Abstract C24: Beyond EMT: Vimentin function in lung cancer metastasis

Vimentin is an intermediate filament protein that has been used in the clinic as a biomarker for metastatic potential and poor patient prognosis across numerous solid tumor types; however, little is known about how vimentin might contribute to cancer progression and metastasis. Here we created a novel genetically engineered mouse model (GEMM) to study the role of vimentin in lung cancer progression and metastasis. We first re-developed an LSL-KrasG12D,LKB1fl/fl mouse model whereby intranasal administration of lentiviral Cre recombinase causes a KrasG12D mutation and LKB1 knockout in the lungs. In this model Kras/Lkb1 GEMM (termed KLV+/+), approximately half the mice that develop primary lung tumors in 8-10 weeks also develop metastasis to the mediastinal lymph nodes in weeks 12-16. We used this model to test the hypothesis that vimentin loss prevents lung cancer metastasis by crossing this GEMM with a Vim-/- mouse to create a novel LSL-KrasG12D, LKB1fl/fl, Vim-/- (termed here KLV-/-) mouse. Our findings show that in the KLV-/- mouse, primary tumor burden and cell proliferation were not significantly different than wild-type KLV+/+ mice; however, KLV-/- mice exhibit significantly less metastasis to the mediastinal lymph nodes compared to their wild type counterpart indicating that vimentin contributes to the metastatic cascade. Consistent with this finding, histological analysis of the primary tumors show that KLV-/- mice exhibit less focal invasion than KLV+/+ and +/- mice, further confirming that KLV-/- mice have reduced invasiveness. Interestingly, vimentin staining in invasive regions of KLV+/+ mice was not found in the cancer cells but rather in fibroblast-like cells surrounding invasive cell buds that we term collective invasion packs (CIPs). These fibroblast-like, vimentin-positive cells also stain positive for alpha smooth muscle actin, which is consistent with cancer-associated fibroblasts (CAFs). Taken together, these results suggest that in this highly metastatic genetic background, vimentin may play a central role in recruiting CAFs to invading cancer cells but not necessarily in cancer cell epithelial to mesenchymal transition (EMT).

Citation Format: Alessandra M. Salgueiro, Melissa Gilbert-Ross, Lauren S. Havel, John Shupe, Allyson E. Koyen, Hans E. Grossniklaus, Gabriel Sica, Adam I. Marcus. Beyond EMT: Vimentin function in lung cancer metastasis. [abstract]. In: Proceedings of the AACR Special Conference: Function of Tumor Microenvironment in Cancer Progression; 2016 Jan 7–10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2016;76(15 Suppl):Abstract nr C24.

Abstract 2290: More than a biomarker: Studying the role of vimentin in lung cancer metastasis

A major mechanism of metastasis is the epithelial to mesenchymal transition (EMT) where epithelial cells from the primary site undergo genetic and epigenetic alterations to become motile and mesenchymal. A canonical marker of EMT and a clinical marker of metastasis is the intermediate filament protein, vimentin. Vimentin expression correlates with increased metastatic potential and poor patient prognosis across most solid tumor types including lung, prostate, and breast cancers; however little is known about how vimentin functions to promote the metastatic cascade. Here we use a lung cancer mouse model to study the role of vimentin in lung cancer metastasis in vivo. We have re-developed an LSL-KrasG12D,LKB1fl/fl mouse model whereby intranasal administration of lentiviral Cre recombinase causes a KrasG12D mutation and LKB1 knockout in the lungs. Consequently, about half of the mice that develop primary lung tumors after 10 weeks also develop metastasis to the mediastinal lymph nodes by week 16. We crossed this mouse with a Vim-/- mouse to create a LSL-KrasG12D,LKB1fl/fl, Vim-/- (termed here KLV) mouse to determine if loss of vimentin prevents lung cancer metastasis in this model. Our findings show that the KLV mouse exhibits significantly less metastasis to the mediastinal lymph nodes (12%) as compared with its vimentin wild type counterpart but does not show a decrease in primary tumor formation. These knockout mice also exhibit less focal invasion at the primary tumor site while maintaining a well-differentiated adenocarcinoma histology independent of vimentin genotype. Taken together, our work indicates that vimentin is critical for lung cancer metastasis but not primary tumor formation.

