The Initiator Methionine tRNA Drives Secretion of Type II Collagen from Stromal Fibroblasts to Promote Tumor Growth and Angiogenesis

Abstract

Expression of the initiator methionine tRNA (tRNAi(Met)) is deregulated in cancer. Despite this fact, it is not currently known how tRNAi(Met) expression levels influence tumor progression. We have found that tRNAi(Met) expression is increased in carcinoma-associated fibroblasts, implicating deregulated expression of tRNAi(Met) in the tumor stroma as a possible contributor to tumor progression. To investigate how elevated stromal tRNAi(Met) contributes to tumor progression, we generated a mouse expressing additional copies of the tRNAi(Met) gene (2+tRNAi(Met) mouse). Growth and vascularization of subcutaneous tumor allografts was enhanced in 2+tRNAi(Met) mice compared with wild-type littermate controls. Extracellular matrix (ECM) deposited by fibroblasts from 2+tRNAi(Met) mice supported enhanced endothelial cell and fibroblast migration. SILAC mass spectrometry indicated that elevated expression of tRNAi(Met) significantly increased synthesis and secretion of certain types of collagen, in particular type II collagen. Suppression of type II collagen opposed the ability of tRNAi(Met)-overexpressing fibroblasts to deposit pro-migratory ECM. We used the prolyl hydroxylase inhibitor ethyl-3,4-dihydroxybenzoate (DHB) to determine whether collagen synthesis contributes to the tRNAi(Met)-driven pro-tumorigenic stroma in vivo. DHB had no effect on the growth of syngeneic allografts in wild-type mice but opposed the ability of 2+tRNAi(Met) mice to support increased angiogenesis and tumor growth. Finally, collagen II expression predicts poor prognosis in high-grade serous ovarian carcinoma. Taken together, these data indicate that increased tRNAi(Met) levels contribute to tumor progression by enhancing the ability of stromal fibroblasts to synthesize and secrete a type II collagen-rich ECM that supports endothelial cell migration and angiogenesis.

Keywords: cell migration; extracellular matrix; initiator methionine tRNA (tRNA(i)(Met)); secretome; tRNA repertoire; tumor angiogenesis; tumor stroma.

Figure 1

Increased Expression of tRNA_i^Met in the Host Animal Promotes Angiogenesis and Growth of Allografts (A) 1 × 10⁶ transformed melanocytes (Tyr::NrasQ61K/°;INK4a−/−) were injected subcutaneously into either wild-type (WT) or 2+tRNA_i^Met transgenic male mice and tumor size monitored. Mice were culled once tumors reached endpoint (15 × 15 mm in size). Kaplan-Meier survival curves were plotted based on time taken to reach endpoint. n = 14 (WT); n = 14 (2+tRNA_i^Met mice); log rank (Mantel-Cox) test; ^∗p < 0.05. (B) 1 × 10⁶ LLC cells were injected subcutaneously into either wild-type (WT) or 2+tRNA_i^Met transgenic male mice, and tumors were harvested at 21 days. The graph on the right shows final tumor volume ± SEM. Tumors were fixed and stained to visualize endomucin for blood vessel identification and vessels counted across the entire tumor section. Graph on the left shows tumor vessel density ± SEM; n = 10 (WT); n = 10 (2+tRNA_i^Met); unpaired t test; ^∗p < 0.05. (C) 0.5 × 10⁶ B16 F0 mouse melanoma cells were injected subcutaneously into either wild-type (WT) or 2+tRNA_i^Met transgenic mice. The time at which palpable tumors were formed was recorded (minimum size of 3 × 3 mm) and then tumors were harvested 22 days from this time point. Tumors were fixed and stained to visualize endomucin for blood vessel identification and vessels counted across the entire tumor section. The graph on the right shows final tumor volume ± SEM. The graph on the left shows tumor vessel density ± SEM; n = 20 (WT); n = 20 (2+tRNA_i^Met); unpaired t test; ^∗p < 0.05; ^∗∗p < 0.005.

