cytokine receptor signaling interaction)*^15–17^. The top eight pathways, as sorted by p value, are shown in **Supplementary Table **S1 and Fig. 5A. The volcano plot analysis also depicts several immune-related highly upregulated genes such as Spib and Ms4a1 and downregulated genes such as Nfkbia, Trem1, Bhlhe40 and Clec4e in Tns1-/- compared to Tns1+/+ (Fig. 5B). These findings are also evident in the heatmap (Fig. 5C), which shows the Z-scores for normalized expression in contrast of the 50 most differentially expressed genes with FDR < 0.05. Some of the upregulated genes in Tns1-/- compared to Tns1+/+ mice include Cd19, Cd22, Spib, and Bank1 and downregulated genes include Trem1, Treml4, Plaur, Cd300e, Acod1, and Clec4e.
Fig. 5RNA sequencing analysis reveals differential expression of genes involved in immune system regulation and cytoskeletal organization. (A) KEGG Pathway Enrichment Analysis of the top eight differentially expressed pathways shows a number of immune system regulated pathways in tamoxifen-treated Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) mice compared to TNS1 expressing controls (Tns1+/+). Bubble size indicates the number of genes identified in each pathway. (B) Volcano plot depicting differentially expressed genes in lung lysates from Tns1-/- and Tns1+/+ control mice. The colored points depict differentially expressed genes (adjusted p < 0.05) with a -log(FDR) >1.303. The negative (blue) and positive (red) log_10_(FC) indicate downregulated genes such as Nfkbia, Trem1, Bhlhe40 and Clec4e and upregulated genes such as Spib and Ms4a1, respectively, in Tns1-/- samples. C. Heatmap showing Z-scores for normalized expression of the top 50 differentially expressed genes with an adjusted p value <0.05. (n = 3 for Tns1 -/- and 5 for Tns1 +/+) and illustrates upregulation of Cd19, Cd22, Spib, and Bank1 and downregulation of Trem1, Treml4, Plaur, Cd300e, Acod1, and Clec4e in Tns1-/- mice compared to Tns1+/+.
For the final transcriptome pathway analysis, we subjected the 171 differentially expressed (68 upregulated and 103 downregulated) genes to STRING: functional protein association networks analysis. Seventeen genes were not recognized by the STRING database and ultimately, 154 gene names and their adjusted log(Fold Change) values were included in the analysis, which led to a network with a PPI enrichment p-value of < 1.0e-16, an average node degree of 7.7, and an average local clustering coefficient of 0.456. We performed supervised Markov Clustering (MCL) with an inflation parameter of 2.2 and found two predominant clusters. Across all clusters, the top Biological Processes by Gene Ontology as sorted by FDR was Regulation of immune system process (GO: 0002682), with an FDR of 1.02 × 10^− 27^ and included 63 of the 154 genes in the network, underscoring the immune system signature that resulted from the knockout of TNS1.
While our results demonstrated that TNS1 knockout had an impact on cell membrane-related signaling (Supplementary Table S2), we were intrigued to find that the top two cluster sets were marked by changes in the immune compartment (Figs. 6 and 7). These two cluster sets are defined by medium confidence where a minimum required interaction score of 0.400 was required. Disconnected nodes were removed from analysis. The size and color of the ring correspond to the log(FC) and the weight and color of the lines connecting the node are a STRING feature, where a darker and thicker line indicates a stronger degree of confidence in the prediction of the interaction.
Fig. 6. Loss of TNS1 results in downregulation and phenotypic alteration of myeloid cells. (A) STRING: functional protein association networks analysis with supervised MCL clustering (inflation parameter of 2.2) of differentially regulated genes (adjusted p < 0.05) revealed differential expression in the myeloid cell cluster (n = 65 downregulated and 6 upregulated genes) in tamoxifen-treated Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) mice compared to TNS1 expressing controls (Tns1+/+). Darker and wider gene outlines indicate a greater level of differential gene expression, while weight and color of the lines connecting the nodes are STRING features that indicate the strength of the prediction of the interaction. (B, C) Mouse lungs were collected for reverse-transcription quantitative polymerase chain reaction analysis of myeloid markers, Trem1 (B) and Mmp12 (C). (D–G) THP1 monocyte cells were transiently transfected with TNS1 siRNA or scrambled control before assessing changes in cell phenotype, including assessment of cell area via phalloidin staining of adherent cells (D), forward scatter-height via flow cytometry for estimation of non-attached cell size (E), migrated distance on fibronectin-coated plates via time lapse image analysis (F), and phagocytosis capacity of adherent THP-1 cells (G). Data is presented as mean ± SD and was analyzed via unpaired Student’s t-test (*p < 0.05).
