Introduction_NNP The_DT concept_NN of_IN lung_NN fibroblasts_NNS as_IN effector_NN cells_NNS in_IN the_DT pathogenesis_NNS of_IN idiopathic_JJ pulmonary_JJ fibrosis_NNS (_( IPF_NNP )_) has_VBZ recently_RB evolved_VBN [_NN 1_CD 2_CD ]_NN ._. 
Lung_NNP fibroblasts_NNS respond_VBP ,_, in_IN vitro_NN ,_, to_TO inflammatory_JJ cytokines_NNS by_IN producing_VBG growth_NN factors_NNS and_CC collagen_NN ,_, resulting_VBG in_IN fibroblast_NN proliferation_NN and_CC extracellular_NN matrix_NN deposition_NN [_NN 2_CD 3_CD 4_CD ]_NN ._. 
In_IN addition_NN ,_, activated_VBN lung_NN fibroblasts_NNS have_VBP been_VBN shown_VBN to_TO produce_VB large_JJ amounts_NNS of_IN inflammatory_JJ cytokines_NNS and_CC chemokines_NNS ,_, in_IN vitro_NN ,_, and_CC hence_RB ,_, these_DT cells_NNS may_MD also_RB have_VB a_DT role_NN as_IN effector_NN -_- inflammatory_JJ cells_NNS [_NN 1_CD 2_CD ]_NN ._. 
This_DT capacity_NN to_TO produce_VB both_DT inflammatory_JJ and_CC fibrotic_JJ factors_NNS could_MD mean_VB that_IN phenotypically_RB altered_VBN lung_NN fibroblasts_NNS act_VBP simultaneously_RB as_IN effector_NN and_CC target_NN cells_NNS ,_, via_IN paracrine_NN and_CC autocrine_NN mechanisms_NNS ,_, perpetuating_VBG the_DT fibrotic_JJ process_NN [_NN 2_CD ]_NN ._. 
Prostanoids_NNP are_VBP important_JJ regulators_NNS of_IN fibroblast_NN function_NN [_NN 5_CD 6_CD 7_CD 8_CD 9_CD ]_NN ._. 
Prostaglandin_NNP (_( PG_NNP )_) E_NN 2_CD is_VBZ thought_VBN to_TO have_VB antifibrotic_JJ properties_NNS in_IN vitro_NN ,_, but_CC also_RB can_MD have_VB proinflammatory_NN effects_NNS both_DT in_IN vivo_NN and_CC in_IN vitro_NN [_NN 10_CD 11_CD 12_CD ]_NN ._. 
Thromboxane_NNP (_( TX_NNP )_) A_DT 2_CD increases_NNS proliferation_NN ,_, and_CC DNA_NNP and_CC RNA_NNP synthesis_NN in_IN several_JJ cell_NN types_NNS ,_, including_VBG fibroblasts_NNS and_CC smooth_JJ muscle_NN like_IN glomerular_NN mesangial_NN cells_NNS [_NN 13_CD 14_CD 15_CD 16_CD ]_NN ._. 
Conversely_RB ,_, prostacyclin_NN (_( PGI_NNP 2_CD )_) decreases_NNS smooth_JJ muscle_NN cell_NN proliferation_NN and_CC collagen_NN synthesis_NN [_NN 17_CD 18_CD ]_NN ._. 
Many_JJ cell_NN types_NNS ,_, including_VBG lung_NN fibroblasts_NNS ,_, contain_VB cyclooxygenase_NN (_( COX_NNP )_) ,_, a_DT proximal_NN enzyme_NN in_IN prostanoid_NN production_NN ,_, and_CC can_MD generate_VB prostanoids_NNS [_NN 19_CD ]_NN ._. 
It_PRP has_VBZ been_VBN previously_RB reported_VBN that_IN IPF_NNP lung_NN fibroblasts_NNS have_VBP decreased_VBN COX-2_NNP expression_NN compared_VBN to_TO normal_JJ lung_NN fibroblasts_NNS and_CC ,_, hence_RB ,_, have_VBP decreased_VBN PGE_NNP 2_CD production_NN [_NN 12_CD 20_CD 21_CD ]_NN ._. 
Because_IN of_IN these_DT findings_NNS and_CC the_DT fact_NN that_IN PGs_NNP are_VBP important_JJ fibroblast_NN regulators_NNS ,_, we_PRP sought_VBD to_TO investigate_VB whether_IN abnormalities_NNS in_IN COX-2_NNP expression_NN could_MD be_VB associated_VBN with_IN an_DT altered_VBN balance_NN between_IN profibrotic_JJ and_CC antifibrotic_JJ PGs_NNP ._. 
We_PRP hypothesized_VBN that_IN fibroblasts_NNS from_IN the_DT lungs_NNS of_IN patients_NNS with_IN IPF_NNP (_( HF-IPF_NNP )_) have_VBP an_DT altered_VBN PG_NNP balance_NN compared_VBN to_TO normal_JJ lung_NN fibroblasts_NNS (_( HF-NL_NNP )_) ._. 
This_DT phenotypical_JJ abnormality_NN could_MD be_VB an_DT important_JJ factor_NN in_IN the_DT pathogenesis_NNS of_IN IPF_NNP ._. 

Materials_NNS and_CC methods_NNS Primary_JJ lung_NN fibroblasts_NNS 
Fibroblasts_NNP from_IN the_DT lungs_NNS of_IN seven_CD patients_NNS (_( 6_CD males_NNS )_) with_IN IPF_NNP (_( HF-IPF_NNP )_) were_VBD harvested_VBN :_: a_DT )_) from_IN excised_VBD lung_NN at_IN the_DT time_NN of_IN lung_NN transplantation_NN ;_: b_SYM )_) during_IN an_DT autopsy_NN performed_VBN within_IN 4_CD hours_NNS from_IN death_NN ;_: or_CC c_SYM )_) during_IN open_NN or_CC transbronchial_NN lung_NN biopsies_NNS at_IN the_DT time_NN of_IN diagnosis_NN ._. 
