METHODS OF DIAGNOSING CHRONIC OBSTRUCTIVE PULMONARY DISEASE (COPD) USING NOVEL MOLECULAR BIOMARKERS

Abstract

The present invention relates to in vitro methods for the diagnosis of chronic obstructive pulmonary disease (COPD), wherein the expression of the marker gene KIAA1199 is determined. In particular, the invention relates to an in vitro diagnostic method of assessing the susceptibility of a subject to develop progressive COPD involving the appearance of irreversible lung damage, wherein the expression of the marker gene KIAA1199 and optionally one or more further marker genes selected from DMBT1, TMSB15A, DPP6, SLC51B, NUDT11, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, PLA1A, FGG, CEACAMS, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2, COMP, ITGA10, CTHRC1, TAL1, FIBIN, BEX5, BEX1, ESM1 and GHRL is determined. The invention also relates to an in vitro method of diagnosing stable COPD or assessing the susceptibility of a subject to develop stable COPD, wherein the expression of KIAA1199 and optionally one or more further marker genes selected from DMBT1, TMSB15A, DPP6, SLC51B, NUDT11, ELFS, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, PLA1A, FGG, CEACAMS, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2, COMP, ITGA10, CTHRC1, TAL1, FIBIN, BEX5, BEX1, ESM1 and GHRL is determined. Furthermore, the invention relates to the use of primers for transcripts of the aforementioned marker genes, the use of nucleic acid probes to transcripts of these marker genes, the use of microarrays comprising nucleic acid probes to transcripts of these marker genes, and the use of antibodies against the proteins expressed from these marker genes in corresponding in vitro methods. In vitro methods of monitoring the progression of COPD are also provided, in which the expression of marker genes according to the invention is determined.

Claims

1. An in vitro method for the diagnosis of chronic obstructive pulmonary disease (COPD), the method comprising determining the level of expression of the gene KIAA1199 in a sample obtained from a subject.

2. (canceled)

3. An in vitro diagnostic method of assessing the susceptibility of a subject to develop progressive chronic obstructive pulmonary disease (COPD) involving the appearance of irreversible lung damage, the method comprising: determining the level of expression of the gene KIAA1199 in a sample obtained from the subject; comparing the level of expression of KIAA1199 in the sample from the subject to a control expression level of KIAA1199 in a healthy subject; and determining whether or not the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage, wherein an increase in the level of expression of KIAA1199 in the sample from the subject as compared to the control expression level of KIAA1199 is indicative of a proneness to develop progressive COPD.

4. The method of claim 3, further comprising: determining the level of expression of one or more further genes selected from the group consisting of DMBT1, TMSB15A, DPP6, SLC51B, NUDT11, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, PLA1A, FGG, CEACAM5, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2, COMP, ITGA10, CTHRC1, TAL1, FIBIN, BEX5, BEX1, ESM1 and GHRL in the sample obtained from the subject; comparing the level of expression of the one or more further genes to a control expression level of the corresponding gene(s) in a healthy subject; and determining whether or not the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage, wherein an increase in the level of expression of KIAA1199, DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject as compared to the control expression level of the corresponding gene(s) is indicative of a proneness to develop progressive COPD, and wherein a decrease in the level of expression of TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TAL1, FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject as compared to the control expression level of the corresponding gene(s) is indicative of a proneness to develop progressive COPD.

5.-9. (canceled)

10. The method of claim 4, wherein the level of expression of DMBT1 and TMSB15A is determined.

11. The method of claim 4, wherein the level of expression of DMBT1, TMSB15A and at least one further gene selected from the group consisting of FGG, CYP1A1, CEACAM5, CTHRC1, NTRK2, RASGRF2, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, DPP6, SLC51B and NUDT11 is determined.

12. The method of claim 4, wherein it is determined that the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage if the level of expression of a majority of the number of genes tested is altered in the sense that (i) the level of expression of KIAA1199, DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject is increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TALL FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject is decreased as compared to the control expression level of the corresponding gene(s).

13. The method of claim 4, wherein it is determined that the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage if the level of expression of a majority of the number of genes tested is altered in the sense that (i) the level of expression of KIAA1199, DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject is at least 3-fold increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TALL FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject is at least 3-fold decreased as compared to the control expression level of the corresponding gene(s).

14. The method of claim 4, wherein it is determined that the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage if the level of expression of at least 70% of the number of genes tested is altered in the sense that (i) the level of expression of KIAA1199, DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject is increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TALL FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject is decreased as compared to the control expression level of the corresponding gene(s).

15. The method of claim 4, wherein it is determined that the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage if the level of expression of at least 70% of the number of genes tested is altered in the sense that (i) the level of expression of KIAA1199, DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject is at least 3-fold increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TALL FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject is at least 3-fold decreased as compared to the control expression level of the corresponding gene(s).

16. (canceled)

17. The method of claim 3, wherein the subject is a human.

18. (canceled)

19. The method of claim 3, wherein the sample obtained from the subject is a lung tissue sample, or the sample obtained from the subject is a transbronchial lung biopsy sample or a bronchoalveolar lavage sample.

20. (canceled)

21. The method of claim 3, wherein the level of expression of KIAA1199 and, if applicable, the one or more further genes in the sample from the subject is determined by determining the level of transcription of the corresponding gene(s).

22. The method of claim 21, wherein the level of transcription is determined using a quantitative reverse transcriptase polymerase chain reaction or a microarray.

23.-25. (canceled)

26. An in vitro method of diagnosing stable chronic obstructive pulmonary disease (COPD) in a subject or assessing the susceptibility of a subject to develop stable COPD, the method comprising: determining the level of expression of the gene KIAA1199 in a sample obtained from the subject; comparing the level of expression of KIAA1199 in the sample from the subject to a control expression level of KIAA1199 in a healthy subject; and determining whether or not the subject suffers from stable COPD or is prone to suffer from stable COPD, wherein a decrease in the level of expression of KIAA1199 in the sample from the subject as compared to the control expression level of KIAA1199 is indicative of stable COPD or a proneness to stable COPD.

27. The method of claim 26, further comprising: determining the level of expression of one or more further genes selected from the group consisting of DMBT1, TMSB15A, DPP6, SLC51B, NUDT11, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, PLA1A, FGG, CEACAM5, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2, COMP, ITGA10, CTHRC1, TAL1, FIBIN, BEX5, BEX1, ESM1 and GHRL in the sample obtained from the subject; comparing the level of expression of the one or more further genes to a control expression level of the corresponding gene(s) in a healthy subject; and determining whether or not the subject suffers from stable COPD or is prone to suffer from stable COPD, wherein an increase in the level of expression of DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject as compared to the control expression level of the corresponding gene(s) is indicative of stable COPD or a proneness to stable COPD, and wherein a decrease in the level of expression of KIAA1199, TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TAL1, FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject as compared to the control expression level of the corresponding gene(s) is indicative of stable COPD or a proneness to stable COPD.

28.-33. (canceled)

34. The method claim 27, wherein it is determined that the subject suffers from stable COPD or is prone to suffer from stable COPD if the level of expression of a majority of the number of genes tested is altered in the sense that (i) the level of expression of DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject is increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of KIAA1199, TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TAL1, FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject is decreased as compared to the control expression level of the corresponding gene(s).

35.-36. (canceled)

37. The method of any onc of claims 27 to 33 claim 27, wherein it is determined that the subject suffers from stable COPD or is prone to suffer from stable COPD if the level of expression of at least 70% of the number of genes tested is altered in the sense that (i) the level of expression of DMBT1, ELF5, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and/or COMP in the sample from the subject is at least 3-fold increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of KIAA1199, TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TALL FIBIN, BEX5, BEX1, ESM1 and/or GHRL in the sample from the subject is at least 3-fold decreased as compared to the control expression level of the corresponding gene(s).

38. (canceled)

39. The method of any one of claim 26, wherein the subject is a human.

40. (canceled)

41. The method of claim 26, wherein the sample obtained from the subject is a lung tissue sample, or the sample obtained from the subject is a transbronchial lung biopsy sample or a bronchoalveolar lavage sample.

42. (canceled)

43. The method of any one of claim 26, wherein the level of expression of KIAA1199 and, if applicable, the one or more further genes in the sample from the subject is determined by determining the level of transcription of the corresponding gene(s).

44. The method of claim 43, wherein the level of transcription is determined using a quantitative reverse transcriptase polymerase chain reaction or a microarray.

45.-47. (canceled)

48. An in vitro diagnostic method of assessing the susceptibility of a subject suffering from stable chronic obstructive pulmonary disease (COPD) to develop progressive COPD involving the appearance of irreversible lung damage, the method comprising: determining the level of expression of the gene KIAA1199 in a sample obtained from the subject; comparing the level of expression of KIAA1199 in the sample from the subject to a control expression level of KIAA1199 in a subject suffering from stable COPD; and determining whether or not the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage, wherein an increase in the level of expression of KIAA1199 in the sample from the subject as compared to the control expression level of KIAA1199 is indicative of a proneness to develop progressive COPD.

49. The method of claim 48, further comprising: determining the level of expression of one or more further genes selected from the group consisting of DMBT1, ELF5, AZGP1, PRRX1, AQP3, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2, COMP, ITGA10, CTHRC1, TAL1, BEX1 and GHRL in the sample obtained from the subject; comparing the level of expression of the one or more further genes to a control expression level of the corresponding gene(s) in a subject suffering from stable COPD; and determining whether or not the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage, wherein an increase in the level of expression of KIAA1199, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2 and/or TAL1 in the sample from the subject as compared to the control expression level of the corresponding gene(s) is indicative of a proneness to develop progressive COPD, and wherein a decrease in the level of expression of DMBT1, ELF5, AZGP1, PRRX1, AQP3, COMP, ITGA10, CTHRC1, BEX1 and/or GHRL in the sample from the subject as compared to the control expression level of the corresponding gene(s) is indicative of a proneness to develop progressive COPD.