Citation Format: Alessandra M. Salgueiro, Melissa Gilbert-Ross, Lauren S. Havel, John Shupe, Adam I. Marcus. More than a biomarker: Studying the role of vimentin in lung cancer metastasis. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 2290. doi:10.1158/1538-7445.AM2015-2290

Vimentin regulates lung cancer cell adhesion through a VAV2-Rac1 pathway to control focal adhesion kinase activity

Vimentin is an intermediate filament protein whose expression correlates with increased metastatic disease, reduced patient survival, and poor prognosis across multiple tumor types. Despite these well-characterized correlations, the molecular role of vimentin in cancer cell motility remains undefined. To approach this, we used an unbiased phosphoproteomics screen in lung cancer cell lines to discover cell motility proteins that show significant changes in phosphorylation upon vimentin depletion. We identified the guanine nucleotide exchange factor (GEF) VAV2 as having the greatest loss of phosphorylation due to vimentin depletion. Since VAV2 serves as a GEF for the small Rho GTPase Rac1, a key player in cell motility and adhesion, we explored the vimentin-VAV2 pathway as a potential novel regulator of lung cancer cell motility. We show that VAV2 localizes to vimentin positive focal adhesions (FAs) in lung cancer cells and complexes with vimentin and focal adhesion kinase (FAK). Vimentin loss impairs both pY142-VAV2 and downstream pY397-FAK activity showing that vimentin is critical for maintaining VAV2 and FAK activity. Importantly, vimentin depletion reduces the activity of the VAV2 target, Rac1, and a constitutively active Rac1 rescues defects in FAK and cell adhesion when vimentin or VAV2 is compromised. Based upon this data, we propose a model whereby vimentin promotes FAK stabilization through VAV2-mediated Rac1 activation. This model may explain why vimentin expressing metastatic lung cancer cells are more motile and invasive.

HOMA-IR and non-HDL-C as predictors of high cholesteryl ester transfer protein activity in patients at risk for type 2 diabetes

BACKGROUND AND AIMS

Metabolic syndrome (MS) and type 2 diabetes are highly associated with an abnormal lipoprotein profile, which may be generated and accentuated by high cholesteryl ester transfer protein (CETP) activity. Given the difficulty in measuring CETP activity, the aim was to identify simple biochemical predictors of high CETP activity.

DESIGN AND METHODS

Eighty five subjects at risk for type 2 diabetes were classified according to the presence of MS. Lipoprotein profile, HOMA-IR and endogenous CETP activity were evaluated.

RESULTS

As expected, MS patients presented higher concentration of glucose, insulin, triglycerides and non-HDL-C and lower HDL-C levels. Moreover, MS patients exhibited increased HOMA-IR and CETP activity. Employing a ROC curve for MS, high CETP activity was defined as >250%ml⁻¹ h⁻¹. The predictive variables of high CETP were non-HDL-C≥160mg/dl (OR=11.1;95%IC=3.3-38.2;p<0.001) and HOMA-IR>2.1 (OR=4.4;95%IC=1.3-14.8;p<0.05).

CONCLUSIONS

High non-HDL-C and insulin resistance were predictors for increased CETP activity which measurement is not accessible for clinical laboratories.

[Central obesity: relationship between conicity index and lipoprotein risk factors for coronary atherosclerosis]

In middle age men the relationship between biochemical and anthropometric parameters was studied associated with central obesity in order to evaluate the risk of coronary atherosclerosis. In 31 males of 30 to 65 years of age, "apparently healthy", we determined the percentile 50 (P50) for body fat percentage (PG), 26.8% (utilizing the skinfold thicknesses) and P75 of Conicity Index (IC), 1.26. Were selected 63 subjects with different values of PG and IC. IC, adjusted for BMI and PG, correlated with Cholesterol (CT), Triglycerides (TG), CT/C-HDL, C-LDL/C-HDL, Apolipoprotein B (ApoB), Apo B/C-HDL, (P < 0.01) in all cases), and cholesterol LDL (C-LDL) (P < 0.05). This population was divided in three groups: Group I, Control, N = 19, PG < = 27% and IC < 1.26; Group II, obese without Central Obesity Predominance (POC), N = 15, PG < or = 27% and IC < or = 1.26; Group III, N = 23, obese with POC, PG > 27% and IC < or = 1.26. We found that Group III vs Group II had: CT 242 +/- 35 vs 205 +/- 40 mg/dl (P < 0.01), C-LDL 165 +/- 41 vs 138 +/- 36 mg/dl (P < 0.05) and CT/C-HDL 6.0 +/- 1.2 VS 5.1 +/- 1.2 (p < 0.05) respectively. Group III vs I showed significant differences for all biochemical parameters and index studied with exception of C-HDL. In subjects with IC > 1.26 vs IC < = 1.26 increased the frequency of coronary risk factors and indicators: CT > or = 240 mg/dl, 59% vs 12% (P < 0.001); C-LDL > or = 160 mg/dl, 55% vs 18% (P < 0.001); Apo B > 120 mg/dl, 72% vs 26% (P < 0.001); CT/C-HDL > 4.5, 86% vs 53% (P < 0.01); C-LDL/C-HDL > 3.0, 86% vs 47% (P < 0.001); Apo B/C-HDL > 2.6, 72% vs 47% (P < 0.05). Our results suggest that the use of IC combined with lipoproteic factors and index will contribute to the detection of males at risk of coronary heart disease.