Of_IN the_DT seven_CD patients_NNS ,_, five_CD subjects_NNS had_VBD advanced_VBN lung_NN fibrosis_NNS and_CC were_VBD receiving_VBG prednisone_NN ±_NN immunosuppressive_JJ agents_NNS ;_: 2_CD patients_NNS were_VBD at_IN an_DT earlier_JJR stage_NN of_IN their_PRP$ disease_NN and_CC were_VBD not_RB receiving_VBG immunosuppressive_JJ drugs_NNS ._. 
The_DT mean_JJ age_NN of_IN the_DT patients_NNS was_VBD 59_CD (_( range_NN 43_CD -_- 71_CD )_) ]_NN ._. 
HF-NL_NNP were_VBD cultured_JJ from_IN five_CD human_JJ lungs_NNS that_WDT arrived_VBD at_IN our_PRP$ transplant_NN center_NN with_IN the_DT intention_NN of_IN being_VBG used_VBN for_IN transplantation_NN ,_, but_CC for_IN various_JJ reasons_NNS could_MD not_RB be_VB transplanted_VBN ;_: these_DT were_VBD macroscopically_RB and_CC microscopically_RB normal_JJ ._. 
The_DT cells_NNS were_VBD harvested_VBN and_CC cultured_JJ as_IN per_IN the_DT protocol_NN described_VBN by_IN Kumar_NNP et_CC al._NN [_NN 22_CD ]_NN ._. 
Briefly_NNP ,_, lung_NN tissue_NN sections_NNS were_VBD finely_RB cut_VB with_IN sterile_JJ scissors_NNS and_CC incubated_JJ with_IN serum_NN free_JJ DMEM_NNP containing_VBG trypsin_NN ,_, DNAse_NNP and_CC collagenase_NN for_IN 30_CD min._NN 
The_DT procedure_NN was_VBD repeated_VBN twice_RB ,_, and_CC the_DT supernatants_NNS were_VBD pooled_VBN and_CC cultured_JJ in_IN one_CD 100_CD mm_NN plate_NN and_CC incubated_JJ at_IN 37_CD °_NN C_NNP in_IN a_DT 5_CD %_NN CO_NNP 2_CD humidified_JJ atmosphere_NN ._. 
Culture_NNP medium_NN (_( DMEM_NNP with_IN 5_CD %_NN fetal_JJ bovine_JJ serum_NN [_NN FBS_NNP ]_NN and_CC penicillin_NN /_NN streptomycin_NN )_) was_VBD replaced_VBN three_CD times_NNS per_IN week_NN and_CC fibroblasts_NNS were_VBD passed_VBN (_( 1_CD :_: 2_CD split_NN )_) at_IN the_DT time_NN they_PRP became_VBD confluent_NN ._. 
On_IN passage_NN 4_CD the_DT cells_NNS were_VBD resuspended_JJ in_IN 1_CD ml_NN of_IN DMEM_NNP with_IN 20_CD %_NN FBS_NNP and_CC DMSO_NNP and_CC frozen_VBN at_IN -_: 70_CD °_NN C._NN 
For_IN each_DT experiment_NN described_VBD below_IN the_DT cells_NNS were_VBD thawed_JJ ,_, cultured_JJ and_CC passed_VBN at_IN least_JJS once_RB ._. 
All_DT the_DT experiments_NNS were_VBD conducted_VBN with_IN cells_NNS at_IN passages_NNS 6_CD to_TO 8_CD ._. 

Inducible_NNP cyclooxygenase_NN (COX)-2_NNP expression_NN and_CC eicosanoid_NN production_NN COX-2_NNP activity_NN was_VBD determined_VBN by_IN measuring_VBG PGE_NNP 2_CD ,_, 6-keto-PGF_NN 1_CD α_NN (_( stable_JJ PGI_NNP 2_CD metabolite_NN )_) ,_, TXB_NNP 2_CD (_( stable_JJ TXA_NNP 2_CD metabolite_NN )_) ,_, and_CC PGF_NNP 2_CD α_NN production_NN in_IN stimulated_VBN fibroblasts_NNS ._. 
HF-IPF_NNP (_( n_NN =_SYM 7_CD )_) and_CC HF-NL_NNP (_( n_NN =_SYM 5_CD )_) were_VBD brought_VBN to_TO >_NN 90_CD %_NN confluency_NN in_IN 100_CD mm_NN plates_NNS and_CC then_RB placed_VBN on_IN serum_NN free_JJ DMEM_NNP for_IN 24_CD hours_NNS to_TO render_VB them_PRP quiescent_JJ ._. 
Fibroblasts_NNP were_VBD then_RB incubated_JJ in_IN DMEM_NNP with_IN 5_CD %_NN FBS_NNP alone_RB or_CC in_IN the_DT same_JJ medium_NN with_IN IL-1β_NNP (_( 2.5_CD ng_NN /_NN ml_NN )_) for_IN 24_CD hours_NNS ._. 
At_IN the_DT end_NN of_IN the_DT incubation_NN period_NN the_DT supernatant_NN was_VBD aspirated_JJ and_CC fresh_JJ media_NNS containing_VBG 30_CD μM_NN of_IN arachidonic_JJ acid_NN was_VBD added_VBN to_TO the_DT plates_NNS ._. 
After_IN 30_CD min_NN of_IN incubation_NN the_DT supernatant_NN was_VBD collected_VBN and_CC saved_VBN at_IN -_: 70_CD °_NN C_NNP for_IN later_RB eicosanoid_NN analysis_NN ._. 
The_DT cells_NNS were_VBD then_RB resuspended_JJ and_CC divided_VBN into_IN two_CD aliquots_NNS ,_, which_WDT were_VBD used_VBN for_IN RNA_NNP and_CC protein_NN extractions_NNS ,_, respectively_RB ._. 