50.-54. (canceled)

55. The method of any one of claim 49, wherein it is determined that the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage if the level of expression of a majority of the number of genes tested is altered in the sense that (i) the level of expression of KIAA1199, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2 and/or TALI in the sample from the subject is increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of DMBT1, ELF5, AZGP1, PRRX1, AQP3, COMP, ITGA10, CTHRC1, BEX1 and/or GHRL in the sample from the subject is decreased as compared to the control expression level of the corresponding gene(s).

56.-57. (canceled)

58. The method of any one of claim 49, wherein it is determined that the subject is prone to develop progressive COPD involving the appearance of irreversible lung damage if the level of expression of at least 70% of the number of genes tested is altered in the sense that (i) the level of expression of KIAA1199, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, HYAL2, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, CST6, NTRK2 and/or TALI in the sample from the subject is at least 3-fold increased as compared to the control expression level of the corresponding gene(s) and (ii) the level of expression of DMBT1, ELF5, AZGP1, PRRX1, AQP3, COMP, ITGA10, CTHRC1, BEX1 and/or GHRL in the sample from the subject is at least 3-fold decreased as compared to the control expression level of the corresponding gene(s).

59. (canceled)

60. The method of claim 48, wherein the subject is a human.

61. (canceled)

62. The method of claim 48, wherein the sample obtained from the subject is a lung tissue sample, or the sample obtained from the subject is a transbronchial lung biopsy sample or a bronchoalveolar lavage sample.

63. (canceled)

64. The method of claim 48, wherein the level of expression of KIAA1199 and, if applicable, the one or more further genes in the sample from the subject is determined by determining the level of transcription of the corresponding gene(s).

65. The method of claim 64, wherein the level of transcription is determined using a quantitative reverse transcriptase polymerase chain reaction or a microarray.

66.-72. (canceled)

73. A method of treating chronic obstructive pulmonary disease (COPD), the method comprising administering a drug against COPD to a subject that has been identified in a method as defined in claim 3 as being prone to develop progressive COPD involving the appearance of irreversible lung damage.

74.-75. (canceled)

76. The method of claim 73, wherein the drug against COPD is bitolterol, carbuterol, fenoterol, pirbuterol, procaterol, reproterol, rimiterol, salbutamol, levosalbutamol, terbutaline, tulobuterol, arformoterol, bambuterol, clenbuterol, formoterol, olodaterol, salmeterol, indacaterol, beclometasone, betamethasone, budesonide, ciclesonide, flunisolide, fluticasone, mometasone, triamcinolone, aclidinium bromide, glycopyrronium bromide, ipratropium bromide, oxitropium bromide, tiotropium bromide, cromoglicate, nedocromil, acefylline, ambuphylline, bamifylline, doxofylline, enprofylline, etamiphylline, proxyphylline, theobromine, theophylline, aminophylline, choline theophyllinate, montelukast, pranlukast, zafirlukast, zileuton, ramatroban, seratrodast, ibudilast, roflumilast, amlexanox, eprozinol, fenspiride, omalizumab, epinephrine, hexoprenaline, isoprenaline, isoproterenol, orciprenaline, metaproterenol, atropine, or a pharmaceutically acceptable salt of any of the aforementioned agents, or any combination thereof.

77. The method of claim 73, wherein the drug against COPD is roflumilast.

78.-81. (canceled)

82. A method of treating or preventing chronic obstructive pulmonary disease (COPD), the method comprising administering a drug against COPD to a subject that has been identified in a method as defined in claim 26 as suffering from stable COPD or as being prone to suffer from stable COPD.

83.-84. (canceled)

85. The method of claim 82, wherein the drug against COPD is bitolterol, carbuterol, fenoterol, pirbuterol, procaterol, reproterol, rimiterol, salbutamol, levosalbutamol, terbutaline, tulobuterol, arformoterol, bambuterol, clenbuterol, formoterol, olodaterol, salmeterol, indacaterol, beclometasone, betamethasone, budesonide, ciclesonide, flunisolide, fluticasone, mometasone, triamcinolone, aclidinium bromide, glycopyrronium bromide, ipratropium bromide, oxitropium bromide, tiotropium bromide, cromoglicate, nedocromil, acefylline, ambuphylline, bamifylline, doxofylline, enprofylline, etamiphylline, proxyphylline, theobromine, theophylline, aminophylline, choline theophyllinate, montelukast, pranlukast, zafirlukast, zileuton, ramatroban, seratrodast, ibudilast, roflumilast, amlexanox, eprozinol, fenspiride, omalizumab, epinephrine, hexoprenaline, isoprenaline, isoproterenol, orciprenaline, metaproterenol, atropine, or a pharmaceutically acceptable salt of any of the aforementioned agents, or any combination thereof.

86.-90. (canceled)

91. A method of treating chronic obstructive pulmonary disease (COPD), the method comprising administering a drug against COPD to a subject suffering from stable COPD, wherein the subject has been identified in a method as defined in claim 48 as being prone to develop progressive COPD involving the appearance of irreversible lung damage.

92.-93. (canceled)

94. The method of claim 91, wherein the drug against COPD is bitolterol, carbuterol, fenoterol, pirbuterol, procaterol, reproterol, rimiterol, salbutamol, levosalbutamol, terbutaline, tulobuterol, arformoterol, bambuterol, clenbuterol, formoterol, olodaterol, salmeterol, indacaterol, beclometasone, betamethasone, budesonide, ciclesonide, flunisolide, fluticasone, mometasone, triamcinolone, aclidinium bromide, glycopyrronium bromide, ipratropium bromide, oxitropium bromide, tiotropium bromide, cromoglicate, nedocromil, acefylline, ambuphylline, bamifylline, doxofylline, enprofylline, etamiphylline, proxyphylline, theobromine, theophylline, aminophylline, choline theophyllinate, montelukast, pranlukast, zafirlukast, zileuton, ramatroban, seratrodast, ibudilast, roflumilast, amlexanox, eprozinol, fenspiride, omalizumab, epinephrine, hexoprenaline, isoprenaline, isoproterenol, orciprenaline, metaproterenol, atropine, or a pharmaceutically acceptable salt of any of the aforementioned agents, or any combination thereof.

95.-107. (canceled)

Description

[0137] The invention is also described by the following illustrative figures. The appended figures show:

[0138] FIG. 1: Study design of the COPD-AUVA study conducted at the Vienna Medical University (see Example 1).

[0139] FIG. 2: Overview of the numbers of subjects of different disease states who underwent the COPD-AUVA study.

[0140] FIG. 3: Overview of healthy subjects (A) and of subjects with either chronic bronchitis but no signs of pulmonary obstruction (COPD “at risk”; “GOLD 0”) at visit 1 (B) or with manifest COPD at visit 1 (C), as well as the development of COPD (severity according to GOLD criteria), bronchitis and smoking habits in these subjects over the period from visit 1 (day 0) to visit 2 (12 months) to visit 3 (36 months). The term “pack years” refers to a person's cigarette consumption calculated as the packs of cigarettes (each pack containing 20 cigarettes) smoked per day, multiplied by the length of cigarette consumption in years. (D) Clinical characteristics of participants in the COPD-AUVA study and changes between baseline and visit 3 (see Example 1).

[0141] FIG. 4: COPD Pathology module 1: Development of chronic bronchitis: Progressive inhibition of adaptive motility of mucosal cells caused by the inhibition of coordinated actin cytoskeleton movements.

[0142] Chronic bronchitis starts with the significant downregulation of genes that control assembly, polymerization, motility, stabilization and energy supply of F actin-mediated cytoskeleton movements (suppression of thymosin beta 15A (TMSB15A), dipeptidyl-peptidase 6 (DPP6), nudix (nucleoside diphosphate linked moiety X)-type motif 11 (NUDT11), and integrin alpha 10 (ITGA10)). At the same time, expression of the RASGRF2 gene known to inhibit Cdc42-mediated polymerization of actin during cellular movements is progressively increased during advancement of COPD (FIGS. 4A and 4D) indicating that the inhibition of cellular motility is not only a leading mechanism in early stages of COPD development, but also part of the progressive membrane destruction in later stages of COPD.

[0143] Of note, reduced expression of these genes is also connected to increasing intensity of bronchial inflammation. This characteristic expression pattern includes the SLC51B gene (FIG. 4D) which is as yet largely known for its capacity to transport steroid-precursor molecules in intestinal cells.

[0144] The compensatory activation of the GTPase RND1 (Rho family GTPase 1) best known for its ability to control the organization of the actin cytoskeleton in response to growth factor stimulation is just increased up to COPD GOLD stage II not only indicating a complete failure of actin-dependent cellular cytoskeleton organization in later stages of COPD, but also the loss of the regenerative capacity, as also demonstrated within Module 3 (see FIGS. 6A-6E). This in turn concurs rather well with the progressive downregulation of the cystatin M/E (CST6) gene being annotated with both functional differentiation of epithelial cells and maintenance of surface integrity.

[0145] As the coordinated action of these molecules is required for controlled movements of epithelial cells during pivotal processes, such as growth, intercalation and extrusion of cells within a cohesive cell layer system, the loss of these functions causes a profound disturbance of membrane integrity allowing for the development of non-specific bronchial inflammation that basically reflects all constituents of ventilated air including combustion products, such as cigarette smoke or welding fumes.