Figure 2

Fibroblasts Isolated from 2+tRNA_i^Met Transgenic Mice Deposit a Pro-migratory ECM (A) Extracellular matrix (ECM) was generated from primary lung fibroblasts isolated from either wild-type (WT) or 2+tRNA_i^Met transgenic mice. See Figure S2 for characterization of these cells. Migration of human umbilical vein endothelial cells (HUVECs) on the ECM was recorded by time-lapse microscopy and analyzed using ImageJ. Data represent ECM generated from seven independent isolates of primary fibroblasts from pools of three WT and three 2+tRNA_i^Met transgenic mice, tracking the migration of at least 28 HUVECs in each replicate. Box and whisker plot represents 5–95 percentile; Mann-Whitney test; ^∗∗∗∗p < 0.0001. (B) ECM was generated from primary fibroblasts isolated from either WT or 2+tRNA_i^Met primary mouse embryonic fibroblasts (MEFs)—these cells are characterized in Figure S2. Migration of immortalized MEFs (iMEFs) on the ECM was recorded by time-lapse microscopy over a 17-hr time course and analyzed using ImageJ. Data represent ECM generated from primary fibroblasts from two pairs of litter-matched WT and 2+tRNA_i^Met transgenic mice, tracking the migration of at least 60 iMEFs in each replicate; n = 3 for pair 1; n = 2 for pair 2. Box and whisker plot represents 5–95 percentile; Mann-Whitney test; ^∗∗∗∗p < 0.0001.

Figure 3

tRNA_i^Met Supports Deposition of Pro-migratory ECM via Release of Secreted Factors (A) Migration of iMEFs on ECM was recorded by time-lapse microscopy over a 17-hr period and analyzed using ImageJ. ECM was generated using immortalized fibroblasts (iMEFs) stably overexpressing either empty vector (iMEF-vector) or tRNA_i^Met (iMEF-tRNA_i^Met). See Figure S3 for characterization of these cells. (B) ECM was generated by iMEFs in the presence of conditioned media from iMEF-vector or iMEF-tRNA_i^Met cells. All data represent ECM generated from at least three independent ECM isolations, tracking the migration of at least 40 iMEFs in each replicate. Box and whisker plot represents 5–95 percentile; Mann-Whitney test; ^∗∗∗∗p < 0.0001. (C) Atomic force microscopy and contact imaging were performed using a JPK NanoWizard II Bio AFM in combination with a Bruker MLCT cantilever and to measure the thickness of ECM following wound scratch or a Nanoworld Arrow TL-1 cantilever with bead attached for force spectroscopy; n = 3 independent ECM generations; ±SEM; ^∗∗∗∗p < 0.0001; ns, no significant difference; Kruskal-Wallis test.

Figure 4

tRNA_i^Met Drives a Secretome that Is Enriched in Type II Collagen (A) The cellular proteome was analyzed following tRNA_i^Met overexpression using quantitative SILAC mass spectrometry. iMEF-vector and iMEF-tRNA_i^Met cells were labeled with heavy and light amino acids. Cells were lysed and proteins separated by gel electrophoresis and analyzed by mass spectrometry. Forward experiment (Fw) corresponds to iMEF-tRNA_i^Met labeled with heavy amino acids combined with iMEF-vector light amino acids, and the reverse experiment (Rev) corresponds to iMEF-vector heavy-labeled amino acids combined with iMEF-tRNA_i^Met light amino acids. ECM proteins are highlighted in red. (B) The secretome was analyzed following tRNA_i^Met overexpression using quantitative SILAC mass spectrometry. Cells were labeled as for (A), and conditioned medium was collected. Proteins were concentrated from conditioned medium using Strataclean beads and subjected to mass spectrometry analysis as for (A). ECM proteins are highlighted in red. (C) A table of the most significantly upregulated secreted proteins identified in conditioned medium following tRNA_i^Met overexpression. Upregulated ECM proteins are highlighted in red. (D) Data from (B) replotted to show the total abundance of secreted proteins identified in the media of iMEF-tRNA_i^Met against the fold change in expression of the same proteins in the iMEF-tRNA_i^Met compared to iMEF-vector. ECM proteins are highlighted in red. (E) ECM was generated from iMEF-vector and iMEF-tRNA_i^Met cells. Incorporation of collagen II into the ECM was assessed by immunostaining using type II collagen-specific antibodies and quantified using the Aerius infrared imaging system (LI-COR Biosciences).