Fig. 7TNS1 knockout reveals upregulation of B cell genes and a trend toward increased lymphocyte presence in the bronchoalveolar lavage fluid. (A) STRING: functional protein association networks analysis with supervised MCL clustering (inflation parameter of 2.2) of differentially regulated genes (adjusted p < 0.05) revealed differential expression in the myeloid cell cluster (n = 0 downregulated and 16 upregulated genes) in tamoxifen-treated Rosa^Cre/+^;Tns1^lox/lox^ (Tns1-/-) mice compared to TNS1 expressing controls (Tns1+/+). Darker and wider gene outlines indicate a greater level of differential gene expression, while weight and color of the lines connecting the nodes are STRING features that indicate the strength of the prediction of the interaction. (B, C) Mouse lungs were collected for reverse-transcription quantitative polymerase chain reaction analysis of B cell markers, Spib (B) and Cd19 (C). (D–G) Mouse bronchoalveolar lavage was performed and BAL fluid was collected for analysis of macrophages (D), neutrophils (E), lymphocytes (F), and eosinophils (G). Unpaired Student’s t-test was utilized for statistical analyses directly or on log-transformed data that did not pass normality testing (*p < 0.05). Data is presented as mean ± SD.
Cluster A (Fig. 6A) is a cluster of 71 predominantly downregulated genes, many of which are involved in macrophage signaling. The top five biological Gene Ontology molecular functions sorted by FDR include Signaling receptor binding (GO:0005102), Signaling receptor regulator activity (GO:0030545), Immunoglobin receptor binding (GO:0034987), Signaling receptor inhibitor activity (GO:0034987), and Protein binding (GO:0005515). The top five TISSUES database Tissue expression pathways sorted by FDR include Macrophage, Bone marrow-derived macrophage, Immune system, Spleen, and Microglia. Notably, TNS1 expression in myeloid/macrophage lineage is corroborated by publicly available databases, such as the Kaminski/Rosas data set from the IPF cell atlas (Supplementary Fig. S9)^18^.
Several of the myeloid markers in Cluster A such as Trem1, Treml4, Plaur, Cd300e, Acod1, and Clec4e are downregulated in Tns1-/- samples. Using combined mRNA from a validation cohort treated identically, we validated changes in expression of markers Trem1 (Fig. 6B) and Mmp12 (Fig. 6C) by RT qPCR. To understand the potential functional impact of these gene and mRNA expression changes, we utilized an in vitro model of TNS1 knockdown in the THP-1 immortalized monocyte/macrophage cell line (Supplemental Fig. S10). THP-1 cells transfected with TNS1-targeting siRNA are smaller than scramble controls as measured by phalloidin staining of adherent cells (Fig. 6D) and forward scatter-height (FSC-H) flow cytometry of non-attached THP-1 cells (Fig. 6E). Adherent THP-1 cells with TNS1 knockdown by siRNA also migrated a significantly shorter distance on fibronectin-coated plates over the 16 h compared to the scrambled siRNA controls (Fig. 6F). Lastly, we observed a decrease in phagocytosis in TNS1-knockdown adherent THP-1 cells compared to scrambled control in a flow cytometric phagocytosis assay, demonstrating impaired macrophage function (Fig. 6G, Supplemental Fig. S11). Functional enrichment analysis of RNA sequencing similarly highlighted several predicted proteins and pathways involved in phagocytosis and differential gene expression revealed several differentially expressed genes that are critical in phagocytosis, all of which were downregulated in TNS1 knockout mice compared to controls (Supplemental Fig. S12).
Beyond the changes in myeloid signaling, we also detected Cluster B, composed of 16 upregulated genes associated with a B-cell signature (Fig. 7A). The top five TISSUES Database Tissue expression pathways are B-lymphocyte (BTO:0000776), Spleen (BTO:0001281), DAUDI cell (BTO:0001561), Urinary bladder carcinoma cell line (BTO:0000792), and plasma cell (BTO:000392). Gene Ontology molecular functions were not reported for this cluster. In Cluster B, several B-cell signaling markers such as Cd19,* Cd22*,* Spib*, and Bank1 were upregulated. Using combined mRNA from a second cohort of mice, we validated the upregulation of Spib (Fig. 7B) and Cd19 (Fig. 7C) by RT qPCR.
Given these findings that suggest TNS1’s impact on the immune compartment, we sought to examine the cell differentials in the Tns1-/- bronchoalveolar lavage. We identified a trend toward increased numbers of lymphocytes in Tns1-/- mouse bronchoalveolar lavage (BAL) three days after knockout compared to Tns1+/+ controls (Fig. 7D-G), indicating that there may be a propensity for increased lymphocytic inflammation in Tns1-/- mouse lungs.
