A loss-of-function polymorphism in the propeptide domain of the LOX gene and breast cancer

Abstract

The lysyl oxidase (LOX) gene reverted Ras transformation of NIH 3T3 fibroblasts and tumor formation by gastric cancer cells, which frequently carry mutant RAS genes. The secreted lysyl oxidase proenzyme is processed to a propeptide (LOX-PP) and a functional enzyme (LOX). Unexpectedly, the tumor suppressor activity mapped to the LOX-PP domain, which inhibited tumor formation and the invasive phenotype of NF639 breast cancer cells driven by human epidermal growth factor receptor-2/neu, which signals via Ras. A single-nucleotide polymorphism, G473A (rs1800449), resulting in an Arg158Gln substitution in a highly conserved region within LOX-PP, occurs with an average 473A allele carrier frequency of 24.6% in the HapMap database, but was present in many breast cancer cell lines examined. Here, we show that the Arg-to-Gln substitution profoundly impairs the ability of LOX-PP to inhibit the invasive phenotype and tumor formation of NF639 cells in a xenograft model. LOX-PP Gln displayed attenuated ability to oppose the effects of LOX, which promoted a more invasive phenotype. In a case-control study of African American women, a potential association of the Gln-encoding A allele was seen with increased risk of estrogen receptor (ER)-alpha-negative invasive breast cancer in African American women. Consistently, LOX gene expression was higher in ER-negative versus ER-positive primary breast cancers, and LOX-PP Gln was unable to inhibit invasion by ER-negative cell lines. Thus, these findings identify for the first time genetic polymorphism as a mechanism of impaired tumor suppressor function of LOX-PP and suggest that it may play an etiologic role in ER-negative breast cancer.

Figure 1

The Arg-to-Gln substitution has no effect on protein expression, secretion or processing. A, Sequence alignment spanning the carboxyl terminus of LOX-PP and the amino terminus of the LOX enzyme; identical amino acids in the propeptide region are in grey, arrow marks the cleavage site, amino acid numbers correspond to the last amino acid of LOX-PP shown for each species. B and C, Expression of the proteins was induced in stable pools of NF639-EV, -Pro-LOX WT, -Pro-LOX Gln, -LOX-PP WT (PP WT) or -LOX-PP Gln (PP Gln) cells with 2 μg/mL Dox for 48 h. Medium (MED) was subjected to immunoprecipitation as described. Immunoblot analysis was performed on WCEs (20 μg) and immunoprecipitates using anti-V5 Ab followed by HRP-labeled protein A. All lanes in the panels were from the same blots, and cut to contiguously align the relevant samples.

Figure 2

LOX-PP Gln variant displays reduced ability to attenuate Her-2/neu-mediated signaling and malignant phenotype. A, The indicated NF639 cells were treated with Dox for 4 h, serum starved in DMEM-0.5% FBS with Dox for 48 h and stimulated with 10% FBS for 90 min for phospho-Akt (P-Akt) and phospho-Erk1/2 (P-Erk1/2), or 15 h for cyclin D1 and E-cadherin (E-cad). WCEs (50 μg) were subjected to immunoblotting as indicated. Quantification of this and two duplicate experiments indicate that the PP WT and PP Gln compared to EV DNA (set to 1.0) gave values for P-Akt of 0.67 ± 0.23 and 1.07 ± 0.12, respectively and for P-Erk1/2 of 0.60 ± 0.11 and 0.77 ± 0.12, respectively and for cyclin D1 of 0.51 ± 0.17 and 0.99 ± 0.11, respectively. E-cadherin was undetectable in the NF639-EV cells, precluding analysis of fold changes compared to this sample. B, NF639 cells were co-transfected, in triplicate, with pCX_bsr vector carrying the indicated cDNAs and a GFP expression vector in DMEM-0.5% FBS. The numbers of GFP positive cells were determined 24 h and 72 h post-transfection by fluorescence microscopy (50 × magnification, 9 fields/well). Only LOX-PP WT significantly reduced numbers of GFP positive cells (*, P = 0.009) (LOX-PP Gln, P = 0.97) at 72 h post-transfection as determined by Student's t test. Data are presented as average GFP positive cells/image relative to EV control at 24 h (set at 100%) from three independent experiments ± SD. C, After induction with Dox for 24 h, NF639-EV, -PP WT, and -PP Gln cells were subjected to Matrigel outgrowth assay. After 5 days, cultures were photographed at 50 × magnification and representative images are shown. D. After induction with Dox for 24 h in DMEM-5% FBS, cells were subjected to Matrigel invasion assay for 20 h. LOX-PP WT significantly reduced cell invasion (*, P = 0.0005) whereas LOX-PP Gln was not effective (P = 0.48). Data for the average invasion from three independent experiments relative to vehicle control (set at 100%) ± SD are presented.