[0146] FIG. 5: COPD Pathology module 2: Bi-phasic activation of mucosal immunity.

[0147] Driven by this loss of cellular cohesion, the bronchus develops a diverse mucosal immune response that combines mechanisms of acute inflammation, such as the expression of fibrinogen (FGG) (FIGS. 5A and 5D), the upregulation of carcinoembryonic antigen-related cell adhesion molecule 5 (CEACAM 5) (FIGS. 5A and 5D), and aryl hydrocarbon receptor (AHR) signaling, the latter characterized by increased expression of cytochrome P450, family 1, subfamily A polypeptide 1 (CYP1A1) and cytochrome P450, family 1, subfamily B polypeptide 1 (CYP1B1) (FIGS. 5A and 5E, 5F). Intensity of AHR signaling is significant, in spite of the increased compensatory expression of the aryl hydrocarbon receptor repressor gene (AHRR), most likely reflecting the continuous impact of smoke. As CEACAMs have recently been shown to act as surface receptors for gram-negative bacteria such as Neisseria meningitidis, Haemophilus influenzae and Moraxella catarrhalis being frequently found in progressive bronchitis, this mechanism is prone to contribute to episodes of intensified bronchial inflammation.

[0148] Nonetheless, neither FGG nor CEACAM5 expression causes short-term worsening of non-reversible pulmonary obstruction (FIG. 5D, middle panel), although the activation of both genes significantly contributes to the intensity of bronchial inflammation (FIG. 4D, right panel). This differs from CYP1A2, KIAA1199 and phospholipase Al member A (PLA1A) expression (FIG. 4b and e) that all correlate with a significant deterioration of pulmonary function. While CYP1A2 expression as part of a smoke-induced AHR signaling response fits well to the current perception of COPD development, the strong correlation of KIAA1199 and PLA1A expression with deterioration of pulmonary function according to GOLD criteria points towards another direction, the complete failure of the bronchial compartment system. KIAA1199 has recently been demonstrated to activate matrix hyaluronidases while phospholipase Al member A (PLA1A) is known to activate T cells in response to non-specific inflammatory stimulation. It has presently been found that the significant upregulation of KIAA1199 is characteristic for the second phase of increased bronchial inflammation in GOLD stages III and IV (FIG. 5B) which follows a phase of non-progressive bronchial inflammation characterizing GOLD stage I (FIG. 5A). Notably, during this stabilization phase both the expression of KIAA1199 and of PLA1A is reduced as well (FIG. 5B). Given the strong proinflammatory impact of a degradation of high molecular mass hyaluronan, these observations indicate that the final increase of inflammatory activity in COPD GOLD stage III and IV is the combined result of permanently disturbed epithelial integrity and a secondary destruction of the hyaluronan matrix within the bronchial wall by the activation of matrix hyaluronidases. This view is supported by the expression pattern of matrix hyaluronidase 2 (HYAL2) itself which represents the leading hyaluronan-degrading enzyme in humans (FIG. 5C).

[0149] FIG. 6: COPD Pathology module 3: The impact of intensified regenerative repair: temporary suspension of progressive bronchial inflammation.

[0150] Maintaining the structural integrity of the mucosa as well as upholding essential components of the bronchial wall is part of effective wound healing and as such an indispensable measure to prevent the intrusion of antigens, allergens and infectious agents into submucosal compartments. It is thus not surprising that various genes guiding functions of epithelial repair are upregulated in response to increased inflammation, as demonstrated in FIG. 6A. However, only a small group of these genes is significantly contributing to the temporary suspension of progressive bronchial inflammation in GOLD stage I, genes known to participate in epithelial regeneration and differentiation, bacterial defense and transepithelial water transport (FIGS. 6A-6C): a) deleted in malignant brain tumors 1 (DMBT1), b) zinc-binding alpha-2-glycoprotein 1 (AZGP1), and c) aquaporin 3 (AQP3). However, this regenerative impulse does not last long as expression of these genes decreases again once progression of inflammation resumes stressing the impact of KIAA1199 expression and matrix degradation on bronchial inflammation. Although further genes closely related to epithelial repair, such as stratifin (SFN), the G protein-coupled orphan receptor 110 (GPR110), the smoke-inducible growth differentiation factor 15 (GDF15), and E74-like factor 5 (ELFS) are expressed throughout a much longer period of COPD development (FIG. 6A), the effectiveness of this wound healing approach is evidently not sufficient to maintain bronchial integrity and to balance bronchial inflammation in the presence of epithelial disintegration and progressive hyaluronan breakdown.

[0151] As a result, simultaneous measurement of DMBT1 and KIAA1199 gene expression is capable of discerning stable from progressive COPD (according to GOLD criteria), if the difference between DMBT1 and KIAA1199 expression exceeds a value of 3.63 (FIG. 6E). The importance of intensified KIAA1199 expression for progressive epithelial inflammation is further stressed by the fact that in chronic inflammatory wound healing of diabetic skin, expression of KIAA1199 is significantly upregulated, whereas in normal skin repair, KIAA1199 expression is reduced (see FIG. 8). It should also be noted that KIAA1199 expression in aged skin is in general significantly higher than in the skin from younger individuals (p<0.01).

[0152] FIG. 7: Expression of KIAA1199 in skin wound healing.

[0153] FIG. 8: COPD Pathology module 4: Scar formation by predominant mesenchymal repair as the result of regenerative failure in the presence of a prevailing structural deficit.

[0154] As in any situation of prevailing unresolved repair that is not life-threatening, activation of “secondary” mesenchymal repair will serve as the exit strategy to remove the structural deficit and to terminate wound healing. During progression of COPD, coordinated gene activation in this regard can be divided into two categories: a) permanent support of mesenchymal repair (expression of NTRK2 and SOS1 genes) (FIGS. 8A and 8B), b) support of mesenchymal repair during both functional “primary” repair and non-functional “secondary” wound healing (expression of COMP, PRRX1 and CTHRC1 genes) (FIGS. 8A-8C).

[0155] As in any form of predominantly mesenchymal repair, expression of genes controlling vascular growth and differentiation is progressively diminished. FIG. 8D provides a synopsis of the expression pattern and relevant annotations for all genes related to vascular outgrowth and repair which are significantly regulated during progression of COPD.

[0156] The invention will now be described by reference to the following examples which are merely illustrative and are not to be construed as a limitation of the scope of the present invention.

EXAMPLES

Example 1

Controlled Prospective Pilot Trial Aimed at Identifying Symptom-Based Molecular Metabolic Markers for Progressive COPD (Vienna COPD-AUVA Study)

[0157] Introduction

[0158] In the context of the present invention, a controlled prospective pilot trial aimed at the identification of symptom-based molecular metabolic markers for progressive COPD was conducted at the Vienna Medical University between 2007 and 2012. The Vienna COPD-AUVA study combined the assessment of validated clinical measures for COPD following in part the overall strategy of the ECLIPSE trial (Vestbo et al., 2011), the largest and most elaborate study addressing progress and variability of COPD.

[0159] For stratification of patients, a three-year analysis (day 0, 12 months, and 36 months) of symptom scoring (St. George Respiratory questionnaire, activity and symptom score), assessment of pulmonary function, cardiopulmonary exercise testing, and radiological evaluation by computer-assisted tomography (high-resolution mode) were combined with whole genome transcription analysis plus quantitative RT-PCR assessment and mass spectrometry proteomics. As shown in FIG. 1, the patients were grouped into three strata, two of which presented at the start of the study with regular lung function, either without any sign of a cardiopulmonary disease (healthy volunteers) or with symptoms of chronic bronchitis (COPD “at risk”), and a group of volunteers having symptoms of chronic bronchitis together with deteriorated lung function (COPD at GOLD stages I-IV).

[0160] Study visits were performed at base line and after 12 and 36 months, respectively. Each visit was performed on an ambulatory basis and included medical history, physical examination, pulmonary function tests (PFT), cardiopulmonary exercise tests (CPET), radiological assessment by computer-assisted tomography (CAT) scans and a bronchoscopy. On each visit, both personal and occupational history was taken as well as smoking history which comprised onset and duration of symptoms related to COPD, production of phlegm (frequency, quantity, and color), intensity of symptoms measured by the St. George Respiratory Questionnaire (SGRQ; activity and symptom score index) and assessment of life quality using the SF-36 questionnaire. The rate of exacerbations (frequency, number of hospitalizations, use of antibiotics, corticosteroids or combined treatment) and the individual medication were also recorded.

[0161] Pulmonary function tests (PFT) were taken at each visit and included blood drawings, body plethysmography, spirometry and quantitative measurement of pulmonary gas exchange at rest and during symptom-limited cardiopulmonary exercise testing (CPET). PFT was performed with an Autobox DL 6200 (Sensor Medics, Vienna, Austria), and CPET on a treadmill using the Sensormedics 2900 Metabolic Measurement Cart. Formulas for calculation of reference values were taken from Harnoncourt et al., 1982. Predicted values were derived from the reference values of the Austrian Society of Pneumology following the recommendations of the European Respiratory Society (Rabe et al., 2007).

[0162] Serum samples were analyzed for complete cellular blood count, electrolytes, glucose, C-reactive protein, fibrinogen, and coagulation parameters.

[0163] Prior to bronchoscopy, CAT scans encompassing high resolution-computed tomography (HRCT) were performed. Following additional informed consent on each visit, bronchoscopy was performed. During bronchoscopy, both bronchoalveolar lavage (BAL) samples and transbronchial biopsy samples (five per segment in each middle lobe) were taken.

