healthmedicinet

Antibodies and reagents

Antibodies: to human integrin ?4 (mAB1964, clone 3E1; mAB clone 450-11A) and ?1 (mAB1951,
clone P4G11) subunits, Chemicon International, Inc. (Temecula, CA) of Millipore (Billerica,
MA), GIBCO BRL (Gaithersburg, MD), or BD Biosciences/Pharmingen; to human 12-LOX,
Oxford Biomedical Research (Oxford, MI); to ERK and phosphorylated ERK (T202/Y204),
Cell Signaling (Danvers, MA). Alexa₄₈₈ goat anti-rabbit IgG or Alexa₅₉₄ goat anti-mouse IgG were from Molecular Probes/Invitrogen (Eugene, OR). The anti-?4
mAbs 450-11A and 439-9B were provided by Dr. Steve Kennel (Oak Ridge National Laboratory,
Oak Ridge, TN).

Human laminin and EGF were from GIBCO BRL or Sigma Aldrich (St. Louis, MO). The 12-LOX-selective
inhibitor, BMD122, formerly called BHPP for N-benzyl-N-hydroxy-5-phenylpentanamide
45], was a generous gift from Biomide Corp. (Grosse Pointe Farms, MI). Other 12-LOX inhibitors:
Baicalein, Calbiochem (San Diego, CA); cinnamyl-3,4-dihydroxy-?-cyanocinnamate (CDC),
Biomol International, LP (Plymouth Meeting, PA). [³H]-12-HETE standard and [¹⁴C]-AA were from NEN Research Products (DuPont Company, Wilmington, DE). ODS-Silica
cartridges were from J.T. Baker Inc. (Phillipsburg, NJ). FuGENE 6 Transfection Reagent
kit was from Boehringer Mannheim (Santa Cruz, CA).

Cell culture and treatments

A431 and Chinese hamster ovary (CHO) cells were obtained from the American Type Culture
Collection (Manassas, VA), and cultured as recommended. Transfectants were selected
and cultured in media with 300 ?g/ml Geneticin (G418; Life Technologies, Inc., Grand
Island, NY). Prostate cancer cell lines PC-3 12-LOX/3.1 have been described previously
51].

For treatment with ?4 mAb (3E1), 5 × 10⁶ A431 cells were grown to sub-confluence in 100 mm Petri dishes and serum-starved
overnight prior to use. Cells were washed with PBS (3×) and stimulated with ?4 antibody
for 5, 15, 30, 60 and 90 min at a concentration of 5 ?g/ml in serum-free DMEM media.
For experiments where the natural ?6?4 ligand was immobilized, the dishes were coated
with laminin (10 ?g/ml) and cells were subsequently harvested as above. Otherwise,
laminin was used at 5 ?g/ml in serum-free media.

Subcellular fractionation

Cells (5×10⁶) were cultured to 80 % confluence [75-cm² flasks; 37 °C; 5 % CO₂ in DMEM containing 10 % (v/v) FBS], rinsed (2×) with PBS buffer, and washed (2×)
in isotonic buffer (134 mM NaCl; 15 mM Tris–HCl, pH 7.6; 5 mM glucose; 1 mM EDTA;
1 mM EGTA) before suspension in homogenization buffer (25 mM Tris–HCl, pH 7.6; 1 mM
EGTA) containing protease inhibitors. Cells were homogenized by sonication (15 sec,
3×, 0 °C) (Vibracell-Microtip) with intervals of 3 min. In some experiments, homogenates
were initially centrifuged at 10,000 × g (10 min; 4 °C) and the resultant supernatant
was considered cytosolic. The membrane fraction represents the pellet obtained after
a one-step centrifugation of the homogenate at 100,000 × g (1 h; 4 °C). The 10,000-
and 100,000 × g pellets, respectively, were rinsed once with homogenization buffer
and resuspended in protease inhibitor-free homogenization buffer. Samples, standardized
by protein concentration, were immediately used for SDS-PAGE.