Figure 3

LOX-PP Gln variant has reduced ability to suppress Her-2/neu tumor formation. Dox inducible cells were pre-treated with 2 μg/mL Dox and injected subcutaneously in the flanks of NCr^nu/nu nude mice (n = 6). A and C, Tumor volumes were measured and tumor growth curves plotted. Bars represent SEM. B and D, Average tumor weights at day 30. Bars represent SEM. The average tumor weight for NF639-LOX-PP WT xenografts were 52% of those for the NF639-EV group (P = 0.03) (B). There is no significant difference between NF639-LOX-PP Gln xenografts compared to NF639-EV control (P = 0.38) (D).

Figure 4

The G473A polymorphism impairs the ability of Pro-LOX to inhibit transformed phenotype. A, NF639-EV, -Pro-LOX WT, -Pro-LOX Gln cells were starved, stimulated and analyzed as in Fig. 2A. Quantification of this and two duplicate experiments indicate that the Pro-LOX WT and Pro-LOX Gln compared to EV DNA (set to 1.0), respectively gave values for phospho-Erk1/2 of 0.91 ± 0.03 and 2.41 ±1.20; phospho-Akt of 0.68 ± 0.10 and 0.98 ± 0.07; cyclin D1 of 0.71 ± 0.32 and 0.97 ± 0.27; vimentin of 1.01 ± 0.02 and 1.61 ± 0.54; and fibronectin (FN) of 0.91 ± 0.16 and 1.44 ± 0.36. B, Cells were subjected to Matrigel outgrowth assay as described in Fig. 2C. Total colonies and colonies with branching structures were counted across 9 images/sample at 50 × magnification. Numbers below represent percentage of branching colonies/total colonies from two independent experiments. C. Cells were subjected to Matrigel invasion assay as in Fig. 2D. Pro-LOX WT significantly reduced cell invasion (*, P = 2.0 e^-5) whereas Pro-LOX Gln significantly increased invasion of these cells (*, P = 0.002). Data for the average invasion from three independent experiments relative to vehicle control (set at 100%) ± SD are presented.

Figure 5

ER negative breast cancers display elevated LOX gene expression and sensitivity to LOX-PP WT but not the Gln variant. A, Box plots of data from the Van de Vijver (set 1), Sotiriou (set 2) and Hess (set 3) breast carcinoma microarray analyses, accessed using the www.oncomine.org database, are plotted on a log scale (n = sample number). Student's t test, performed through the Oncomine 3.0 software, showed the difference in LOX expression between the two ER groups was significant (P = 3.7 e^-4 (set 1), 2.1 e^-4 (set 2), 1.6 e^-4 (set 3), respectively). B and C, MDA-MB-231 (B) and Hs578T (C) cells were pre-treated with recombinant LOX-PP WT or LOX-PP Gln proteins at 55.5 nM or 111 nM or water vehicle control for 24 h in DMEM-10% FBS or DMEM-5% FBS, respectively. Cells were then subjected, in triplicate, to a Matrigel invasion assay for 24 h (B) or 16 h (C) in the presence of the same doses of proteins. Data for the average invasion from three independent experiments relative to vehicle control (set at 100%) ± SD are presented. LOX-PP WT significantly suppressed invasion of MDA-MB-231 cells at 55.5 nM (P = 3.2 e^-6) and 111 nM (P = 0.0033), whereas LOX-PP Gln was not effective at 55.5 nM (P = 0.08) or 111 nM (P = 0.57) as determined by Student's t test. Similarly, LOX-PP WT (WT) significantly reduced invasion of Hs578T cells at 55.5 nM (P = 0.0046) and 111 nM (P = 0.0004) while LOX-PP Gln (Gln) was ineffective [55.5 nM (P = 0.8), 111 nM (P = 0.9)]. D. Hs578T cells were transfected with human pcDNA4-V5/His-LOX-PP WT or pcDNA4-V5/His-LOX-PP Gln or EV control for 24 h in DMEM-5% FBS. Cells were subjected to invasion assay for 16 h. Only expression of LOX-PP WT significantly reduced cell invasion (*, P = 9.5 e^-8) (LOX-PP Gln, P = 0.57).