The_DT above_IN experiments_NNS were_VBD repeated_VBN in_IN HF-IPF_NNP (_( n_NN =_SYM 2_CD )_) and_CC HF-NL_NNP (_( n_NN =_SYM 2_CD )_) using_VBG serum_NN free_JJ media_NNS conditions_NNS ._. 
Prostanoids_NNP were_VBD measured_VBN by_IN modified_VBN stable_JJ isotope_NN dilution_NN assays_NNS that_WDT used_VBD gas_NN chromatography_NN -_- negative_JJ ion_NN -_- chemical_JJ ionization_NN mass_NN spectrometry_NN as_RB previously_RB described_VBD [_NN 23_CD ]_NN ._. 
Briefly_NNP ,_, deuterium_NN -_- labeled_JJ internal_JJ standards_NNS of_IN PGE_NNP 2_CD ,_, PGF_NNP 2_CD α_NN ,_, TXB_NNP 2_CD ,_, and_CC 6-keto-PGF_NN 1_CD α_NN were_VBD added_VBN to_TO the_DT supernatants_NNS with_IN isopropyl_NN alcohol_NN ._. 
Isopropyl_NNP alcohol_NN was_VBD removed_VBN by_IN evaporation_NN under_IN nitrogen_NN ._. 
After_IN acidification_NN to_TO pH_NN 3.5_CD ,_, the_DT samples_NNS were_VBD extracted_VBN on_IN preconditioned_JJ C-18_NNP PrepSep_NNP columns_NNS (_( Fisher_NNP Scientific_NNP ,_, Fair_NNP Lawn_NNP ,_, NJ_NNP )_) ,_, and_CC eluted_JJ with_IN ethyl_NN acetate_NN ._. 
The_DT extract_NN was_VBD then_RB converted_VBN to_TO a_DT pentafluorobenzyl_NN ester_NN by_IN treatment_NN with_IN a_DT mixture_NN of_IN 12.5_CD %_NN pentafluorobenzyl_NN bromide_NN in_IN acetonitrile_NN and_CC disopropylethylamine_NN at_IN room_NN temperature_NN for_IN 30_CD min._NN 
After_IN evaporation_NN of_IN reagents_NNS ,_, the_DT residue_NN was_VBD subjected_VBN to_TO TLC_NNP plates_NNS ,_, using_VBG the_DT solvent_JJ system_NN chloroform_NN /_NN ethanol_NN (_( 93_CD :_: 7_CD ,_, vol_NN /_NN vol_NN )_) for_IN PGF_NNP 2_CD α_NN and_CC TXB_NNP 2_CD ,_, and_CC ethyl_NN acetate_NN /_NN methanol_NN (_( 93_CD :_: 2_CD ,_, vol_NN /_NN vol_NN )_) for_IN 6-keto-PGF_NN 1_CD α_NN and_CC PGE_NNP 2_CD ._. 
Then_RB PGF_NNP 2_CD α_NN was_VBD converted_VBN to_TO trimethylsilyl_NN ether_NN derivative_JJ by_IN treatment_NN with_IN N,O-bis_NNP (_( trimethylsilyl_NN )_) trifluoroacetamide_NN and_CC dimethylformamide_NN ._. 
The_DT methoxime_NN derivative_NN of_IN TXB_NNP 2_CD ,_, PGE_NNP 2_CD and_CC 6-keto-PGF_NN 1_CD α_NN was_VBD made_VBN by_IN treatment_NN with_IN 2_CD %_NN methoxamine_NN hydrochloride_NN in_IN pyridine_NN at_IN 70_CD °_NN C_NNP for_IN 60_CD min_NN ,_, followed_VBN by_IN evaporation_NN of_IN the_DT pyridine_NN ,_, addition_NN of_IN water_NN ,_, and_CC extraction_NN with_IN ethyl_NN acetate_NN ._. 
Derivatization_NNP was_VBD completed_VBN by_IN formation_NN of_IN the_DT trimethylsilyl_NN derivatives_NNS by_IN treatment_NN with_IN N,O-bis_NNP (_( trimethylsilyl_NN )_) trifluoroacetamide_NN and_CC pyridine_NN ._. 
Eicosanoids_NNP were_VBD quantified_VBN by_IN measuring_VBG the_DT ratio_NN of_IN the_DT intensity_NN of_IN ions_NNS m_NN /_NN z_SYM 569/573_CD for_IN PGF_NNP 2_CD α_NN ,_, m_NN /_NN z_SYM 614/618_CD for_IN TXB_NNP 2_CD and_CC 6-keto-PGF_NN 1_CD α_NN ,_, and_CC m_NN /_NN z_SYM 524/528_CD for_IN PGE_NNP 2_CD ._. 
An_DT analytical_JJ blank_NN for_IN each_DT of_IN these_DT products_NNS was_VBD determined_VBN by_IN measuring_VBG the_DT amount_NN of_IN nondeuterated_JJ material_NN ,_, detected_VBD after_IN extracting_VBG and_CC analyzing_VBG 2_CD ml_NN of_IN saline_NN to_TO which_WDT the_DT deuterium-labeled_JJ internal_JJ standards_NNS had_VBD been_VBN added_VBN ._. 