Discussion
TNS1 has an established role as an integrin-binding protein that associates with focal and fibrillar adhesions, mediating fibronectin fibrillogenesis, along with regulation of focal adhesion related signaling^1,3,10^. However, despite its multifunctional structure and dispersed tissue/cell expression, the role of TNS1 in non-stromal cells is unknown. Here, we have developed and characterized a novel mouse model that enables temporal regulation and conditional knockout of TNS1. Through RNA sequencing analysis of these mouse lungs, we found differential TNS1-driven regulation of lung immune signaling.
This murine model of conditional TNS1 knockout allows improved characterization of TNS1 effects compared to previously available models^11^. In contrast to the previous method of TNS1 knockout, these mice do not demonstrate any overall pathological changes as a result of conditional TNS1 knockout up to twelve months, including no evidence of cystic kidney lesions. We noted one case of hydronephrosis in the three-month TNS1 knockout cohort, something that has been previously reported in mice, especially on C57Bl/6 background^19,20^ and was not evident at the later time points analyzed. Considering that kidney pathology differed in our conditional TNS1 knockout compared to the constitutive method, these findings suggest that TNS1 may play an important role during development of the kidney, although we cannot exclude the possibility that we may see emergent renal pathology in mice of advanced age. Furthermore, we observed evidence of peritoneal inflammation and fibrosis, which appears to result from intraperitoneal tamoxifen administration, subsiding over time and up to the 12-month time point that we analyzed. In the Tns1+/+ and Tns1-/- female mice, we noted the presence of endometrial hyperplasia and cystic structures surrounding the ovaries, which has also been previously reported in association with tamoxifen administration in both mice and humans^21–23^. Importantly, these pathological abnormalities did not appear to associate with the TNS1 knockout cohorts.
Interestingly, we found that TNS1 knockout results in robust changes to the immune compartment of the lung. Specifically, we have identified a strongly downregulated myeloid RNA signature in the lungs of TNS1 knockout mice that occurs in the absence of lung injury. Functionally, loss of TNS1 corresponded to smaller cell area, decreased migration distance, and reduced phagocytosis efficiency in THP-1 cells compared to scrambled control cells, indicating significant functional impairment in the absence of TNS1. Additional analysis of RNA sequencing suggests that many of the differentially regulated genes are involved in membrane signaling and we hypothesize that TNS1 may play a role in signaling by immune receptors at the membrane given its known function in interacting with transmembrane integrins^24–26^. Our data also suggests that these transmembrane interactions are dependent on TNS1, as we observe impaired migration in adherent THP-1 cells with TNS1 knockdown compared to controls. Together, this suggests a novel role for TNS1, which may be critical for homeostatic signaling in myeloid cells such as dendritic cells, monocytes, and macrophages.
However, we also noted induction of a B-cell signature in the TNS1 knockout mouse, in contrast to downregulation of a myeloid signature. This finding further substantiates the hypothesis that TNS1 may have a role in immune system regulation. Using the Human Protein Atlas, the IPF cell atlas and RNA sequencing of human immune cells^18,27–30^, it is observed that TNS1 expression is abundant in the monocyte/macrophage and B-cell populations. Supporting an effect of loss of TNS1 on the immune compartment, we find a trend toward elevated numbers of lymphocytes in BAL fluid of Tns1-/- mice. Taken together, these findings suggest that TNS1 may modify the immune response to injury beyond what is expected in its regulation of fibronectin assembly and mesenchymal function and fate^10,31^. Further studies will be needed to determine the mechanistic basis for these alterations in immune cell function.
Given our findings, it will be of interest to determine how TNS1 modifies the inflammatory response to lung injury and fibrosis using this transgenic approach, the development of immune cell phenotypes, homing of immune cells to the lung, as well as signaling and immune responses in individual immune cell types. While beyond the scope of this report, these studies are ongoing.
In conclusion, we have developed a mouse that enables conditional knockout of TNS1 and avoids the deleterious complications of germline deletion. We have unexpectedly found that global TNS1 knockout results in gene expression changes in myeloid and lymphoid immune compartments, suggesting a role for TNS1 in these cell lineages. Ongoing research will determine the mechanisms by which TNS1 modifies immune signaling and lung signatures.