[0164] Biological analysis was performed in transbronchial lung biopsies taken during bronchoscopy from two pulmonary localizations (5 each) of the middle-lobe after radiological assessment by computer-assisted tomography (CAT) scans including high-resolution scanning. CAT scans were used for the assessment of emphysema formation as well as for the exclusion of tumor development and infection. During the controlled observational period, combined assessment of clinical and molecular development was finally possible in 120 volunteers. Biomarkers were identified in each case by means of the individual changes of pulmonary function and clinical symptoms characteristic for the progression of COPD. As a result, this approach makes use of the well-known variability of clinical phenotypes in COPD and their variable course of progression while at the same time identifying the very set of biomolecules responsible for this type of disease progression.

[0165] Clinical Analysis

[0166] The study protocol was approved by the ethical committee of the Medical University of Vienna (ClinicalTrials.gov Identifier: NCT00618137). Following informed consent during screening, individuals were stratified at visit 1 (day 0) if they fulfilled the following criteria:

TABLE-US-00002 TABLE 2 Stratification of subjects at visit 1 (day 0). Inclusion criteria Occupational history Healthy Controls Age 18-70 years No occupation with No history or clinical findings suggestive of any disease increased exposure towards Never Smoker combustion products, Normal pulmonary function test at study entry particularly no welding or professional car driving COPD, at risk′ Age 18-70 years Professional car driver Chronic bronchitis according to WHO with repeated episodes of or welder with increased phlegm production occupational exposure No history or clinical findings suggestive of bronchial asthma towards combustion products Normal PFT according to GOLD criteria at study entry of at least 10 years Smoking history of at least 10 years No history or clinical findings suggestive of cardiovascular or malign disease COPD manifest Age 18-70 years Professional car driver Chronic bronchitis according to WHO with repeated episodes of or welder with increased phlegm production occupational exposure No history or clinical findings suggestive of bronchial asthma towards combustion products Pathological PFT according to GOLD criteria at study entry of at least 10 years Smoking history of at least 10 years No history or clinical findings suggestive of cardiovascular or malign disease

[0167] 396 individuals were screened, 185 of whom met the study criteria. 136 participants finished visit 2 after 12 months, and 120 completed the final visit after 36 months of controlled observation. Throughout the study, all participants were residing and occupied in the greater Vienna area in order to ensure comparable environmental conditions. The control group consisted of 16 healthy volunteers who had never smoked (7 females and 9 males; mean age 36±12.2 years), as also shown in Table 2 above. None of the healthy participants developed any symptom of pulmonary disease during the study period. At the start of the study, 104 participants presented with clinical symptoms of chronic bronchitis according to WHO definition, 55 of whom did not have signs of non-reversible bronchial obstruction (GOLD “at risk”), while the other 49 participants showed bronchial obstruction ranging from GOLD stage I to IV as determined by PFT (see FIG. 3D). All participants in the COPD and COPD “at risk” groups were active cigarette smokers with a smoking history of more than 10 pack years, except for one welder who in addition to a daily expectoration of phlegm reported about frequent episodes of bronchial infection (>2 per year) without radiological signs of bronchiectasis. 64 participants were working as taxi or bus drivers (53%) and 40 active welders (33%) with a previous exposure to welding fumes of more than 10 years.

[0168] At visit 1, the majority of participants with manifest COPD had bronchial obstruction GOLD stage II and III (n=38), while the remaining subjects were in COPD GOLD stage I (n=9) and IV (n=2) (see FIG. 3D). Mean age in GOLD stages I and II was 50±9.5 and 56±10.4 yrs. respectively, compared to 52±9.0 yrs. in GOLD stage III and 63±11 yrs. in GOLD stage IV. 29% of the participants in the GOLD “at risk” group were already presenting with a continuous daily expectoration of sputum, and sputum was frequently discolored (yellow, green, brown) in 27%.

[0169] During controlled observation (36 months), 14 participants (12%) had a progression of disease according to GOLD, 7 (13%) in the GOLD “at risk” group, 1 (11%) in GOLD I, 3 (12%) in GOLD II, and 3 (25%) in GOLD Ill. Improvement of bronchial obstruction according to GOLD was observed in 13 individuals (5 participants in both GOLD stage I and II, and 3 cases in GOLD stage Ill and IV), mostly connected to a cessation of cigarette smoking.

[0170] As part of the observational design of the study, participants were not specifically encouraged to stop smoking. Accordingly, smoking habits changed only slightly: only 5 participants of the “COPD at risk” group (9%) and 2 participants in the “manifest COPD” group (4%) stopped smoking during the observational period, while 31% reduced cigarette smoking (data not shown). These changes did not significantly alter both occurrence and intensity of chronic bronchitis symptoms, as 27 participants (23%) demonstrated improvement and deterioration of cough and sputum production.

[0171] Biological/Molecular Analysis (Gene Transcription in Pulmonary Tissue)

[0172] RNAlater (Ambion, lifetechnologies) was used for tissue asservation. The lung biopsy material was disrupted using Lysing Matrix D ceramic balls in a Fastprep 24 system (MP Biomedical, Eschwege). A chaotropic lysis buffer (RLT, RNeasy Kit, Qiagen, Hilden) was used, followed by a phenol/chloroform extraction and subsequent clean up using the spin column approach of the RNeasy Mini Kit (Qiagen, Hilden) according to the manufacturer's manual, including a DNase I digestion on the chromatography matrix. RNA quantification was done spectrophotometrically using a NanoDrop 1000 device (Thermo Scientific) and quality control was performed on the Agilent 2100 Bioanalyzer. A cut off for the amount of 1 microgram and a RNA integrity number of 7.0 was chosen.

[0173] Total RNA samples were hybridized to Human Genome U133plus 2.0 array (Affymetrix, St. Clara, CA), interrogating 47,000 transcripts with more than 54,000 probe sets.

[0174] Array hybridization was performed according to the supplier's instructions using the “GeneChip® Expression 3′ Amplification One-Cycle Target Labeling and Control reagents” (Affymetrix, St. Clara, Calif.). Hybridization was carried out overnight (16h) at 45° C. in the GeneChip® Hybridization Oven 640 (Affymetrix, St. Clara, Calif.). Subsequent washing and staining protocols were performed with the Affymetrix Fluidics Station 450. For signal enhancement, antibody amplification was carried out using a biotinylated anti-streptavidin antibody (Vector Laboratories, U.K.), which was cross-linked by a goat IgG (Sigma, Germany) followed by a second staining with streptavidin-phycoerythrin conjugate (Molecular Probes, Invitrogen). The scanning of the microarray was done with the GeneChip® Scanner 3000 (Affymetrix, St. Clara, Calif.) at 1.56 micron resolution.

[0175] The data analysis was performed with the MAS 5.0 (Microarray Suite statistical algorithm, Affymetrix) probe level analysis using GeneChip Operating Software (GCOS 1.4) and the final data extraction was done with the DataMining Tool 3.1 (Affymetrix, St. Clara, Calif.).

[0176] CEL files were imported and processed in R/Bioconductor (Gentleman et al., 2004). Briefly, data was preprocessed using quantile normalization (Gentleman et al., 2004) and combat (Johnson et al., 2007), linear models were calculated using limma (Smyth GK, 2005) and genes with a p-value of the f-statistics <5e-3 were called significant. Those genes were grouped into 20 clusters of co-regulated genes. The procedure of modeling and clustering was repeated for GOLD and phlegm as covariates.

[0177] For subsequent Gene Ontology (GO)-analysis it was necessary to separate the effects of GOLD and phlegm on gene expression. To this end, the GOLD classifications were grouped into “no COPD” (healthy and GOLD 0) and “COPD” (GOLD grades I-IV). Similarly, phlegm was reclassified into a “phlegm” group (productive or severe) and a “no phlegm” group (health or no/dry). Based on these reclassifications, gene expression was modeled using a 2×2 factorial design, resulting in five different lists of genes: (1) genes which are regulated with phlegm in the presence of COPD, (2) genes which are regulated with phlegm in the absence of COPD, (3) genes which are regulated with COPD in the presence of COPD, (4) genes which are regulated with COPD in the absence of COPD and finally (5) genes which are regulated differently with COPD, depending on whether there is phlegm or not.

[0178] These lists were annotated with respect to their biological functions as catalogued in the Gene Ontology (GO) database using the ClueGO plugin for the Cytoscape framework.