Western_JJ analysis_NN After_IN washing_VBG with_IN PBS_NNP at_IN pH_NN 7.4_CD ,_, pellets_NNS were_VBD lyzed_JJ in_IN solubilization_NN buffer_NN containing_VBG 50_CD mM_NN TRIS_NNP at_IN pH_NN 8_CD ,_, 1_CD %_NN Tween_NNP 20_CD ,_, 10_CD mM_NN phenylmethylsulphonyl_NN fluoride_NN ,_, diethyldithiocarbamic_JJ acid_NN ,_, leupeptin_NN and_CC pepstatin_NN A_DT (_( all_DT from_IN Sigma_NNP Chemical_NNP )_) ,_, sonicated_JJ ,_, boiled_VBD with_IN gel_NN loading_NN buffer_NN (_( 62.5_CD mM_NN TRIS-HCl_NNP ,_, at_IN pH_NN 6.8_CD ,_, 10_CD %_NN glycerol_NN ,_, 2_CD %_NN SDS_NNP ,_, 5_CD %_NN β-mercaptoethanol_JJ ,_, and_CC bromophenol_NN blue_JJ )_) ,_, and_CC centrifuged_JJ at_IN 15,000_CD x_SYM g_SYM for_IN 10_CD min._NN 
Equal_NNP amounts_NNS of_IN protein_NN (_( 70_CD to_TO 100_CD μg_NN )_) were_VBD separated_VBN by_IN electrophoresis_NNS ._. 
SDS-PAGE_NNP was_VBD performed_VBN using_VBG a_DT 7.5_CD %_NN separating_VBG gel_NN with_IN a_DT 4_CD %_NN stacking_VBG gel_NN ._. 
The_DT resolved_VBN proteins_NNS were_VBD transferred_VBN electrophoretically_RB to_TO nitrocellulose_NN membranes_NNS (_( Hybond-_NNP ECL_NNP ,_, Amersham_NNP Corp_NNP ._. )_) ._. 
After_IN transfer_NN ,_, the_DT filters_NNS were_VBD incubated_JJ overnight_JJ at_IN 4_CD °_NN C_NNP in_IN a_DT blocking_VBG solution_NN (_( 20_CD mM_NN TRIS_NNP base_NN ,_, 137_CD mM_NN sodium_NN chloride_NN at_IN pH_NN 7.6_CD ,_, 5_CD %_NN powdered_JJ milk_NN ,_, 3_CD %_NN BSA_NNP )_) ,_, and_CC incubated_JJ with_IN primary_JJ polyclonal_NN rabbit_NN antibodies_NNS against_IN COX-2_NNP at_IN a_DT dilution_NN 1_CD :_: 1000_CD (_( Cayman_NNP Chemical_NNP ,_, Ann_NNP Arbor_NNP ,_, MI_NNP )_) ,_, for_IN 1_CD hour_NN at_IN room_NN temperature_NN ._. 
The_DT filters_NNS were_VBD washed_VBN (_( TBS_NNP -_- 0.1_CD %_NN Tween_NNP 20_CD at_IN pH_NN 7.6_CD )_) and_CC incubated_JJ with_IN horseradish_NN peroxidase_NN linked_VBD secondary_JJ antibodies_NNS at_IN a_DT dilution_NN 1_CD :_: 4000_CD (_( Amersham_NNP )_) ._. 
After_IN washing_VBG ,_, the_DT membranes_NNS were_VBD incubated_JJ with_IN luminol_NN based_VBD chemiluminescence_NN reagent_NN (_( DuPont_NNP NEN_NNP Research_NNP Products_NNPS ,_, Boston_NNP ,_, MA_NNP )_) ._. 

Northern_JJ analysis_NN Cell_NNP pellets_NNS were_VBD lyzed_JJ and_CC RNA_NNP extracted_VBD using_VBG the_DT RNeasy_NNP method_NN ®_NN (_( Qiagen_NNP )_) ,_, following_VBG the_DT manufacturer_NN 's_POS instructions_NNS ._. 
RNA_NNP was_VBD quantified_VBN by_IN determining_VBG light_JJ absorbance_NN at_IN 260_CD nm_NN and_CC then_RB fractioned_JJ by_IN electrophoresis_NNS (_( 10_CD μg_NN per_IN lane_NN )_) on_IN a_DT 1_CD %_NN agarose_NN MOPS_NNP /_NN formaldehyde_NN gel_NN ._. 
The_DT RNA_NNP was_VBD denatured_JJ prior_RB to_TO loading_NN by_IN incubating_VBG the_DT RNA_NNP at_IN 65_CD °_NN C_NNP for_IN 10_CD min_NN in_IN a_DT loading_NN buffer_NN comprising_VBG 1_CD X_NNP MOPS_NNP ,_, 50_CD %_NN formamide_NN ,_, 6.5_CD %_NN formaldehyde_NN ,_, 5_CD %_NN glycerol_NN ,_, 0.1_CD mM_NN EDTA_NNP ,_, 0.025_CD %_NN bromophenol_NN blue_JJ ,_, 0.025_CD %_NN xylene_NN cyanol_NN ._. 
The_DT RNA_NNP was_VBD transferred_VBN by_IN gravity_NN -_- assisted_VBN capillary_JJ method_NN with_IN 6_CD X_NNP SSC_NNP to_TO nylon_NN hybridization_NN membrane_NN ,_, and_CC then_RB fixed_VBN to_TO the_DT membrane_NN by_IN UV_NNP crosslinking_VBG (_( Stratalinker_NNP 1200_CD μj_NN /_NN cm_NN 2_CD )_) ._. 
Prehybridization_NNP and_CC hybridization_NN were_VBD performed_VBN at_IN 42_CD °_NN C_NNP and_CC using_VBG Quick_NNP Hyb_NNP ®_NN (_( Stratagene_NNP )_) as_IN hybridization_NN solution_NN ._. 
The_DT COX-2_NNP probe_NN was_VBD random_JJ primed_VBN following_VBG the_DT directions_NNS of_IN the_DT manufacturer_NN (_( Megaprime_NNP ®_NN ,_, Amersham_NNP /_NN Pharmacia_NNP )_) ._. 
The_DT membrane_NN was_VBD then_RB washed_VBN at_IN a_DT final_JJ stringency_NN of_IN 0.2_CD X_NNP SSC_NNP ,_, 0.1_CD %_NN SDS_NNP ,_, at_IN 60_CD °_NN C_NNP for_IN 30_CD min._NN 
The_DT membrane_NN was_VBD wrapped_VBN in_IN plastic_NN wrap_NN and_CC exposed_VBN to_TO Kodak_NNP XR_NNP film_NN at_IN -_: 70_CD °_NN C_NNP with_IN intensifier_NN screen_NN overnight_JJ ._. 