Figure 5

Collagen II Production Is Required for tRNA_i^Met to Drive Production of a Pro-tumorigenic ECM 1 × 10⁶ LLC cells were injected subcutaneously into either WT or 2+tRNA_i^Met transgenic male mice, followed by daily administration of ethyl-3,4-dihydroxybenzoate (DHB) (40 mg/kg) or vehicle control. Tumors were harvested 21 days following injection. Tumor volumes were measured and tumor sections were stained for collagen II or endomucin. (A and B) Collagen II staining in LLC tumors harvested from WT and 2+tRNA_i^Met mice treated with vehicle or DHB as indicated (A). Collagen II staining was quantified by using the Image J “threshold color” plug-in to highlight the brown staining in the left panels and to render this as a grayscale image as shown in the right-hand panels. Bar, 50 μm. Graph in (B) shows quantification of collagen II staining in LLC tumors harvested from WT and 2+tRNA_i^Met mice treated with vehicle or DHB. (C) Graphs show tumor volumes (right-hand graph) and quantification of endomucin-positive tumor blood vessel density (left-hand graph) in harvested tumors from WT and 2+tRNA_i^Met mice treated with vehicle or DHB as indicated. Values are mean ± SEM; n = 8 (WT; vehicle); n = 7 (2+tRNA_i^Met; vehicle); n = 8 (WT; DHB); n = 7 (2+tRNA_i^Met; DHB); unpaired t test; ^∗p < 0.05; ^∗∗p < 0.001; ns, no significant difference.

Figure 6

Collagen II Secretion Is Required for tRNA_i^Met to Drive Production of a Pro-migratory ECM (A) ECM was generated from iMEFs in the presence of conditioned media from iMEF-tRNA_i^Met cells treated with non-targeting (siNT) or type-II-collagen-specific siRNA (siCollagen II). Migration of immortalized MEFs (iMEFs) on the ECM was recorded by time-lapse microscopy over a 17-hr time course and analyzed using ImageJ. See Figures S6A–S6C for a schematic representation of the protocol for this experiment, the validation of the siRNA, and representative examples of the migration track plots. (B) ECM was generated from primary lung fibroblasts treated with non-targeting (siNT) or type-II-collagen-specific siRNA (siCollagen II). Two independent pools of fibroblasts from 2+tRNA_i^Met mice were used as indicated. Migration of human umbilical vein endothelial cells (HUVECs) on the ECM was recorded by time-lapse microscopy and analyzed using ImageJ. (C) ECM was generated from iMEF-vector, iMEF-tRNA_i^Met, and iMEF-tRNA_i^Met with type II collagen II stable knockdown (iMEF-tRNA_i^Met Collagen II CRISPR). Incorporation of collagen II into the ECM was assessed by immunostaining using type-II-collagen-specific antibodies, quantified using the Aerius infrared imaging system (LI-COR Biosciences); unpaired t test; ^∗p < 0.05; ^∗∗p < 0.005. Migration of iMEFs was recorded as previously described. Data represent ECM generated from at least three independent ECM isolations, tracking the migration of at least 40 iMEFs in each replicate. Box and whisker plot represents 5–95 percentile; Kruskal-Wallis test; ^∗∗∗∗p < 0.0001; ns, not significant. See Figures S6D–S6F for a schematic representation of the protocol for this experiment, the validation of the CRISPR, and representative examples of the migration track plots.

Figure 7

Type II Collagen Expression Predicts Poor Prognosis in Ovarian Cancer A tissue microarray (TMA) containing 62 cases of ovarian cancer was stained for type II collagen. Examples of tumor cores displaying low (left side) and high (right side) histoscores for collagen II are displayed in (A). (B) shows a Kaplan-Meier analysis of patient survival when the cohort is divided into tumors with low (histoscore < 50) and high (histoscore > 50) collagen II expression. Patients with tumors displaying a collagen II histoscore greater than 50 have a 2.992-fold increased hazard ratio for death; p = 0.0002 (log rank test); n = 62. See Figure S7 for Kaplan-Meier analysis of the relationship between collagen II levels in the high-grade serous subset of these tumors.