[0179] Results of Combined Clinical and Molecular Analysis

[0180] Activation of Epithelial Repair Mechanisms

[0181] Systematic analysis of the significant changes of gene expression during COPD development reveals a differentiated picture: As shown in FIGS. 6A to 6D, mechanisms of regeneration and repair commence as soon as the chronic inflammatory process in the peripheral bronchial tree is established. This is already the case in persistent or repeatedly manifesting bronchitis (COPD “at risk”). The functions associated with this kind of aberration from the normal equilibrium, in ontological terms still only potential COPD, include mediators involved in the regulation of embryonic epidermal and pulmonary growth, such as ELFS (E74-like factor 5; ETS domain transcription factor) which confers spatially controlled outgrowth of epithelial structures (Metzger et al., 2008; Yaniw et al., 2005) as well as mucosal immunity of the lung (Lei et al., 2007). Not surprisingly, the expression of ELFS is accompanied by a significant upregulation of stratifin (SFN) conferring increased epidermal regeneration and differentiation (Medina et al., 2007), yet also reduced deposition of matrix proteins including collagen I (Chavez-Munoz et al., 2012) and reduced functions of non-specific surface immunity (Butt et al., 2012). This regenerative phase of repair involves not only the G protein-coupled orphan receptor GPR110 and the smoke-inducible growth differentiation factor 15 (GDF15) (Wu et al., 2012), a member of the bone morphogenic protein-transforming growth factor-beta superfamily, but also mediators directing differentiated epithelial repair, such as the zinc-binding alpha-2-glycoprotein 1 (AZGP1), and the DMBT1 gene (deleted in malignant brain tumors 1) which is strongly upregulated during acute but resolving bacterial inflammation in enteral epithelia during appendicitis (Kaemmerer et al., 2012), suggesting a functional relevance for mucosal defense (Diegelmann et al., 2012). The almost identical expression profile of DMBT1 and AZGP1, a mediator capable of inducing a strong epithelial transdifferentiation in tumor cells (Kong et al., 2010), suggests an as yet undefined combinatory effect of both mediators on cellular differentiation during epithelial regeneration. Notably, the expression of these genes is strongly increased in individuals with COPD GOLD I and decreases significantly with progression of COPD, as also shown in FIG. 6A. In line with this observation, all mediators conveying epithelial regeneration and differentiation were found to be significantly downregulated during the transition from COPD stage III to COPD stage IV.

[0182] Activation of mediators of regenerative repair was also found in individuals demonstrating significant symptoms of bronchial inflammation, as demonstrated by a uniform increase of gene expression of SFN, GPR110 (see also FIG. 6D), and aquaporin 3 (AQP3) (see FIG. 6A) being an additional mediator known to guide proliferation and differentiation of epithelial cells (Nakahigashi et al., 2011; Kim et al., 2010). However, expression of these factors did not further increase with an increase of severity of bronchial inflammation, much in contrast to mediators capable of intensifying inflammation on epithelial surfaces, such as the carcinoembryonic antigen-related cell adhesion molecule 5 (CEACAMS) (see FIGS. 5A and 5D), or factors being part of the preferentially mesenchymal wound healing response during inflammatory repair (Agarwal et al., 2012; Agarwal et al., 2013), such as the cartilage oligomeric matrix protein (COMP) (see FIGS. 8A and 8C). The study design allowed as well for the measurement of changes of gene expression occurring throughout the study period of 3 years, possibly indicating significant changes of repair during short-term progression of COPD. Here, a significant downregulation of GPR110 and DMBT1 genes correlating with deteriorated lung function according to GOLD was found, as also shown in FIGS. 6B and 6D. This decrease of regenerative gene activity started already in GOLD stage II, where it was accompanied by a striking increase of repair functions related to mesenchymal wound healing (see also FIG. 8).

[0183] Progressive Activation of Mesenchymal Repair

[0184] During later stages of COPD, expression of mediators favoring mesenchymal repair became increasingly prominent. This did not only relate to the increased expression of the COMP gene (see FIGS. 8A and 8C), but also to the expression of potent activators of mesenchymal stem cells, such as the son of sevenless homolog 1 (SOS1) gene, a guanine nucleotide exchange factor for RAS proteins acting as the cognate receptor for hepatocyte growth factor, and to the paired related homeobox 1 gene (PRRX1), a transcriptional co-activator of RAS transcription factors belonging to the HOX family of early differentiation factors able to induce mesenchymal outgrowth in liver cirrhosis (Jiang et al., 2008) as well as epithelial-to-mesenchymal transition (EMT) during cancer development (Ocaña et al., 2012). While their pattern of expression indicates that both COMP and PRRX1 genes take also part in the regenerative phase of wound healing characterizing GOLD stage I and II, their later increase during transition from GOLD stage III to IV suggests an additional involvement in the progressive scarring of the airways. Increased expression of pro-fibrotic factors is further demonstrated by the striking increase of expression of neurotrophic tyrosine kinase receptor type 2 (or tropomyosin receptor kinase B receptor; TrkB) (NTRK2). NTRK2/TrkB, thus far known to act as high affinity receptor for various neurotrophic growth factors during nerve development, is also capable of promoting resistance of mesenchymal cells towards apoptosis and anoikis (Frisch et al., 2013). The combined increase of profibrotic mediators includes as well the expression of the collagen triple helix repeat containing 1 gene (CTHRC1) capable of conferring fibrotic organ dystrophy (Spector et al., 2013). Notably, while the increased expression of CTHRC1 starts only at GOLD stage II, cumulative activation of NTRK2/TrkB is a hallmark throughout progression of COPD in general, suggesting a permanent contribution of NTRK2/TrkB signaling to the aberrant repair response in the peripheral airways during COPD development. This view is further supported by the observation that a disturbed TrkB axis may contribute to experimental pulmonary fibrosis (Avcuogiu et al., 2011).

[0185] With the exception of COMP expression, where clinical deterioration correlates with worsening of bronchial obstruction according to GOLD (see also FIG. 8C), neither increased long-term expression of NTRK2 (see also FIG. 8B), nor of PRRX1 (see also FIG. 8B) or CTHRC1 genes (see also FIG. 8C) demonstrate a comparable short-term impact on bronchial obstruction during the controlled 3-year observational study period. Corresponding results were obtained when assessing the correlation of gene expression with progressive bronchial inflammation: while the expression of all genes favoring mesenchymal repair is increased as a result of intensified bronchitis, significant changes were only found for the PRRX1 and CTHRC1 genes (see also FIGS. 8B and 8C).

[0186] Loss of Structural Integrity of Epithelial Surfaces

[0187] Unexpectedly, the present analysis revealed a very significant downregulation of expression of a group of genes which guide movement, distribution and activation of the cellular cytoskeleton and which, as a result, are likely to profoundly influence structural integrity and barrier function of the mucosal surface. The downregulation of these genes takes place already during establishment of chronic bronchitis, well before the establishment of bronchial obstruction according to GOLD, as also shown in FIG. 4A. The genes closely connected to this development are thymosin beta 15 A (TMSB15A), dipeptidyl-peptidase 6 (DPP6), nudix (nucleoside diphosphate linked moiety X)-type motif 11 (NUDT11), integrin alpha 10 (ITGA10), cystatin E/M (CST6), and PRICKLE2 (data not shown). Notably, the two genes most significantly decreased during progression of COPD, TMSB15A and DPP6, are also significantly downregulated in correlation with symptoms of increased bronchial inflammation (see also FIG. 4B). Beta thymosins are controllers of both composition and sequestration of the actin cytoskeleton (Hannappel, 2007; Huff et al., 2001; Malinda et al., 1999), by that influencing membrane structure, surface stability and cellular phenotype (Husson et al., 2010).

[0188] One of the outcomes of elevated levels of beta thymosins during wound healing seems to be a protection from fibrotic aberrations of repair (De Santis et al., 2011), in part by preventing the expression of a-smooth muscle stress fibers preventing them from a transdifferentiation into myofibroblasts most characteristic for fibrotic tissue development. Currently, little is known about the function of DPP6 in regenerative wound healing. However, DPP6, a member of the S9B family of membrane-bound serine proteases which is lacking any detectable protease activity, has recently been demonstrated to confer membrane stability and controlled outgrowth of cells during nerve development including close control of cell attachment and motility (Lin et al., 2013). Moreover, given its proven association with and control of membrane-bound ion channel expression and activation (Jerng et al., 2012), in particular of voltage-gated potassium channels, expression of DPP6 is also capable of controlling the resting membrane potential (Nadin et al., 2013), thereby controlling both activity and intracellular distribution of the actin cytoskeleton (Mazzochi et al., 2006; Chifflet et al., 2003).

[0189] Combined with the striking reduction of TMSB15A gene expression, the significant decrease of DPP6 expression suggests a severe disturbance of regular movement and distribution of the cellular actin skeleton, reducing physicochemical integrity of the epithelial lipid bilayers. As this occurs already very early in COPD development, this finding could indicate an initiating and possibly predisposing mechanism leading to non-specific surface inflammation.

[0190] Cystatin M/E (CST6), on the other side, is an epithelium-specific protease inhibitor belonging to the cystatin family of secreted cysteine protease inhibitors indispensable for the physiological regulation of protease activity during growth and differentiation of epithelial structures. CST6 is expressed both in dermal and bronchial epithelia where it characterizes the status of functional differentiation (Zeeuwen et al., 2009). Significant downregulation of CST6 has already been shown to cause a marked disturbance of both surface integrity and differentiation status in the dermis of mice (Zeeuwen et al., 2010). Progressive downregulation of CST6 as observed during advancement of COPD is thus likely to destabilize the intricate balance between proteases and protease inhibitors, by that contributing to a loss of surface stability as well as cellular adhesion and differentiation in the regenerating bronchial epithelium. Within this context, significant downregulation of two other genes intricately involved in the regulation of cell adhesion and motility has also been observed, namely of integrin a10 (ITGA10) being part of differentiated mesenchymal structures, and the nudix (nucleoside diphosphate linked moiety X)-type motif hydrolase 11 (NUDT11), capable of hydrolyzing diphosphoinositol polyphosphates derived from cellular lipid bilayer structures, and diadenosine polyphosphates, mostly based on adenosine triphosphate (ATP).