Statistical_NNP methods_NNS All_DT results_NNS are_VBP presented_VBN as_IN medians_NNS with_IN their_PRP$ range_NN ._. 
Comparisons_NNP between_IN HF-IPF_NNP and_CC HF-NL_NNP were_VBD done_VBN using_VBG the_DT Mann_NNP -_- Whitney_NNP test_NN ._. 
A_DT P_NN value_NN of_IN <_NN 0.05_CD 
was_VBD considered_VBN significant_JJ ._. 

Results_NNS Baseline_NNP and_CC stimulated_VBN COX-2_NNP activity_NN in_IN HF-IPF_NNP and_CC HF-NL_NNP Unstimulated_NNP eicosanoid_NN production_NN was_VBD similar_JJ in_IN both_DT HF-IPF_NNP and_CC HF-NL_NNP (_( Fig._NN 1_CD ,_, a-d_JJ )_) ._. 
When_WRB fibroblasts_NNS were_VBD stimulated_VBN with_IN IL-1β_NNP there_EX was_VBD a_DT significant_JJ and_CC similar_JJ upregulation_NN of_IN PGE_NNP 2_CD production_NN in_IN both_DT HF-IPF_NNP and_CC HF-NL_NNP (_( 28.35_CD [_NN range_NN :_: 9.09_CD -_: 89.09_CD ]_NN versus_CC 17.12_CD [_NN 8.58_CD -_: 29.33_CD ]_NN ng_NN /_NN 10_CD 6_CD cells_NNS /_NN 30_CD min_NN ,_, respectively_RB ;_: P_NN =_SYM 0.25_CD ;_: [_NN Fig._NN 1_CD a_DT ]_NN )_) ._. 
IL-1β-stimulated_NNP production_NN of_IN TXB_NNP 2_CD (_( stable_JJ metabolite_NN of_IN the_DT active_JJ TXA_NNP 
2_CD )_) ,_, PGF_NNP 2_CD α_NN ,_, and_CC 6-keto-PGF_NN 1_CD α_NN (_( stable_JJ metabolite_NN of_IN PGI_NNP 2_CD )_) increased_VBN modestly_RB in_IN every_DT case_NN ,_, except_IN TXB_NNP 2_CD production_NN by_IN HF-NL_NNP ,_, which_WDT decreased_VBD (_( 0.75_CD [_NN 0.15_CD -_: 2.58_CD ]_NN ng_NN /_NN 10_CD 6_CD cells_NNS /_NN 30_CD min_NN at_IN baseline_NN versus_CC 0.61_CD [_NN 0.21_CD -_: 1.64_CD ]_NN ng_NN /_NN 10_CD 6_CD cells_NNS /_NN 30_CD min_NN with_IN IL-1β_NNP stimulation_NN )_) (_( Fig._NN 1_CD b_SYM )_) ._. 
Results_NNS of_IN PGE_NNP 2_CD production_NN were_VBD similar_JJ when_WRB experiments_NNS were_VBD conducted_VBN in_IN serum_NN free_JJ media_NNS conditions_NNS (_( results_NNS not_RB shown_VBN )_) ._. 
IL-1β_NNP stimulated_VBN TXB_NNP 2_CD production_NN was_VBD significantly_RB greater_JJR in_IN HF-IPF_NNP (_( 1.92_CD [_NN 1.27_CD -_: 2.57_CD ]_NN ng_NN /_NN 10_CD 6_CD cells_NNS /_NN 30_CD min_NN )_) than_IN in_IN HF-NL_NNP (_( 0.61_CD [_NN 0.21_CD -_: 1.64_CD ]_NN ng_NN /_NN 10_CD 6_CD cells_NNS /_NN 30_CD min_NN ;_: P_NN =_SYM 0.007_CD )_) (_( Fig._NN 1_CD b_SYM )_) ;_: baseline_NN TXB_NNP 2_CD production_NN was_VBD not_RB significantly_RB different_JJ between_IN the_DT two_CD cell_NN groups_NNS (_( 1.73_CD [_NN 0.77_CD -_: 2.53_CD ]_NN versus_CC 0.75_CD [_NN 0.15_CD -_: 2.58_CD ]_NN ng_NN /_NN 10_CD 6_CD cells_NNS /_NN 30_CD min_NN ,_, in_IN HF-IPF_NNP and_CC HF-NL_NNP ,_, respectively_RB ;_: P_NN =_SYM 0.17_CD [_NN Fig._NN 1_CD b_SYM ]_NN )_) ._. 
Because_IN PGI_NNP 2_CD and_CC TXA_NNP 2_CD have_VBP opposing_VBG effects_NNS in_IN vivo_NN ,_, we_PRP calculated_VBD the_DT ratio_NN of_IN their_PRP$ metabolites_NNS (_( 6-keto-PGF_NN 1_CD α_NN :_: TXB_NNP 2_CD )_) and_CC found_VBD a_DT significantly_RB lower_JJR ratio_NN in_IN HF-IPF_NNP at_IN baseline_NN (_( 0.08_CD [_NN 0.04_CD -_: 0.52_CD ]_NN versus_CC 0.12_CD [_NN 0.11_CD -_: 0.89_CD ]_NN in_IN HF-IPF_NNP and_CC HF-NL_NNP ,_, respectively_RB ;_: P_NN =_SYM 0.028_CD )_) and_CC a_DT similar_JJ trend_NN under_IN stimulated_VBN conditions_NNS (_( 0.24_CD [_NN 0.05_CD -_: 1.53_CD ]_NN versus_CC 1.08_CD [_NN 0.51_CD -_: 3.79_CD ]_NN in_IN HF-IPF_NNP and_CC HF-NL_NNP ,_, respectively_RB ;_: P_NN =_SYM 0.09_CD [_NN Fig._NN 2_CD ]_NN )_) ._. 