Methods
Generation of Tns1lox, lox and Tns1−/− mice
Tns1^lox, lox^ mice were generated de novo in collaboration with the University of Wisconsin-Madison Genome Editing and Animal Model (GEAM) core, flanking Tns1 exon 8 (ENSMUST00000169786.8) with loxP cassettes. In order to minimally impact critical functional elements of the splice acceptor, the 5’ loxP cassette was targeted for insertion over 500 bp away from the exon splice acceptor. The 3’ loxP site was targeted for insertion over 100 bp away from the exon 8 splice donor. We first attempted to knockin both loxP cassettes in one round of microinjection, using a single sgRNA and single-stranded oligo (ssODN) donor for each loxP cassette (Supplementary Table S3). A mixture of both in vitro transcribed sgRNAs (50 ng/µl total), both ssODNs (50 ng/µl total), and Cas9 protein (40 ng/µl) was microinjected into C57Bl6/J mouse embryos before the embryos were implanted into recipient mothers. C57Bl6/J mice were purchased from Jackson Labs (Bar Harbor, ME). To identify loxP knockin integration, tail snips from resulting pups were sampled, the region amplified via polymerase chain reaction, subcloned, and Sanger sequenced (Fig. 1A-C, Supplementary Fig. S1, S2). A mouse line carrying the 3’ loxP cassette, but not the 5’ loxP cassette, was successfully recovered. This line was used as embryo donor to specifically introduce the 5’ loxP cassette (Supplementary Table 1) as describe above. An allele carrying both 5’ and 3’ loxP cassettes was identified, but this allele also carried a duplication of part of the 5’ single-stranded oligodeoxynucleotide (ssODN) homology arm (possibly due to mixed homology-directed repair (HDR) – non-homologous end joining (NHEJ) event whereby single-stranded invasion of the ssODN followed by synthesis beyond the knockin construct and the generation of a double-stranded end, which was then ligated to the other end of the double stranded break by NHEJ^32^. Thus, the net result is an 80 bp insertion, rather than the expected 34 bp loxP insertion.
The resultant Tns1^lox, lox^ mice were backcrossed for over ten generations onto C57Bl6/J background to create the B6.Tns1^lox/lox^ mice. Subsequently, B6.Tns1^lox/lox^ mice were crossed with the tamoxifen-inducible Cre reporter strain (B6.129-Gt(ROSA)26Sor^tm1(cre/ERT2)Tyj/J^); strain 008463 from Jackson Labs, now referred to as “Rosa^Cre/+^;Tns1^lox/lox^”. Mating potential was assessed by comparing the number of litters/mating pair and number of pups/litter in n = 5 mating pairs from the following colonies: C57Bl6/J, B6.Tns1^lox/lox^ and Rosa^Cre/+^;Tns1^lox/lox^. One-Way ANOVA with Šídák’s post-hoc test was used for statistical analyses (p < 0.05 was considered statistically significant). In other experiments, when Rosa^Cre/+^;Tns1^lox/lox^ mice are treated with tamoxifen to induce Tns1 recombination, they are referred to as Tns1-/- and when the negative controls, either B6.Tns1^lox, lox^ or Rosa^Cre/+^:Tns1^+/+^ mice, are given tamoxifen, they are referred to as Tns1+/+.
Animals
Mouse protocols were approved by the Institutional Animal Care and Use Committee (IACUC protocol number M005823) at the University of Wisconsin-Madison and adhered to the guidelines set by the National Institutes of Health and the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC), as well as the Animal Research: Reporting of In Vivo Experiments (ARRIVE) guidelines. The mice, generated de novo as described above, were housed under controlled temperature (~ 74 °F), humidity (~ 30%), and illumination (12-hour light/12-hour dark cycle), and laboratory chow and water were provided ad libitum. Animals received daily care and health checks and weekly cage changes. For mating potential comparisons, in-house bred C57Bl6/J mice were used as controls, along with mice from the aforementioned B6.Tns1^lox/lox^ colony. For experiments, male and female mice aged 14.4 ± 5.2 (mean ± SD) weeks from Rosa^Cre/+^;Tns1^lox/lox^ colony or negative controls expressing TNS1 as described above were housed in groups (five or less) whenever possible. Mice were treated with daily doses of tamoxifen (100 mg/kg/day, IP; carcinogen, reproductive hazard/teratogen) for seven days, followed by confirmation of exon eight deletion 14 days (Fig. 1C and Supplementary Fig. S2) as well as 12 months later (Supplementary Fig. S1). Mice were regularly weighed and the weights were compared using Two Way ANOVA (p < 0.05 was considered statistically significant). In some cases, mice were anesthetized with ketamine (100 mg/kg, Zoetis Inc., Parsippany-Troy Hills, NJ) and xylazine (15 mg/kg, Akorn Pharmaceuticals, Lake Forest, IL) before delivering a single dose of intratracheal 50 µl 0.9% normal saline (NS, B. Braun Medical Inc., Bethlehem, PA). At experimental endpoint, mice were anesthetized via intraperitoneal injection of ketamine (200 mg/kg, Zoetis Inc.) and xylazine (30 mg/kg, Akorn Pharmaceuticals), and euthanized by exsanguination through severing the carotid artery or cardiac puncture before tissue sample collection.