[0191] The consequence of these changes in gene expression is expected to be a disintegration of the epithelial barrier function, probably starting on the cellular level (continuous shear stress within the cellular lipid bilayer due to uncoordinated accumulation and movements of the actin cytoskeleton attached to it), and aggravated by disintegration of the extracellular matrix composition itself. This is supported by the significant increase of gene expression of the KIAA1199 gene during progression of CORD from GOLD stage I to GOLD stage IV (see FIG. 5B). Increased expression of KIAA1199, in addition to mediating cellular attachment and contact inhibition (Tian et al., 2013), has just recently been demonstrated to cause the leakage of endoplasmatic reticulum (ER) contents into the cytosol of cancer cells (Evensen et al., 2013). Moreover, increased expression of KIAA1199 is capable of activating hyaluronidases (HAase), enzymes capable of degrading high-molecular mass hyaluronic acid (HMM-HA), one of the major constituents of the extracellular matrix (Toole, 2004). Biological responses triggered by hyaluronic acid (HA) depend on the HA polymer length. HMM-HA has strong anti-inflammatory properties (Kothapalli et al., 2007), whereas low-molecular-mass HA promotes inflammation and concomitant cellular proliferation (Pure et al., 2009). In support of this view, degradation of HA has been shown to trigger skin inflammation by generation of low molecular weight fragments of HA (Yoshida et al., 2013).

[0192] In line with this, expression of HA synthases (HAS1-3) is not changed during progression of COPD (see FIG. 5G), while the hyaluronidase 2 (HYAL2) gene is upregulated between GOLD stages I and III (see also FIG. 5C). Indeed, the pattern of expression of both HYAL1 and HYAL2 follows the expression pattern of KIAA1199, showing a downregulation during the most intense regenerative phase of repair in COPD progression (chronic bronchitis and CORD GOLD 1). Upregulation of KIAA1199 in turn is synchronous to that of the PLAZA gene (see FIG. 5B) which is a phosphatidylserine-specific phospholipase expressed in macrophages stimulated by typical mechanisms of surface immunity, such as toll-like receptor 4 (TLR4) signaling (Wakahara et al., 2007). Both intensified KIAA1199 and PLA1A expression were found to be connected to short-term worsening of pulmonary function according to GOLD criteria (see also FIG. 5B).

[0193] Decrease of Pro-Angiogenic Mediators During Progression of COPD Effective organ repair involves mechanisms concomitantly directing spatially controlled epithelial, mesenchymal and endothelial outgrowth. However, in contrast to gene functions contributing to epithelial and mesenchymal repair, gene expression promoting angiogenesis and vascular differentiation was found to decrease as soon as chronic bronchitis was present. During development of COPD (GOLD stage I and II), this pattern of gene expression proceeded significantly, as also shown in FIG. 8D. Even the increase of Bex1 and Ghrelin (GHRL) gene expression occurring at GOLD stage I is rather small and insignificant compared to gene functions aimed at the regeneration of epithelial outgrowth, such as DMBT1 and AZGP1. Some of the functions, such as FIBIN (fin bud initiation factor homolog), ESM1 (endothelial cell-specific molecule 1) and ghrelin (GHRL) are known to act, in part, as mediators in the early phases of organ development. For instance, FIBIN takes part in mesodermal lateral plate development (Wakahara et al., 2007) which is crucial for early vasculogenesis (Paffett-Lugassy et al., 2013), ESM1 mediates VEGF-A-dependent signaling (Zhang et al., 2012) and is typically expressed in growing vascular tissue which includes tumor angiogenesis (Zhang et al., 2012; Roudnicky et al., 2013; Chen et al., 2010) and regenerative wound healing (Béchard et al., 2001).

[0194] Ghrelin, on the other hand, is a typical marker of microvascular development (Li et al., 2007; Wang et al., 2012; Rezaeian et al., 2012) being vital for continuous epithelial oxygen and energy supply preventing excessive apoptosis characteristic for emphysema development (Mimae et al., 2013). BEX1 and BEX5 (Brain Expressed, X-Linked 1 and 5) are genes encoding adapter molecules interfering with p75NTR signaling events. p75NTR is one of the two receptors central to nerve growth factor (NGF) signaling. While BEX1 is known to induce sustained cell proliferation under conditions of growth arrest in response to NGF, much less is known about its possible involvement in angiogenesis and vessel formation, although NGF signaling itself is well-known to promote angiogenesis (Cantarella et al., 2002). One possible interaction could be that reduced BEX1 gene expression would increase p75NTR signaling efficacy causing increased endothelial apoptosis, as the blockade of p75NTR signaling significantly decreases endothelial apoptosis (Han et al., 2008; Caporali et al., 2008). The BEX5 promoter, in turn, contains regulatory binding sites for TALI (T-cell acute lymphocytic leukemia 1), a direct transcriptional activator of angiopoietin 2, which is significantly upregulated during angiogenesis (Deleuze et al., 2012). TAL1, however, is downregulated as well during progression of COPD, as also shown in FIG. 8D.

[0195] Stage-Dependent Activation of the Immune Response

[0196] Based on the significant changes of gene expression measured during progression of COPD, four sequential phases of gene expression were distinguished: Phase 1 is characterized by a rapid increase of genes involved in the acute immune response, such as fibrinogen (FGG) (Duvoix et al., 2013; Cockayne et al., 2012), and products of aryl hydrocarbon receptor (AHR) signaling, such as CYP1A1 (cytochrome P450, family 1, subfamily A, polypeptide 1) and CYP1B1 (cytochrome P450, family 1, subfamily B, polypeptide 1) expression, as also shown in FIGS. 5A to 5E. This includes as well an increased expression of carcinoembryonic antigen (CEA)-related cell adhesion molecules (CEACAMs), particularly of the CEACAMS gene (see FIGS. 5A and 5D). At this early stage, still representing chronic bronchitis without significant changes of pulmonary function (COPD “at risk”), expression of genes mediating functions of primarily adaptive immunity, such as RASGRF2 (Ras protein-specific guanine nucleotide-releasing factor 2), KIAA1199 or CXCL3 was not significantly changed (see also FIGS. 5H and 5F). At phase 2 (representing GOLD stage I), expression of these genes remained stable or even decreased to some extent (see FIGS. 4A and 5A), probably reflecting the stabilizing outcome of regenerative repair efforts which was most intense at GOLD stage I (see also FIG. 6A). However, phase 3 which includes GOLD stages II and III was characterized by a significant increase of expression of all genes related to immunity including genes indicating increased AHR signaling, such as CYP1A1, CYP1A2 and CYP1B1 (see also FIGS. 5A, 5E and 5F). The latter ones most likely reflect the impact of cigarette smoking, ail the more as three quarters of the participants were still actives smokers at this stage (see FIG. 3C). Increased gene expression reflecting intensified AHR signaling could be demonstrated in spite of elevated levels of the aryl hydrocarbon receptor repressor (AHRR) gene known to inhibit AHR signaling events, particularly during GOLD stages II and III.

[0197] Nonetheless, short-term analysis of gene expression addressing a development of COPD over a period of 3 years (see also FIGS. 5A and 5D, middle) indicates that the overall impact of AHR signaling on the deterioration of pulmonary function is more important than the additional expression of CEACAMS which, comparable to FGG expression (see also FIG. 5D), seems to reflect the intensity of bronchitis much better. Phase 4 representing GOLD stage IV shows a striking downregulation of the majority of immune-related functions upregulated during earlier phases of COPD development, comparable to the regulation of genes controlling cellular regeneration and differentiation. interestingly, however, this does not apply to the expression of KIAA1199 and RASGRF2 genes which are both upregulated even at GOLD stage IV, the latter one being again capable of influencing cellular movements by inhibition of the actin cytoskeleton (Calvo et al., 2011): RASGRF2 belongs to a group of activators of the GTPase RAS involved as well in the activation of T cells and required for the induction of NF-AT, IL-2 and TNF-α (Ruiz et al., 2007).

[0198] Within this context, the slow yet constant and highly significant upregulation of the guanine-nucleotide exchange factor (GEF) son of sevenless homolog 1 (SOS1) (see FIG. 8A), capable of continually activating RAS, could significantly contribute to the chronic inflammatory process facilitating the bronchial wall scarring characteristic for late stage COPD.

[0199] Members of the carcinoembryonic antigen-related cell adhesion molecule (CEACAM) family serve as cellular receptors for typical gram-negative bacteria frequently colonizing the surface of the human airways, such as Neisseria meningitidis, Haemophilus influenzae and Moraxella catarrhalis expressing opacity (Opa) proteins (Muenzner et al., 2010; Bookwalter et al., 2008; Muenzner et al., 2005). It was recently suggested that non-typable Haemophilus influenzae and Moraxella catarrhalis are able to increase the expression of their respective receptors on host cells (Klaile et al., 2013). However, no correlation between the expression of members of the CEACAM family and COPD was found under the conditions employed in that study. In the present study, only the expression of the CEACAM5 gene was significantly increased up to GOLD stage III, in that following the inflammatory reaction in general, while significantly decreasing afterwards in GOLD stage IV. This does not, however, exclude the aggravation of mucosa! inflammation as a result of a persistent upregulation of CEACAM5, all the more as the expression of CEACAM5 was found to be increased in combination with a growing intensity of bronchial inflammation (see FIG. 5D).

[0200] Conclusions

[0201] Between 2007 and 2012, a controlled prospective pilot trial was conducted in finally 120 volunteers in order to identify metabolic markers indicative of the progression of COPD. By adopting parts of the design of the ECLIPSE trial (Vestbo et al., 2011), the largest and most elaborate study performed thus far to identify clinical markers describing both progress and variability of COPD, the Vienna COPD study combined controlled assessment of validated clinical measures with unsupervised assessment of genome-wide gene transcription in pulmonary tissue representing the focus of COPD pathology (Hogg JC, 2004 (b)). The correlation of gene expression with clinical development was based a) on the extent of non-reversible pulmonary obstruction at visit 1 (according to the Global Initiative for Obstructive Lung Disease; GOLD), b) on the worsening of non-reversible obstruction according to GOLD between visit 1 and 3 (covering a period of three years), and c) on symptoms indicative of an increasing intensity of bronchitis being recorded during structured clinical history at visits 1 and 3.