Baseline_NNP and_CC stimulated_VBN COX-2_NNP expression_NN Western_JJ blot_NN in_IN unstimulated_JJ fibroblasts_NNS showed_VBD no_DT detectable_JJ COX-2_NNP protein_NN in_IN either_DT group_NN of_IN cells_NNS ,_, while_IN IL-1β_NNP significantly_RB induced_VBD COX-2_NNP to_TO a_DT similar_JJ degree_NN in_IN IPF_NNP and_CC normal_JJ lung_NN fibroblasts_NNS (_( Fig._NN 3_CD )_) ._. 
Northern_NNP blot_NN showed_VBD minimal_JJ COX-2_NNP mRNA_NN in_IN unstimulated_JJ cells_NNS and_CC significant_JJ upregulation_NN of_IN COX-2_NNP mRNA_NN expression_NN when_WRB stimulated_VBN with_IN IL-1β_NNP in_IN both_DT HF-IPF_NNP and_CC HF-NL_NNP (_( Fig._NN 
4_CD )_) ._. 

Discussion_NNP Several_JJ factors_NNS modulate_VBP fibroblast_NN proliferation_NN and_CC collagen_NN production_NN ,_, including_VBG mitogenic_JJ cytokines_NNS (_( e.g._NN ,_, transforming_VBG growth_NN factor_NN β_NN [_NN TGFβ_NNP ]_NN ,_, platelet_NN -_- derived_VBN growth_NN factor_NN [_NN PDGF_NNP ]_NN ,_, basic_JJ fibroblast_NN growth_NN factor_NN [_NN bFGF_NN ]_NN )_) ,_, eicosanoids_NNS (_( i.e._NN ,_, PGE_NNP 2_CD ,_, TXB_NNP 2_CD ,_, and_CC PGI_NNP 2_CD )_) ,_, and_CC antifibrogenic_JJ cytokines_NNS (_( e.g._NN IFN-γ_NNP )_) [_NN 1_CD 2_CD 3_CD ]_NN ._. 
It_PRP is_VBZ very_RB likely_JJ that_IN a_DT complex_JJ interaction_NN among_IN these_DT factors_NNS exists_VBZ in_IN the_DT tissue_NN repair_NN process_NN ,_, and_CC it_PRP is_VBZ possible_JJ that_IN pathologic_JJ fibrosis_NNS ,_, as_IN in_IN IPF_NNP ,_, results_VBZ from_IN phenotypical_JJ alterations_NNS in_IN fibroblasts_NNS that_WDT affect_VBP the_DT "_'' normal_JJ "_'' interaction_NN of_IN these_DT factors_NNS ._. 
Our_PRP$ results_NNS show_VBP that_DT stimulation_NN of_IN primary_JJ cultures_NNS of_IN human_JJ lung_NN fibroblasts_NNS with_IN the_DT proximal_NN cytokine_NN IL-1β_NNP upregulates_NNS COX-2_NNP protein_NN and_CC mRNA_NN expression_NN to_TO a_DT similar_JJ degree_NN in_IN normal_JJ and_CC IPF_NNP fibroblasts_NNS ._. 
TXA_NNP 2_CD production_NN tended_VBD to_TO be_VB greater_JJR in_IN IPF_NNP than_IN in_IN normal_JJ fibroblasts_NNS at_IN baseline_NN ;_: when_WRB stimulated_VBN with_IN IL-1β_NNP this_DT difference_NN became_VBD statistically_RB significant_JJ ._. 
The_DT ratio_NN of_IN PGI_NNP 2_CD to_TO TXA_NNP 2_CD metabolites_NNS was_VBD lower_JJR in_IN IPF_NNP fibroblasts_NNS at_IN baseline_NN and_CC with_IN IL-1β_NNP stimulation_NN ._. 
The_DT above_IN results_NNS suggest_VBP that_IN a_DT decreased_VBN PGI_NNP 2_CD :_: TXA_NNP 2_CD ratio_NN could_MD be_VB a_DT phenotypic_JJ alteration_NN present_JJ in_IN IPF_NNP fibroblasts_NNS ,_, resulting_VBG in_IN a_DT loss_NN of_IN their_PRP$ capacity_NN to_TO autoregulate_NN proliferation_NN and_CC extracellular_NN matrix_NN production_NN ._. 
The_DT effects_NNS of_IN PGs_NNP on_IN cell_NN proliferation_NN and_CC collagen_NN production_NN have_VBP been_VBN widely_RB studied_VBN in_IN different_JJ cell_NN types_NNS [_NN 13_CD 14_CD 15_CD 16_CD 17_CD 26_CD ]_NN ._. 
TXA_NNP 2_CD has_VBZ been_VBN studied_VBN extensively_RB because_IN of_IN its_PRP$ apparent_JJ role_NN in_IN atherosclerosis_NNS ,_, due_JJ to_TO its_PRP$ prothrombotic_JJ and_CC mitogenic_JJ activities_NNS on_IN vascular_NN smooth_JJ muscle_NN cells_NNS [_NN 15_CD 16_CD ]_NN ._. 
These_DT mitogenic_JJ effects_NNS are_VBP potentiated_JJ by_IN growth_NN factors_NNS [_NN 15_CD 16_CD 27_CD 28_CD ]_NN ._. 
In_IN vascular_NN smooth_JJ muscle_NN cells_NNS TXA_NNP 2_CD stimulates_NNS synthesis_NN of_IN bFGF_NN and_CC increases_VBZ the_DT expression_NN of_IN the_DT proto_JJ -_- oncogenes_NNS c-fos_JJ ,_, c-myc_JJ ,_, and_CC egr-1_NN ,_, which_WDT are_VBP associated_VBN with_IN entry_NN into_IN the_DT cell_NN growth_NN cycle_NN [_NN 15_CD ]_NN ._. 