Measurement of 12-lipoxygenase activity by LC-MS

12(S)-HETE was measured by liquid chromatography-mass spectrometry as previously described.
Cells (8×10⁵) were seeded into six well plates and serum-starved overnight. The following day
media was replaced with phenol red-free RPMI media. Cells were then stimulated with
3E1 in the presence of 10 ?m AA in 1 % fatty acid-free BSA. AA untreated cells served
as a control. As an additional control, AA was incubated in wells without cells to
measure spontaneous oxidation of AA into 12(S)-HETE, and this value was subtracted
from cell-generated 12(S)-HETE values. The detailed lipid extraction protocol has
been described 52]. For measurement of 12(S)-HETE production in parental A431 cells stimulated with
3E1 as a function of time, cells were incubated with 3E1 for the indicated times,
washed 1× with serum-free, phenol red-free media, and finally treated with 10 ?M AA
(in 1 % fatty acid-free BSA) for 15 min. Similarly, media from 12-LOX knock down (KD)
cell lines plus control cell lines were collected after 6 h incubation with AA alone,
or AA with 3E1 (added together for 6 h). 5 ?L of 15-HETE-d8 was added as an internal
standard to monitor extraction efficiency. Samples were clarified by centrifugation
at 1877 × g for 5 min. Supernatants were subjected to solid phase extraction using
Strata-X 33 ?m Polymeric Reversed Phase columns (30 mg/1 mL; Phenomenex, Torrance,
CA), followed by elution of lipid extracts with methanol, evaporation under a stream
of nitrogen, and reconstitution in 50 ?L LC-MS grade methanol. Ammonium acetate (50 ?L,
35 mM) was added before LC-MS analysis. Samples were analyzed as biological triplicates.

Immunoprecipitation

Cells were lysed in cold buffer (1 % Triton X-100; 150 mM NaCl; 10 mM Tris, pH 7.4;
1 mM EDTA; 1 mM EGTA, pH 8.0; 0.2 mM sodium ortho-vanadate; 0.2 mM PMSF; 0.01 % aprotinin;
5 ?g/ml leupeptin; 0.5 % NP-40), and subsequently clarified (10,000 × g; 10 min).
Supernatants were immunoprecipitated with 4–6 ?l of antibody against human 12-LOX,
anti-?1, or the anti-?4 subunit for 2 h, followed by 40 ?l Sepharose 4B-conjugated
protein G at 4 °C overnight. Immune complexes were washed (3×) in lysis buffer, and
used for SDS-PAGE. Whole cell lysates were used for input controls.

Western blotting

Performed as per standard techniques with horseradish peroxidase-conjugated secondary
anti-IgG diluted 1:4500, and enhanced chemiluminescence (ECL) (both: Amersham, Arlington
Heights, IL) for detection.

Expression constructs and transfection

Dr. Filippo Giancotti (Memorial Sloan-Kettering Cancer Center, NY) kindly provided
expression constructs encoding wild-type or mutant, truncated human ?4 subunits. These
were engineered in the eukaryotic expression vector pRC-CMV (Invitrogen Corp., San
Diego, CA) as described 53]; pRC-CMV-?4 (full-length ?4 subunit cDNA); pCMV-?4 ?854-1752 (tail-less, truncated
?4 lacking the cytoplasmic domain); and pCMV-?4 ?70-660 (headless, truncated ?4, extracellular
sequences replaced by a c-myc epitope tag). The full-length cDNA encoding human 12-LOX was subcloned into the EcoRI/XbaI
sites of pcDNA3.1 (Invitrogen) from pCMV-12-LOX, gifted from Dr. Colin D. Funk (Queen’s
University, Kingston, Ontario, CA). Expression vectors of full-length ?4 and 12-LOX
used neomycin as a selection marker. As deletion constructs of ?4 contained no selective
marker, they were cotransfected with neomycin encoding vector, pcDNA3.1. Cells grown
in 6-well plates were transfected with 3–12 ?g of pcDNA3.1, pCMV-?4, pcDNA-12-LOX,
pCMV-?4 tail-less or pCMV-?4 headless using the FuGENE 6 Transfection Reagent following
the manufacturer’s protocol. Neomycin-resistant cells were selected in 300 ?g/ml geneticin.
Knockdown of gene expression by shRNA was performed using Lentiviral pGIPZ constructs
targeted to unique regions of the 12-LOX gene, which were purchased from Open Biosystems
(Rockford, IL): V2LHS_112083 (#1), V2LHS_112086 (#2), V2LHS_112087 (#3), V3LHS_335849
(#5), V3LHS_335846 (#6), RHS4346 (#8). A431 stable transfections were achieved with
2 ?g plasmid DNA, using Lipofectamine LTX (Invitrogen), followed 48 h later by selection
in DMEM containing 1 ug/mL puromycin for 3 weeks (Invitrogen, Grand Island, NY). Authorization
to use lenti-based vectors for transfection was granted by the WSU Institutional Biosafety
Committee as protocol IBC 02-51-11.