Bronchoalveolar lavage
Mice were treated with tamoxifen daily for seven days (100 mg/kg/day, IP). One or 15 days later, 0.9% NaCl was delivered intratracheally, followed by collection of BAL fluid three days later, as previously described^33^. Following anesthesia, mice were exsanguinated and intubated with a catheter and lavaged five times with 0.8-1.0 mL of 0.9% NaCl. Cells were isolated from the collected BAL fluid, and total cells counted. A proportion of the cells was stained using Wright Giemsa stain (Fisher Scientific, Kalamazoo, MI), and counted to assess the ratio of neutrophils, macrophages, lymphocytes, and eosinophils out of a total of 200 cells. Kolmogorov-Smirnov test was used to test data normality. Data sets that did not pass normality test (lymphocyte and eosinophil counts), data was log-transformed. Unpaired, two-tailed Student’s t-test was utilized for statistical analyses (p < 0.05 was considered statistically significant).
Blood panel analysis
For complete blood count and complete metabolic panel, terminal blood collection was performed via cardiac puncture. Blood was aliquoted into one ethylenediaminetetraacetic acid (EDTA)-coated tube for complete blood count (CBC) and one Lithium Heparin-coated tube for chemistry analysis. Whole blood in EDTA at room temperature was used for CBC analysis on Zoetis Vetscan HM5 (Zoetis Services LLC, Parsippany, NJ) within eight hours of collection. Heparinized whole blood was centrifuged (4000 RPM for ten minutes at room temperature) to collect plasma. Plasma was loaded into the Preventative Care Profile Plus rotors (Zoetis #10023238) and subjected to analysis via Zoetis Vetscan VS2. Individual markers were compared between the two groups using unpaired, two-tailed Student’s t-test, adjusted with the Benjamini-Hochberg method to account for FDR (p < 0.05 was considered statistically significant).
Polymerase chain reaction
DNA was digested from mouse tail clips using Direct PCR (Tail) lysis reagent (Viagen Biotech Inc., Los Angeles, CA) supplemented with proteinase K (Thermo Fisher Scientific, Waltham, MA), and overnight shaking at 55 °C. Subsequently, the reaction was quenched by heating to 85 °C for 45 min before PCR in Taq buffer with KCl (Fermentas, Waltham, MA), 2 mM MgCl_2_ (Fisher Scientific, Hampton, NH), 0.2 mM dNTP (Thermo Fisher Scientific), and Taq DNA Polymerase (Fisher Scientific). PCR cycling included 1 cycle at 94 °C (3 min), 35 cycles at 94 °C (45 s), 57 °C (60 s) then 72 °C (75 s), and 1 cycle at 72 ° (10 min). The resulting samples were mixed with 6X sample dye before being run on 2% agarose gels for 30 min at 100 V and read using GE LAS4000 chemiluminescence imager. One-Way ANOVA with Šídák’s post-hoc test was used to compare data (p < 0.05 was considered statistically significant).
Reverse transcription quantitative polymerase chain reaction (RT-qPCR)
Whole mouse lungs were used for isolation of RNA. Lungs were flash frozen and suspended in RNA STAT-60 (Amsbio, Abingdon, UK) and then bullet blended with RNase free 0.9–2.0 mm stainless steel beads SSB14B-RNA (Next Advance, Troy, NY) until homogenized. Additional RNA STAT-60 followed by chloroform were added prior to centrifugation. RNA was precipitated from the resulting colorless upper aqueous phase and the lower phenol phase containing protein was utilized for protein extraction (see below). RNA pellets were washed with 75% ethanol and resuspended in nuclease free water. cDNA was then synthesized using the iScript cDNA synthesis kit (170–8841 Bio-Rad, Hercules, CA). Primers were designed and purchased from Integrated DNA Technologies (Coralville, IA) or the University of Wisconsin-Madison Biotechnology Center (Madison, WI). Primer pairs are listed in Supplementary Tables S4 and S5 and were tested at different concentrations to determine optimal efficiency with satisfactory melting curves. Succinate dehydrogenase complex flavoprotein subunit A (Sdha) was used as a housekeeping gene control. iTaq Universal SYBR Green Supermix (172–5120 Bio-Rad) was used to for RT-qPCR following the manufacturer’s recommendations in an Applied Biosystems 7500 multi-color StepOne™ Real-Time PCR Detection System. Unpaired, two-tailed Student’s t-test (p < 0.05 was considered statistically significant) was applied to the ΔΔCt values to assess statistical significance.