In_IN addition_NN ,_, TXA_NNP 2_CD increases_NNS proliferation_NN of_IN fibroblasts_NNS [_NN 13_CD ]_NN and_CC smooth_JJ muscle_NN -_- like_IN glomerular_NN mesangial_NN cells_NNS [_NN 14_CD ]_NN ._. 
On_IN the_DT other_JJ hand_NN ,_, PGI_NNP 2_CD decreases_NNS vascular_NN smooth_JJ muscle_NN cell_NN proliferation_NN and_CC collagen_NN and_CC glycosaminoglycane_NN synthesis_NN ,_, via_IN activation_NN of_IN adenylyl_NN cyclase_NN and_CC subsequent_JJ production_NN of_IN cAMP_NN [_NN 17_CD ]_NN ._. 
Betaprost_NNP ,_, an_DT analog_NN of_IN PGI_NNP 2_CD ,_, decreases_NNS procollagen_NN I_PRP and_CC III_NNP mRNA_NN expression_NN in_IN cardiac_JJ fibroblasts_NNS [_NN 18_CD ]_NN ._. 
These_DT effects_NNS may_MD counteract_VB the_DT profibrotic_JJ effects_NNS seen_VBN with_IN TXA_NNP 2_CD and_CC it_PRP is_VBZ possible_JJ that_IN an_DT alteration_NN of_IN a_DT "_'' normal_JJ "_'' physiologic_JJ balance_NN between_IN PGI_NNP 2_CD and_CC TXA_NNP 2_CD could_MD increase_VB tendency_NN towards_IN fibrogenesis_NNS ._. 
It_PRP is_VBZ important_JJ to_TO mention_VB that_IN our_PRP$ experiments_NNS were_VBD conducted_VBN at_IN similar_JJ passage_NN levels_NNS (_( passage_NN 6_CD to_TO 8_CD )_) in_IN both_DT groups_NNS ,_, since_IN senescence_NN of_IN fibroblasts_NNS is_VBZ associated_VBN with_IN a_DT shift_NN from_IN the_DT biosynthesis_NNS of_IN PGI_NNP 2_CD to_TO TXA_NNP 2_CD [_NN 24_CD 25_CD ]_NN ._. 
It_PRP is_VBZ possible_JJ that_IN the_DT difference_NN seen_VBN in_IN our_PRP$ study_NN between_IN HF-IPF_NNP and_CC HF-NL_NNP could_MD result_VB from_IN comparing_VBG fibroblasts_NNS of_IN different_JJ ages_NNS ._. 
HF-IPF_NNP might_MD have_VB been_VBN harvested_VBN from_IN fibrotic_JJ lesions_NNS where_WRB fibroblasts_NNS had_VBD previously_RB undergone_VBN a_DT greater_JJR number_NN of_IN cell_NN divisions_NNS than_IN HF-NL_NNP ,_, obtained_VBN from_IN nonfibrotic_JJ lungs_NNS ._. 
Although_IN this_DT is_VBZ a_DT possibility_NN ,_, the_DT age_NN -_- related_VBN shift_NN in_IN PG_NNP production_NN has_VBZ only_RB been_VBN shown_VBN at_IN very_RB high_JJ cell_NN passages_NNS and_CC has_VBZ not_RB been_VBN documented_VBN in_IN vivo_NN ._. 
We_PRP also_RB found_VBD that_IN both_DT HF-IPF_NNP and_CC HF-NL_NNP had_VBD similar_JJ PGE_NNP 
2_CD production_NN at_IN baseline_NN ,_, and_CC a_DT similar_JJ increase_NN when_WRB stimulated_VBN with_IN IL-1β_NNP ._. 
PGE_NNP 2_CD can_MD decrease_VB fibroblast_NN proliferation_NN and_CC collagen_NN synthesis_NN ,_, and_CC increase_NN collagen_NN degradation_NN [_NN 5_CD 6_CD 7_CD 8_CD ]_NN ._. 
Recent_JJ reports_NNS suggesting_VBG decreased_VBN COX-2_NNP expression_NN and_CC PGE_NNP 2_CD production_NN in_IN IPF_NNP fibroblasts_NNS have_VBP received_VBN significant_JJ attention_NN [_NN 12_CD 20_CD 21_CD ]_NN ._. 
In_IN our_PRP$ study_NN we_PRP found_VBD that_IN both_DT COX-2_NNP protein_NN expression_NN and_CC PGE_NNP 2_CD production_NN were_VBD upregulated_JJ to_TO a_DT similar_JJ degree_NN in_IN IPF_NNP and_CC normal_JJ lung_NN fibroblasts_NNS ._. 
We_PRP believe_VBP that_IN differences_NNS in_IN methodology_NN and_CC patient_NN selection_NN may_MD explain_VB the_DT discrepancies_NNS with_IN other_JJ studies_NNS ._. 
Vancheri_NNP and_CC collaborators_NNS [_NN 20_CD ]_NN found_VBD that_IN TNF-α-stimulated_NNP fibrotic_JJ lung_NN fibroblasts_NNS had_VBD decreased_VBN COX-2_NNP expression_NN and_CC PGE_NNP 2_CD production_NN ,_, but_CC they_PRP further_RBR showed_VBD that_IN these_DT findings_NNS were_VBD a_DT result_NN of_IN decreased_VBD expression_NN of_IN TNF-α_NNP receptors_NNS ._. 
The_DT latter_JJ finding_NN would_MD argue_VB against_IN a_DT primary_JJ defect_NN in_IN COX-2_NNP expression_NN ,_, since_IN no_DT other_JJ stimulus_NN ,_, other_JJ than_IN TNF-α_NNP ,_, was_VBD tested_VBN ._. 