[0202] This analysis revealed changes of gene expression indicative of six major deviations from regular maintenance of pulmonary structure and defense: (1) Progressive loss of functions guiding epithelial and (2) vascular regeneration combined with (3) persistent and increasing activation of mechanism of fibroproliferative repair, together indicating a transition from regenerative to fibrotic repair during progression of COPD; (4) intensifying bronchial inflammation being antagonized at GOLD stage I when regenerative repair activity is highest, and culminating afterwards at GOLD stages II and III; (5) a complete loss of structural maintenance at GOLD stage IV connected to a finally failing immunity, both suggestive of the formation of scar tissue; and lastly, a rapid and persistent downregulation of functions controlling the intracellular distribution, aggregation and sequestration of actin polymers which form the cytoskeleton (6). The latter finding is of particular interest as the changes in the transcription of the corresponding genes, in particular the downregulation of TMSB15A, DPP6, NUDT11 and PRICKLE2, were already observed at GOLD stage 0 (COPD “at risk”), well before any change of pulmonary function was measurable. This striking loss occurs together with a significant increase of functions determining bronchial inflammation suggesting that these changes might be the first to predispose the bronchi to persistent inflammation. The outcome of such an early and simultaneous downregulation of the TMSB15A, DPP6, NUDT11 and PRICKLE2 genes will be discussed in the following.

[0203] Thymosin beta 15A (TMSB15A) belongs to the group of WH2 (WASP-homologue 2) domain binding proteins which are necessary for the depolymerization of actin filaments during cellular movements (Husson et al., 2010; Hertzog et al., 2004). Formation and rapid movement of actin filaments in turn are indispensable for processes such as cell division, intercalation and cellular extrusion. This applies as well to the regulation of apicobasal cell polarity (Nishimura et al., 2012), and even more important, to the formation and maintenance of tight and adherens junctions (Shen et al., 2005; Calautti et al., 2002). These complex membrane dynamics are not only an answer to external and internal stress, but also part of regular tissue growth and as such energy-dependent. The assembly of the actin skeleton is highly dynamic and creates a layer of epidermal cells acting as an impenetrable fluid-like shield composed of the constantly moving lipid border of the cells (Guillot et al., 2013). Thus, a persistent downregulation of TMSB15A is likely to prevent any fast adaptive arrangement of the surface lipid layers during cellular movements causing repeated perturbations of the epithelial barrier function.

[0204] DPP6, on the other hand, is known to stabilize the membrane potential by acting on membrane-bound potassium channels, and has also a profound impact on the organization of the actin cytoskeleton (Chifflet et al., 2003), supporting the perception of a failing barrier function. The same applies to the downregulation of NUDT11 gene expression. The nucleoside diphosphate linked moiety X (nudix)-type motif 11 (NUDT11) gene encodes a type 3 diphosphoinositol polyphosphate phosphohydrolase which generates energy-rich phosphates essential for vesicle trafficking, maintenance of cell-wall integrity in Saccharomyces and for the mediation of cellular responses to environmental salt stress (Dubois et al., 2002). As the adaptive assembly of F and G actin fibers within the cytoskeleton occurs in seconds, it is easily conceivable that energy-rich diphosphoinositol polyphosphates being integral constituents of any cell membrane will be utilized as rapidly accessible source of energy.

[0205] These findings point towards a synchronized dysregulation of genes necessary for upholding the epithelial barrier. Moreover, the downregulation of the PRICKLE2 gene was also shown to be vital for the formation of polarized epithelial layers during mouse embryogenesis (Tao et al., 2012). Decreased expression of all four genes (i.e., TMSB15A, DPP6, NUDT11 and PRICKLE2), however, was associated with significantly increased bronchial inflammation, suggesting a functional correlation between the downregulation of genes that guide functionally interrelated features of cytoskeleton assembly with the activation of bronchitis. This sheds a new light on the progression of bronchial inflammation as it indicates a direct connection between the loss of a protective epithelial shield and the aggravation of chronic bronchitis. Based on the physicochemical nature of such an effect, penetration of the epithelial membranes by any potential antigen or allergen is likely to be enhanced, particularly during intensified repair due to repeated smoke-induced damage or following viral infections. This could not only explain the remarkable heterogeneity of inflammatory conditions characteristic for COPD, but also the observation that the capacity to achieve intense cellular regeneration in spite of ongoing inflammation might be helpful in suppressing pro-inflammatory gene expression.

[0206] This view is further supported by the significant downregulation of the protease inhibitor cystatin M/E (CST6) during progression of COPD (see also FIG. 4A). CST6 is known to control the homeostasis of the stratum corneum, its deficiency in mice causing severe ichthyosis and neonatal lethality (Zeeuwen et al., 2009). The progressive loss of a protease inhibitor in later phases of COPD known to preserve the integrity of epithelial structures wiii most likely contribute to a failure of the protective barrier function, not only by a disintegration of the epithelial layer but also by facilitating the breakdown of the matrix itself.

[0207] In this context, the strong upregulation of the KIAA1199 gene which has been demonstrated to significantly increase the activity of matrix hyaluronidases, is probably equally important, as this upregulation is directly associated with a significant worsening of lung function, even within the relatively short observational period of the present study (see also FIG. 5B). It has recently been shown that matrix structures containing large amounts of high molecular mass hyaluronan as well as the inhibition of hyaluronidase activity protect against both inflammation and cancer progression (Tian et al., 2013). In summary, these findings provide the first conclusive evidence for a progressive breakdown of bronchial surface integrity during the course of COPD development causing growing non-specific bronchial inflammation that varies with frequency and intensity of the physicochemical assaults attacking the bronchial surfaces.

[0208] According to results described herein, the response to these assaults is a slow progressive scarring process in the peripheral bronchi, whereby the combined upregulation of CTHRC1, SOS1 and NTRK2 genes (see also FIG. 8A) is likely to indicate mechanisms of preferentially mesenchymal wound healing while the stage dependent expression of the PRRX1 and COMP genes suggests their participation in regular organ repair as well demonstrating the ambiguity between regular matrix support during regenerative repair and scar formation as a result of a progressive failure of the organ's regenerative repair capacity.

[0209] This fits well to the progressive downregulation of genes mainly controlling functions of regenerative growth of the vascular tree as demonstrated by the concomitant decrease of the expression of FIBIN, TAL1, BEX1/5, and Ghreiin (GHRL) genes (see also FIG. 8D). Here again, the increasing capacity of the peripheral lung to employ mechanisms of preferentially regenerative repair during GOLD stage I becomes evident as BEX1 and GHRL increase at this stage while progressively decreasing during further progression of COPD.

[0210] Thus, in the COPD AUVA study, the clinical progression of COPD has been successfully correlated with the biological analysis of gene expression in pulmonary tissue. In particular, it has been demonstrated that the expression of the genes KIAA1199, DMBT1, ELFS, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and COMP is increased in pulmonary tissue samples from subjects prone to develop progressive COPD, while the expression of the genes TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TAL1, FIBIN, BEXS, BEX1, ESM1 and GHRL is decreased in pulmonary tissue samples from subjects prone to develop progressive COPD, as compared to the expression of the corresponding genes in pulmonary tissue samples from healthy subjects. These molecular biomarkers can thus be used for assessing the susceptibility/proneness of a subject to develop progressive COPD in accordance with the present invention, particularly in the method of the second aspect of the invention. Moreover, it has also been demonstrated that the expression of the genes DMBT1, ELFS, AZGP1, PRRX1, AQP3, SFN, GPR110, GDF15, RASGRF2, RND1, FGG, CEACAM5, AHRR, CXCL3, CYP1A1, CYP1B1, CYP1A2, NTRK2 and COMP is increased in pulmonary tissue samples from subjects suffering from or prone to suffer from stable COPD, while the expression of the genes KIAA1199, TMSB15A, DPP6, SLC51B, NUDT11, PLA1A, HYAL2, CST6, ITGA10, CTHRC1, TAL1, FIBIN, BEXS, BEX1, ESM1 and GHRL is decreased in pulmonary tissue samples from subjects suffering from or prone to suffer from stable COPD, as compared to the expression of the corresponding genes in pulmonary tissue samples from healthy subjects, indicating that these biomarkers are suitable for diagnosing stable COPD or assessing the susceptibility of a subject to develop stable COPD in accordance with the invention, particularly in the method of the third aspect of the invention.

REFERENCES

[0211] Agarwal P, et al. J Biol Chem. 2012; 287(27):22549-59. doi:10.1074/jbc.M111.335935.

[0212] Agarwal P, et al. Matrix Biol. 2013; 32(6):325-31. doi:10.1016/j.matbio.2013.02.010.

[0213] Avcuoglu S, et al. Am J Respir Cell Mo! Biol. 2011; 45(4):768-80. doi:10.1165!rcmb.2010-01950C.

[0214] Béchard D, et al. J Biol Chem. 2001; 276(51):48341-9.

[0215] Bookwalter J E, et al. Infect Immun. 2008; 76(1):48-55.

[0216] Butt A Q, et al. J Biol Chem. 2012; 287(46):38665-79. doi:10.1074/jbc.M112.367490.

[0217] Calautti E, et al. J Cell Biol. 2002; 156:137-48.