Western blot
Protein from mouse whole lung lysates from the lower red phenol phase following RNA isolation (RNA STAT-60 protocol, as described above) was precipitated in acetone. The protein pellet was dispersed in 0.3 M guanidine hydrochloride (Sigma Aldrich, St. Louis, MO) in 95% ethanol + 2.5% glycerol (Fisherbrand, Pittsburg, PA), centrifuged, and washed according to the Amsbio protocol. The resulting pellet was solubilized in 5% sodium dodecyl sulfate (SDS, Sigma Aldrich)/urea buffer (EM Science, Cherry Hill, NJ) supplemented with 1% sodium pyrophosphate (Sigma Aldrich), 1% protease inhibitor cocktail (Thermo Fisher Scientific), 200 μm Na-orthovanadate (ICN, Biomedical Inc. Costa Mesa, CA), and 0.1% phenylmethanesufonyl fluoride (PMSF, Sigma Aldrich). The samples were heated for three minutes at 100 °C and 20 µL of sample were loaded and run with polyacrylamide gel electrophoresis and analyzed by Western blot, with primary and corresponding Horseradish Peroxidase (HRP)-conjugated secondary antibodies, before being developed by enhanced chemiluminescence (ECL) (GE, Boston, MA). Antibodies included Tensin1 Polyclonal Antibody (rabbit from Invitrogen # PA5-116023), β-actin monoclonal antibody (mouse from Sigma, #A1978), Goat Anti-Rabbit IgG H & L Chain Specific Peroxidase Conjugate (EMD Millipore # 401393), and Goat Anti-Mouse IgG (H + L) Secondary Antibody HRP (Invitrogen # 32430). Images were obtained using GE LAS4000 chemiluminescence imager. Densitometry of selected blots was analyzed via ImageJ and unpaired, two-tailed Student’s t-test was applied to compared samples (p < 0.05 was considered statistically significant).
Immunohistochemistry and veterinary pathologist review
Mouse whole lungs were inflated with 10% neutral buffered formalin (Fisherbrand), while the remaining vital organs, including the heart, medial lobe of the liver, spleen, and kidneys, were removed. All tissues were fixed in 10% neutral buffered formalin for 24 h before being submerged in 70% EtOH and subjected to H&E or trichrome staining (HistoWiz, Brooklyn, NY, USA or University of Wisconsin-Madison Comparative Pathology Laboratory). Microscopic tissue assessment was performed by a board-certified veterinary pathologist on tissues from Tns1-/- mice (n = 2 males, 1 female) compared to the tamoxifen-treated B6.Tns1^lox, lox^ mouse (n = 1 male) and the untreated B6.Tns1^lox, lox^ mouse (n = 1 male) to assess any changes in the overall tissue histopathology as a result of TNS1 knockout 14 days post-tamoxifen administration. At select time points, post-tamoxifen delivery the mouse organs were also examined, including the lung, aorta, heart, kidneys, adrenal glands, pancreas, spleen, liver, ovaries, and fallopian and uterine tubes (in female mice only). This was done at three month (n = 1 Tns1+/+ male, 1 Tns1-/- male, and 3 Tns1-/- females), six month (n = 3 Tns1+/+ males, 2 Tns1-/- males, and 2 Tns1-/- females), and 12 month (n = 1 Tns1+/+ male, 1 Tns1-/- male, and 3 Tns1-/- females) time points. Select organs were analyzed from additional mice at three months post-tamoxifen, including ovaries, fallopian and uterine tubes (n = 1 Tns1+/+ female and 1 Tns1-/- female), lungs (n = 1 Tns1+/+ male), and kidneys (n = 1 Tns1-/- male).
RNA preparation for sequencing
Whole lung RNA lysates from the Tns1-/- mouse cohort (n = 1 male, 2 females) were compared to age-matched tamoxifen-treated B6.Tns1^lox, lox^ littermate controls (n = 2 males, 3 females). All mice were treated with tamoxifen (100 mg/kg/day, IP) for seven days, followed by a one-time dose of intratracheal 50 µl 0.9% normal saline (NS, B. Braun Medical Inc.). Fourteen days later, the left lung was flash frozen for RNA extraction and sequencing. RNA was processed as described above under RT-qPCR with the addition of the Ambion (Austin, TX) AM1907 Turbo DNA-free kit to remove any remaining DNA prior to synthesis of cDNA.
RNA sequencing with Bioinformatics Resource Core (BRC) within the UW Biotechnology Center (UWBC) and quality control
Samples were submitted to the University of Wisconsin-Madison Biotechnology Center Gene Expression Center for RNA sequencing and to the Bioinformatics Resource Core for RNA sequencing analysis. RNA was analyzed for quality control (QC) (Nanodrop and Agilent Bioanalyzer). All samples met QC metrics by 28 S/18S rRNA ratio and RNA Integrity Number (RIN). Bulk RNA sequencing analysis was then performed using the Illumina NovaSeq6000 S4 System (2 × 150 bp, ~ 66 M reads/sample) with the Illumina TruSeq Stranded mRNA (polyA enrichment) Library. Sequencing reads were first demultiplexed. Next, low-abundance genes with an average read count below a threshold of 1.0 in two or more samples were filtered and then reads were trimmed using the trimming software Skewer.