In_IN another_DT study_NN ,_, Keerthisingam_NNP et_CC al._NN [_NN 21_CD ]_NN reported_VBD that_IN fibrotic_JJ lung_NN fibroblasts_NNS had_VBD decreased_VBN COX-2_NNP expression_NN and_CC PGE_NNP 2_CD production_NN in_IN response_NN to_TO TGFβ_NNP stimulation_NN ._. 
This_DT study_NN differed_VBD from_IN ours_PRP in_IN that_IN a_DT different_JJ stimulus_NN was_VBD used_VBN ._. 
Of_IN significance_NN is_VBZ the_DT fact_NN that_IN the_DT COX-2_NNP gene_NN is_VBZ known_VBN to_TO be_VB NF-κB_NNP dependent_JJ ,_, and_CC IL-1β_NNP ,_, but_CC not_RB TGFβ_NNP ,_, is_VBZ a_DT potent_JJ inducer_NN of_IN NF-κB_NNP activation_NN ._. 
Hence_RB ,_, the_DT pathway_NN involved_VBN in_IN the_DT induction_NN of_IN the_DT COX-2_NNP gene_NN by_IN IL-1β_NNP and_CC TGFβ_NNP may_MD be_VB different_JJ ._. 
Furthermore_RB ,_, a_DT significant_JJ proportion_NN of_IN the_DT fibroblasts_NNS used_VBN in_IN the_DT study_NN by_IN Keerthisingam_NNP et_CC al._NN [_NN 21_CD ]_NN were_VBD obtained_VBN from_IN patients_NNS with_IN systemic_JJ sclerosis_NN ,_, which_WDT makes_VBZ their_PRP$ fibroblast_NN population_NN more_RBR heterogeneous_JJ ._. 
Wilborn_NNP et_CC al._NN [_NN 12_CD ]_NN also_RB reported_VBD a_DT decreased_VBN production_NN of_IN PGE_NNP 2_CD by_IN IL-1β-stimulated_NNP IPF_NNP fibroblasts_NNS ,_, due_JJ to_TO decreased_VBD COX-2_NNP expression_NN [_NN 12_CD ]_NN ._. 
There_EX is_VBZ a_DT possibility_NN that_DT patient_NN selection_NN may_MD have_VB differed_VBD between_IN the_DT two_CD studies_NNS ._. 
However_RB ,_, we_PRP feel_VBP certain_JJ that_IN the_DT diagnostic_JJ accuracy_NN of_IN our_PRP$ patient_NN population_NN was_VBD high_JJ ,_, due_JJ to_TO the_DT fact_NN that_IN 5_CD out_IN of_IN a_DT total_NN of_IN 7_CD IPF_NNP subjects_NNS included_VBN in_IN our_PRP$ study_NN underwent_VBD lung_NN transplantation_NN with_IN confirmatory_NN pathology_NN results_NNS consistent_JJ with_IN IPF_NNP ._. 
The_DT other_JJ 2_CD subjects_NNS had_VBD biopsy_NN -_- proven_VBN IPF_NNP ._. 
In_IN addition_NN ,_, our_PRP$ results_NNS were_VBD similar_JJ when_WRB comparing_VBG lung_NN fibroblasts_NNS obtained_VBN from_IN 5_CD subjects_NNS with_IN advanced_JJ stage_NN IPF_NNP with_IN those_DT of_IN 2_CD subjects_NNS at_IN an_DT earlier_JJR stage_NN of_IN their_PRP$ disease_NN ,_, who_WP had_VBD received_VBN no_DT therapy_NN ._. 
Although_IN the_DT reasons_NNS for_IN our_PRP$ different_JJ results_NNS are_VBP unclear_JJ ,_, the_DT fact_NN that_IN we_PRP found_VBD similar_JJ COX-2_NNP expression_NN and_CC PGE_NNP 2_CD production_NN in_IN normal_JJ and_CC IPF_NNP lung_NN fibroblasts_NNS suggests_VBZ that_DT loss_NN of_IN COX-2_NNP expression_NN is_VBZ not_RB a_DT universal_JJ characteristic_NN of_IN fibroblasts_NNS cultured_JJ from_IN the_DT lungs_NNS of_IN subjects_NNS with_IN IPF_NNP ._. 

Conclusion_NNP We_PRP have_VBP found_VBN that_WDT fibroblasts_NNS cultured_JJ from_IN normal_JJ and_CC IPF_NNP human_JJ lungs_NNS have_VBP a_DT significant_JJ and_CC similar_JJ induction_NN of_IN the_DT COX-2_NNP enzyme_NN when_WRB stimulated_VBN with_IN IL-1β_NNP ,_, but_CC that_IN IPF_NNP fibroblasts_NNS produced_VBD more_JJR thromboxane_NN and_CC had_VBD a_DT significantly_RB lower_JJR prostacyclin_NN :_: thromboxane_NN ratio_NN ._. 
We_PRP hypothesize_NN that_IN the_DT lower_JJR PGI_NNP 2_CD :_: TXA_NNP 2_CD ratio_NN seen_VBN in_IN HF-IPF_NNP may_MD be_VB a_DT phenotypic_JJ alteration_NN that_WDT plays_VBZ a_DT role_NN in_IN the_DT pathogenesis_NNS of_IN IPF_NNP ._. 

Abbreviations_NNP COX_NNP =_SYM cyclooxygenase_NN ;_: HF_NNP =_SYM human_JJ fibroblasts_NNS ;_: NL_NNP =_SYM normal_JJ lungs_NNS ;_: IPF_NNP =_SYM idiopathic_JJ pulmonary_JJ fibrosis_NNS ;_: IFN_NNP =_SYM interferon_NN ;_: IL_NNP =_SYM interleukin_NN ;_: PG_NNP =_SYM prostaglandin_NN ;_: TX_NNP =_SYM thromboxane_NN ;_: PGI_NNP 2_CD =_SYM prostacyclin_NN ._. 