[0218] Calvo F, et al. Nat Cell Biol. 2011; 13(7):819-26. doi:10.1038/ncb2271.

[0219] Cantarella G, et at. FASEB J. 2002; 16(10):1307-9.

[0220] Caporali A, et al. Circ Res. 2008; 103(2):e15-26. doi:10.1161/CIRCRESAHA.108.177386.

[0221] Chavez-Muñoz C, et al. J Cell Biochem. 2012; 113(8):2622-32. doi:10.1002/jcb.24137.

[0222] Chen L Y, et al. J Int Med Res. 2010; 38(2):498-510.

[0223] Chifflet S, et al. Exp Cell Res. 2003; 282(1):1-13.

[0224] Cockayne D A, et al. PLoS One. 2012; 7(6):e38629. doi:10.1371/journal.pone.0038629.

[0225] Cole S P C, et al. Monoclonal Antibodies and Cancer Therapy. 1985; 27:77-96.

[0226] De Santis M, et al. Respir Res. 2011; 12:22. doi:10.1186/1465-9921-12-22.

[0227] Deleuze V, et al. PLoS One. 2012; 7(7):e40484. doi:10.1371/journal.pone.0040484.

[0228] Diegelmann J, et al. J Biol Chem. 2012; 287(1):286-98. doi:10.1074/jbc.M111.294355.

[0229] Ding C, et al. J Biochem Mol Biol. 2004; 37(1):1-10.

[0230] Dubois E, et al. J Biol Chem. 2002; 277:23755-63.

[0231] Duvoix A, et al. Thorax. 2013; 68(7):670-6. doi:10.1136/thoraxjn1-2012-201871.

[0232] Evensen N A, et al. J Natl Cancer Inst. 2013; 105(18):1402-16. doi:10.1093/jnci/djt224.

[0233] Frisch S M, et al. J Cell Sci. 2013; 126(Pt1):21-9. doi:10.1242/jcs.120907.

[0234] Gentleman R, et al. Genome Biology. 2004; 5:R80.

[0235] Green, M R et al. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press. Fourth Edition. 2012. ISBN: 978-1936113422.

[0236] Guillot C, et al. Science. 2013; 340:1185-9.

[0237] Halbert R J, et al. Eur Respir J. 2006; 28:523-32.

[0238] Han Y, et al. Biochem Biophys Res Commun. 2008; 366(3):685-91.

[0239] Han M K, et al. Am J Respir Crit Care Med. 2010; 182:598-604.

[0240] Hannappel E. Ann NY Acad Sci. 2007; 1112:21-37. doi:10.1196/annals.1415.018.

[0241] Harlow E, et al. Using Antibodies: A Laboratory Manual. Cold Spring Harbor Laboratory Press. 1998. ISBN: 978-0879695446.

[0242] Harnoncourt K, et al. Osterreich Arzteztg. 1982; 37:1640-2.

[0243] Hertzog M, et al. Cell. 2004; 117:611-623.

[0244] Hogg J C, et al. N Engl J Med. 2004; 350:2645-2653. (a)

[0245] Hogg J C. Lancet. 2004; 364:709-721. (b)

[0246] Huff T, et al. Int J Biochem Cell Biol. 2001; 33:205-220. doi:10.1016/S1357-2725(00)00087-X.

[0247] Hurst J R, et al. N Engl J Med. 2010; 363:1128-1138.

[0248] Husson C, et at. Ann NYAcad Sci. 2010; 1194:44-52. doi:10.1111/j.1749-6632.2010.05473.x.

[0249] Jerng H H, et al. PLoS One. 2012; 7(6):e38205. doi:10.1371/journal.pone.0038205.

[0250] Jiang F, et al. Exp Biol Med (Maywood). 2008; 233(3):286=96. doi:10.3181/0707-RM-177.

[0251] Johnson W E, et al. Biostatistics. 2007; 8(1):118-127.

[0252] Kaemmerer E, et al. Histopathology. 2012; 60(4):561-9.doi:10.1111/j.1365-2559.2011.04159.x.

[0253] Kim N H, et al. J Invest Dermatol. 2010; 130(9):2231-9. doi:10.1038/jid.2010.99.

[0254] Klaile E, et al. Respir Res. 2013; 14:85. doi:10.1186/1465-9921-14-85.

[0255] Köhler G, et al. Nature. 1975; 256(5517):495-7.

[0256] Kong B, et al. Oncogene. 2010; 29(37):5146-58. doi:10.1038/onc.2010.258.

[0257] Kothapalli D, et al. J Cell Biol. 2007; 176:535-44.

[0258] Kozbor D, et al. Immunol Today. 1983; 4(3):72-9.

[0259] Lei W, et al. Am J Physiol Lung Cell Mol Physiol. 2007; 293(5):L1359-68.

[0260] Li A, et at. Biochem Biophys Res Commun. 2007; 353(2):238-43.

[0261] Lin L, et al. Nat Commun. 2013; 4:2270. doi:10.1038/ncomms3270.

[0262] Malinda K M, et al. J Invest Dermatol. 1999; 113(3):364-8. doi:10.1046/j.1523-1747.1999.00708.x.

[0263] Mazzochi C, et al. Am J Physiol Renal Physiol. 2006; 291(6):F1113-22.

[0264] Medina A, et al. Mol Cell Biochem. 2007; 305:255-64.

[0265] Metzger D E, et al. Dev Biol. 2008; 320(1):149-60. doi:10.1016/j.ydbio.2008.04.038.

[0266] Mimae T, et al. Thorax. 2013. doi:10.1136/thoraxjn1-2013-203867.

[0267] Muenzner P, et al. J Cell Biol. 2005; 170(5):825-36. doi:10.1083/jcb.200412151

[0268] Muenzner P, et al. Science. 2010; 329(5996):1197-201. doi:10.1126/science.1190892.

[0269] Murray C J L, et al. Lancet. 1997; 349:1498-504.

[0270] Nadin B M, et al. PLoS One. 2013; 8(4):e60831. doi:10.1371/journal.pone.0060831.

[0271] Nakahigashi K, et al. J Invest Dermatol. 2011; 131(4):865-73. doi:10.1038/jid.2010.395.

[0272] Nishimura T, et al. Cell. 2012; 149(5):1084-97. doi:10.1016/j.ce11.2012.04.021.

[0273] Ocaña O H, et al. Cancer Cell. 2012; 22(6):709-24. doi:10.1016/j.ccr.2012.10.012.

[0274] Paffett-Lugassy N, et al. Nat Cell Biol. 2013; 15(11):1362-9. doi:10.1038/ncb2862.

[0275] Pauwels, R A et al. Am J Respir Crit Care Med. 2001; 163(5):1256-76.

[0276] Price D, et al. Prim Care Respir J. 2011; 20(1):15-22. doi:10.4104/perj.2010.00060.

[0277] Pure E, et al. Cell Signal. 2009; 21(5):651-5. doi:10.1016/j.cellsig.2009.01.024.

[0278] Rabe K F, et al. Am J Respir Crit Care Med. 2007; 176(6):532-55.

[0279] Rezaeian F, et al. Am J Physiol Heart Circ Physiol. 2012; 302(3):H603-10. doi:10.1152/ajpheart.00390.2010.

[0280] Roudnicky F, et al. Cancer Res. 2013; 73(3):1097-106. doi:10.1158/0008-5472.CAN-12-1855.

[0281] Ruiz S, et al. Mol Cell Biol. 2007; 27(23):8127-42.

[0282] Shen L, et al. Mol Bio Cell. 2005; 16:3919-36.

[0283] Smyth G K. Iimma: linear models for microarray data. In: Gentleman R, et al. Bioinformatics and Computational Biology Solutions using R and Bioconductor. 2005. Springer, N.Y., pages 397-420.

[0284] Spector I, et al. Am J Pathol. 2013; 182(3):905-16. doi:10.1016/j.ajpath.2012.11.004.

[0285] Tao H, et al. Dev Biol. 2012; 364(2):138-48. doi:10.1016/j.ydbio.2012.01.025.

[0286] Tian X, et al. Nature. 2013; 499(7458):346-9. doi:10.1038/nature12234.

[0287] Toole B P. Nat Rev Cancer. 2004; 4(7):528-39.

[0288] US Burden of Disease Collaborators. JAMA. 2013; 310(6):591-608.

[0289] Vestbo J, et al. N Engl J Med. 2011; 365(13):1184-92.

[0290] Vestbo J, et al. Am J Respir Crit Care Med. 2013; 187(4):347-65. doi:10.1164/rccm.201204-0596PP.

[0291] Wang L, et al. Peptides. 2012; 33(1):92-100. doi:10.1016/j.peptides.2011.11.001.

[0292] Wakahara T, et al. Dev Biol. 2007; 303(2):527-35.

[0293] Wedzicha J A. Thorax. 2000; 55:631-632.

[0294] Wu Q, et al. Innate Immun. 2012; 18(4):617-26. doi:10.1177/1753425911429837.

[0295] Yaniw D, et al. Cell Res. 2005; 15(6):423-9.

[0296] Yoshida H, et al. Proc Natl Acad Sci USA. 2013; 110(14):5612-7. doi:10.1073/pnas.1215432110.

[0297] Zeeuwen P L J M, et al. J Invest Dermatol. 2009; 129:1327-38. doi:10.1038/jid.2009.40.

[0298] Zeeuwen P L J M, et al. FASEB J. 2010; 24:3744-55. doi:10.1096/fj.10-155879.

[0299] Zhang S M, et al. Biotech Histochem. 2012; 87(3):172-8. doi:10.3109/10520295.2011.577754.