Bioinformatic analysis of transcriptomic data adhered to recommended ENCODE guidelines and best practices for RNA-Seq ^34^. Alignment of adapter-trimmed (Skewer v0.1.123 ^35^), 2 × 150 (paired-end, PE) base pair strand-specific Illumina reads to the Mus musculus GRCm38.p5 genome (assembly accession NCBI GCA 000001635.7) was achieved with the Spliced Transcripts Alignment to a Reference (STAR v2.5.3a) software^36^, a splice-junction aware aligner, using annotation provided by Ensembl release 91. Expression estimation was performed with RNASeq by Expectation Maximization (RSEM v1.3.0, ^37^). To test for differential gene expression among individual group contrasts, expected read counts obtained from RSEM were used as input into edgeR (v3.16.5, ^38^). Inter-sample normalization was achieved with the trimmed mean of M-values (TMM, ^39^) method. Statistical significance of the negative-binomial regression test was adjusted with a Benjamini-Hochberg FDR correction at the 5% level^40^. Prior to statistical analysis with edgeR, independent filtering was applied and required genes to have a count-per-million (CPM) above k in n samples, where k is determined by minimum read count (10 reads) and by the sample library sizes where n is determined by the number of biological replicates in each group. The validity of the Benjamini-Hochberg FDR multiple testing procedure was evaluated by inspection of the uncorrected p-value distribution.
STRING database pathway analysis
Differential gene expression analysis of the Tns1-/- mouse cohort compared to tamoxifen treated Tns1+/+ controls was reported using the ‘edgeR glm approach’. Log fold of control (log(FC)), log counts per million (log(CPM)), the likelihood ratio (LR), unadjusted p value (p), and adjusted p value or FDR were reported. Data was sorted by FDR. The STRING online database (String-db.org) was used to construct PPI networks^41–43^. For all differentially expressed genes with an FDR < 0.05 (N = 171 genes), the gene name and log(FC) values were entered into the ‘Proteins with Values/Ranks - Functional Enrichment Analysis’ function of the STRING database with the option ‘Submit data to normal geneset analysis’. The STRING database recognized 154 of these genes, with the majority of unrecognized genes being predicted sequences. Supervised MCL clustering with an inflation parameter of 2.2 was applied. Clusters were exported to Cytoscape and formatted such that a darker and wider gene outline indicates a higher level of differential gene expression.
Functional enrichment analyses
Functional enrichment analyses were identified using the top five Molecular Function and Cellular Component (Gene Ontology) and Tissue Expression (TISSUE), sorted by FDR. The enrichment of biological pathways supplied by KEGG was also reported^15–17^. For heatmap analysis, the Z-score of a given expression value is the number of standard deviations away from the mean of all the expression values for that gene. The Z-score was scaled across rows of genes normalized by TMM to determine gradation of color. Genes and samples were clustered using Euclidean distances and an elbow function was applied to estimate the number of gene clusters present. Calculated relationships were depicted by dendrograms drawn at the top (samples) and to the left (genes) of the heatmap.
Cell culture
THP-1 cells (TIB-202, ATCC, Manassas, VA) passages 13–22 were seeded in appropriate densities (see below) and treated with 100 nM phorbol 12-myristate 13-acetate (PMA, Sigma Aldrich) for 24 h in RPMI 1640 growth medium (Corning, Corning, NY), containing 10% fetal bovine serum (Cytiva, Wilmington, DE), 1% penicillin-streptomycin with amphotericin B (Corning), and 2% sodium pyruvate (Corning) to induce differentiation and attachment. Cells were then transfected with 25nM TNS1 or scrambled siRNA in Opti-MEM™ reduced-serum medium (Thermo Fisher Scientific) for 24 h. siRNA were from Qiagen, including human TNS1-1 sequence AGCGGTATGTTCCGCTCTCAA (Allstars SI00134092)^10^ and scrambled control (1027310). Following transfection, cells were incubated with starvation RPMI medium containing 0.1% bovine serum albumin (BSA, Sigma Aldrich). Cells were then used for subsequent experiments.
Phagocytosis assay
THP-1 cells were plated at a density of 2.0 × 10⁵ cells/well in a 96-well plate and manipulated as described above. After TNS1 was knocked down, cells were subsequently treated with 100 µg/mL pHrodo™ (Thermo Fisher Scientific) at 37 °C for 1 h. Control cells were incubated with 10 µM cytochalasin D (Sigma-Aldrich) at 37 °C both before and during pHrodo™ treatment. After treatment, cells were detached using TrypLE™ Express (Thermo Fisher Scientific), and resuspended in FA3 buffer, composed of 1X DPBS (Ca2 + Mg2+ free, Corning) with 10mM HEPES (Corning), 2mM EDTA (Millipore-Sigma) and 1% FBS (Cytiva). The cells were then plated into a 96-well plate (Corning) for flow cytometry analysis. Flow cytometric analysis was performed on Cytek Aurora Spectral Flow Cytometer at the University of Wisconsin Carbone Cancer Center Flow Cytometry Laboratory. Forward scatter (FSC) files were down sampled to start with the same number of events in each sample ahead of analysis. Cells were gated for singlets using Side scatter height vs. Side scatter area (SSC-H vs. SSC-A). Samples were then analyzed for expression of pHrodo™ Deep Red. One way ANOVA with Šídák’s post-hoc test was utilized for statistical analysis (p < 0.025 was considered statistically significant).