DNA fragmentation assay

Cells (2.5×10⁶) were grown to sub-confluence in 10 cm tissue culture dishes and serum-starved (18 h)
prior to use. Cells were washed with PBS (3×), treated with varying BMD122 concentrations
for 24 h, and subsequently stimulated with 3E1 antibody for 5, 15, 30, 60 and 90 min.
For DNA isolation, cells from each time point were harvested and lysed with lysis
buffer (200 ?l) for 5 min, clarified at 500 × g (5 min), and the resulting pellet
was re-extracted using 200 ?l lysis buffer (2 min) and re-clarified. Supernatants
were pooled and treated with SDS (1 %) and DNase-free RNase (5 mg/ml) (Ambion, Austin,
TX) for 2 h (56 °C), followed by proteinase K (2.5 mg/ml) (Ambion, Austin, TX) treatment
for 2 h (37 °C). Finally, samples were extracted (1x) with alkaline phenol/chloroform/isoamyl
alcohol (25:24:1) and DNA was precipitated with 0.3 M sodium acetate (pH 5.2) and
ethanol. DNA laddering was assayed from equal numbers of cells, or 20 ?g resolved
on a 1.2 % agarose gel followed by ethidium bromide staining.

Migration assays

Modified Boyden chambers (Becton Dickinson, Bedford, MA) were coated with human laminin
(5 ?g/ml; 2 h; 25 °C) on the upper and lower surfaces, and seeded with A431 cells
(5×10⁵/ml) in DMEM-0.1 % BSA. Antibodies to ?4 integrin were preincubated with aliquots
of cells for 20 min prior to seeding. EGF (1 ng/ml) was added to the lower chamber
as a chemoattractant. The final concentration of 12-LOX pharmacological inhibitors
added to the lower chambers was: 10 ?M CDC or baicalein, or 20 ?M BMD122. All conditions
were tested in triplicate. After 3 h, inserts were fixed in a Quick-Fix solution,
double-stained with hematoxylin and eosin (HE), and mounted for observation and counting.
The number of migrated cells (12 fields × 100) was counted in a double-blind manner.
Alternately, inserts with 8 ?m pores (BD Falcon; Franklin Lakes, NJ) were coated with
100 ul of phenol red-free, basement membrane and matrix growth factor-reduced Matrigel
(BD Bioscience, Bedford, MA) (250 ug/ml; 1 h; 37 °C; excess liquid removed). Inserts
were seeded from confluent, serum-starved (overnight) A431 cells (5 × 10⁵) in 0.5 ml serum-free media. Where noted, cells were pre-treated with 25 ?M BMD122
for 1 h prior to 30 min treatments with the following: 3E1 antibody (3 ug/well) or
the natural ligand, laminin (10 ?g/ml). The lower chamber contained serum-free medium
with EGF (2 ng/ml), and complete media with or without serum served as positive and
negative controls, respectively. After 24 h, transmigrant cells on the underside of
the insert were stained with Azure AB/Eosin Y using the Diff Quick Stain Kit (IMEB,
Inc., San Marcos, CA) and washed twice with distilled water. After removing residual,
non-migrated cells, membranes were cut from the inserts, dissolved in 10 % acetic
acid and assayed for dye content at an absorbance of OD₅₇₀. Results are the mean of three samples.

Real-time PCR

Isolated RNA (2 ?g) (NucleoSpin RNAII kit; Macherey-Nagel, Bethlehem, PA) was reverse-transcribed
(High Capacity Reverse Transcription Kit; Applied Biosystems, Foster City, CA) for
real-time PCR (Taqman Gene Expression Master Mix, ALOX12 (HS00167524) and GAPDH primers;
Applied Biosystems, Foster City, CA). All sample reactions were run in triplicate
on the AB 7500 Fast Real Time PCR System. Relative expression of 12-LOX was quantified
by the Ct value measured against the internal standard GAPDH using the 7500 Fast System
SDS Software v1.4.0 (Applied Biosystems).