Rapamycin-insensitive up-regulation of MMP2 and other genes in tuberous sclerosis complex 2-deficient lymphangioleiomyomatosis-like cells

Abstract

Increased matrix metalloproteinase (MMP) activity has been implicated in the pathogenesis of lymphangioleiomyomatosis (LAM). The objective of this study was to investigate how tuberous sclerosis complex (TSC) 1 or TSC2 deficiency alters MMP expression and regulation. We studied immortalized cells that lack TSC2 derived from an angiomyolipoma of a patient with LAM, a TSC2 addback derivative, and murine embryonic fibroblast cells that lack Tsc1 or -2 and respective controls. Global gene expression analysis was performed in the angiomyolipoma and derivative cell lines. MMP levels in the conditioned media from these cells were analyzed by zymography and ELISA. We found increased MMP-2 expression in cells lacking TSC1/TSC2 compared with their respective controls by zymography. MMP-2 overproduction by these cells was not affected by rapamycin treatment. Gene expression analysis confirmed increased MMP-2 gene expression that was not affected by rapamycin. Furthermore, multiple other genes were found to be overexpressed in rapamycin-treated TSC2-deficient cells compared with TSC2(+) cells. We conclude that TSC1/TSC2 deficiency leads to MMP-2 overproduction that is rapamycin-insensitive, and that several genes exhibit similar patterns, suggesting that TSC1/TSC2-dependent, but mammalian target of rapamycin-independent, pathways may be involved in the pathogenesis of LAM.

Figure 1.

Secretion of matrix metalloproteinase (MMP)-2 by tuberous sclerosis complex (TSC) 1– or TSC2-deficient cells. Representative zymography (Zym: MMP-2) of conditioned media and Western blots (WB: pS6, tubulin) of cell lysates are shown. The first two lanes represent a TSC2-deficient human angiomyolipoma (AML) cell and addback control line; the last four lanes represent Tsc2- or Tsc1-null murine embryonic fibroblast (MEF) cells and respective controls. In TSC1/TSC2-deficient cells, pS6 (S240/244) levels are increased, reflecting mammalian target of rapamycin complex (mTORC) 1 activation.

Figure 2.

Lack of effect of rapamycin treatment on MMP-2 secretion by TSC1/TSC2–deficient cells. (A) Representative zymography (Zym: MMP-2) of conditioned media and Western blots (WB: pS6, tubulin) of lysates from TSC2-deficient human AML/lymphangioleiomyomatosis (LAM) cells and Tsc1- or Tsc2-null MEF cells with and without rapamycin treatment for 24 hours are shown. Rapamycin effectively suppressed pS6 (S240/244), but had no effect on MMP-2 secretion by the TSC1/TSC2-deficient cells. (B) ELISA quantification of MMP-2 in the conditioned media of human AML/LAM cells. Means (±SD) of triplicate experiments are presented. Rapamycin-treated TSC2-null cells secreted the same amount of MMP-2 as untreated TSC2-null cells (P = 0.54), whereas TSC2⁺ addbacks secreted much lower MMP-2 (*P < 0.001).

Figure 3.

Lack of effect of mTOR or Rheb knockdown on MMP-2 secretion by the TSC2-deficient AML cell line. (A) Representative zymography (Zym: MMP-2) of conditioned media and Western blots (WB: mTOR, Rheb, pS6, tubulin) of lysates from TSC2-deficient human AML cells. RNA interference with small interfering RNA (siRNA) targeting mTOR (left panel) or Rheb (right panel) successfully reduced expression of the target proteins and reduced pS6 (S240/244) levels, but had no effect on MMP-2 levels in the media. (B) ELISA quantification of MMP-2 levels in conditioned media of TSC2⁻ AML cells treated with siRNA targeting mTOR, or in Rheb. Means (±SD) of triplicate experiments are shown. Compared with control cells, knockdown of mTOR did not affect MMP-2 secretion (P = 0.48), whereas knockdown of Rheb caused a minor increase in MMP-2 production (P < 0.01).