THP-1 cell size estimation
THP-1 cells were seeded onto FN-coated (Millipore-Sigma) 8-well chamber µ-slides (Ibidi, Munich, Germany) slides at a density of 6.25 × 10^4^ cells/well and treated with PMA (as above). After TNS1 knockdown, cells were fixed for 20 min in 4% PFA and stained with Rhodamine Phalloidin Reagent (Abcam) for 1 h at room temperature. Images were obtained on a Nikon Ti2 Eclipse Microscope. Three 40X images (n = 15–55 cells/image) for each condition were analyzed for area using the CellPose plugin for ImageJ, while manually excluding incorrect segmentation^44^. FSC-Height (FSC-H) was assessed from phagocytosis flow cytometry experiments, above. Duplicate singlet gates were utilized to ensure doublets were excluded and the mean of the FSC-H was analyzed. Data is presented as mean ± SD and was analyzed via Unpaired Student’s t-test (* p < 0.05).
Migration assay
THP-1 cells were plated at a density of 1.0 × 10^6^ cells in a 6-well plate and treated with PMA (as above). Following TNS1 knockdown, cells were detached using TrypLE™ Express and then re-seeded onto fibronectin-coated 8-well µ-slides at a density of 6.25 × 10^4^ cells/well. Cells were allowed to reattach for 20 min at 37 °C before imaging. Cells were placed in Tokai Hit chamber and 10X brightfield images were taken every 15 min for 16 h on a Nikon Ti-E Inverted Microscope. Two 10X hyperstacks (n = 2–11 cells/hyperstack) for each condition were analyzed using TrackMate package on ImageJ with manual exclusion of improperly identified cells to provide migration distance^45,46^. Data is presented as mean ± SD and was analyzed via Unpaired Student’s t-test (* p < 0.05).
Supplementary Information
Below is the link to the electronic supplementary material.
Supplementary Material 1
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Duan, J. et al. mi R-152/TNS 1 axis inhibits non-small cell lung cancer progression through Akt/m TOR/Rho A pathway. Biosci. Rep.4110.1042/BSR 20201539 (2021).10.1042/BSR 20201539 PMC 778504033269380 · doi ↗ · pubmed ↗
- 2Adams, T. S. et al. Single-cell RNA-seq reveals ectopic and aberrant lung-resident cell populations in idiopathic pulmonary fibrosis. Sci Adv 6, eaba (1983). 10.1126/sciadv.aba 1983 (2020).10.1126/sciadv.aba 1983 PMC 743950232832599 · doi ↗ · pubmed ↗
- 3Nemoto, S., Uchida, K. & Ohno, H. Implications of Genetic Factors Underlying Mouse Hydronephrosis: Cautionary Considerations on Phenotypic Interpretation in Genetically Engineered Mice. Int. J. Mol. Sci.2510.3390/ijms 25137203 (2024).10.3390/ijms 25137203 PMC 1124151339000307 · doi ↗ · pubmed ↗
- 4Calderwood, D. A. et al. Integrin beta cytoplasmic domain interactions with phosphotyrosine-binding domains: a structural prototype for diversity in integrin signaling. Proc. Natl. Acad. Sci. U. S. A. 100, 2272–2277, (2003). 10.1073/pnas.26279199910.1073/pnas.262791999 PMC 15133012606711 · doi ↗ · pubmed ↗
- 5Monaco, G. et al. RNA-Seq Signatures Normalized by m RNA Abundance Allow Absolute Deconvolution of Human Immune Cell Types. Cell Rep 26, 1627–1640 e 1627, (2019). 10.1016/j.celrep.2019.01.04110.1016/j.celrep.2019.01.041PMC 636756830726743 · doi ↗ · pubmed ↗
- 6Karlsson, M. et al. A single-cell type transcriptomics map of human tissues. Sci. Adv.710.1126/sciadv.abh 2169 (2021).10.1126/sciadv.abh 2169 PMC 831836634321199 · doi ↗ · pubmed ↗
- 7Uhlen, M. et al. A genome-wide transcriptomic analysis of protein-coding genes in human blood cells. Science 36610.1126/science.aax 9198 (2019).10.1126/science.aax 919831857451 · doi ↗ · pubmed ↗
- 8Bernau, K. et al. Megakaryoblastic leukemia-1 is required for the development of bleomycin-induced pulmonary fibrosis. Respir Res.1610.1186/S 12931-015-0206-6 (2015). doi:Artn 45.10.1186/s 12931-015-0206-6PMC 439277825885656 · doi ↗ · pubmed ↗