Dynamic gene expressions of peripheral blood mononuclear cells in patients with acute exacerbation of chronic obstructive pulmonary disease: a preliminary study.

INTRODUCTION: Acute exacerbation of chronic obstructive pulmonary disease (AECOPD) is a serious event that is responsible for the progress of the disease, increases in medical costs and high mortality.
METHODS: The aim of the present study was to identify AECOPD-specific biomarkers by evaluating the dynamic gene expression profiling of peripheral blood mononuclear cells (PBMCs) from patients with AECOPD on days 1, 3 and 10 after hospital admission and to compare the derived data with data from healthy controls or patients with stable COPD.
RESULTS: We found that 14 genes were co-differentially upregulated and 2 downregulated greater than 10-fold in patients with COPD or AECOPD compared with the healthy individuals. Eight co-differentially upregulated genes and six downregulated genes were identified as a panel of AECOPD-specific genes. Downregulation of TCF7 in PBMCs was found to be associated with the severity of COPD. Dynamic changes of Aminolevulinate-delta-synthase 2 and carbonic anhydrase I had similar patterns of Digital Evaluation Score System scores and may serve as potential genes of interest during the course of AECOPD.
CONCLUSION: Thus, our findings indicate a panel of altered gene expression patterns in PBMCs that can be used as AECOPD-specific dynamic biomarkers to monitor the course of AECOPD.

Introduction

Chronic obstructive pulmonary disease (COPD) is an inflammation-based syndrome characterized by progressive deterioration of pulmonary function and increasing airway obstruction [1]. COPD is a major and growing public health burden, ranking as the fourth leading cause of death in the world [2]. In China, it is the fourth leading cause of mortality in urban areas and the third leading cause in rural areas [3]. Patients with COPD often experience a sudden deterioration, termed acute exacerbations of chronic obstructive pulmonary disease (AECOPD), along with a progressive decline in lung function; AECOPD becomes more frequent and severe when the severity of disease increases [4,5]. There is a great need for early and sensitive diagnosis and novel therapeutic targets for the disease, especially for patients with AECOPD in whom COPD is diagnosed in the late phase of disease, when they have significant or irreversible impairment [6].

The progress of COPD is accelerated by the occurrence of the exacerbation induced by multiple factors, including infection. AECOPD is a serious event that is related to decreased health status, increased medical and social costs and increased mortality [7]. Inflammatory cells (for example, lymphocytes, monocytes or macrophages, and their products) could interact with each other or with structural cells in the airways and the lung parenchymal and pulmonary vasculature, leading to the worsening of COPD [8]. Increased numbers of CD8+ lymphocytes were suggested as one of COPD’s characteristics, being present only in smokers who develop the disease [9]. Increased pulmonary inflammatory mediators in patients with COPD could attract inflammatory cells from the circulation, amplify the inflammatory process and induce structural changes [9].

Peripheral blood mononuclear cells (PBMCs) act as a critical component in the immune system to fight infection and adapt to intruders and play an important role in the development of AECOPD. Gene expression profiles of PBMCs were found to be disease-specific and associated with severity [10]. PBMC samples were suggested as easy to gather and important to the discovery of biomarkers for diagnosis and therapeutic management of COPD [11,12], although gene expression changes in lung tissues were noted to be associated with COPD [13-15]. The aim of the present study was to determine AECOPD-specific biomarkers of PBMCs using the concept of clinical bioinformatics and integrating genomics, bioinformatics, clinical informatics and systems biology [16-18]. We translated all clinical measures, including patient complaints, history, therapies, clinical symptoms and signs, physician’s examinations, biochemical analyses, imaging profiles, pathologies and other measurements, into digital format using a digital evaluation scoring system. PBMCs were isolated from healthy volunteers and patients with stable COPD or AECOPD, and we investigated the disease specificity that we inferred from clinical informatics analysis to search for COPD- or AECOPD-specific genes and dynamic biomarkers for AECOPD.

Material and methods

Patient population

The present study was approved by the Ethical Evaluation Committee of Zhongshan Hospital and designed using a case–control approach. From among 220 candidates comprising blood donors (60 healthy controls), inpatients (80 patients with AECOPD) and outpatients (80 patients with stable COPD) in Zhongshan Hospital, patients with AECOPD, patients with stable COPD and healthy controls matched for age and sex were recruited into the study between October 2011 and March 2012. The inclusion criteria for patients with COPD were as follows: (1) forced expiratory volume in 1 second (FEV1) <80% of predicted value adjusted for age, weight and height, and (2) an improvement in FEV1 following bronchodilator inhalation <12% of baseline FEV1. Patients with asthma who had a persistent airflow obstruction were excluded. Stable COPD was defined according to American Thoracic Society/European Respiratory Society consensus criteria as no requirement for increased treatment above maintenance therapy, other than bronchodilators, for 30 days [1]. AECOPD was the reason for hospital admission and was characterized as a worsening of the patient’s respiratory symptoms that was beyond normal day-to-day variations and led to a change in medication [4,19]. Healthy controls enrolled were blood donors at Zhongshan Hospital. Subjects with respiratory diseases, or any family history of lung disease, were excluded. PBMCs were harvested once from healthy controls and patients with stable COPD, as well as from patients with AECOPD, on the admission day and 3 and 10 days after the admission. Informed consent was given by the subjects themselves before they underwent lung function tests, high-resolution computed tomography and blood collection. The time points used in the present study were selected on the basis of our previous study for collecting plasma samples from healthy controls and from patients with stable COPD or AECOPD. The details of the study design are explained in Figure 1.

Figure 1

Details of the study design. Healthy volunteers and patients with stable chronic obstructive pulmonary disease (sCOPD) or acute exacerbation of COPD (AECOPD) at day 1 (D1), day 3 (D3) or day 10 (D10) of hospital admission of hospital were recruited into the present study according to the criteria stated in the text. All clinical information was collected and transferred into the clinical informatics database using the Digital Evaluation Score System. mRNAs of peripheral blood monocytes were harvested, and gene expression profiles were measured by human gene expression array and subjected to bioinformatics analysis. AECOPD-specific biomarkers were selected by integrating gene functional networks and profiles with clinical informatics data.

Digital evaluation score system

The Digital Evaluation Score System (DESS) is a score index used to translate clinical descriptions and information into clinical informatics, as described previously [20]. Using this instrument, we took into account patient symptoms and signs, biochemical analyses and clinical imaging for patients with stable COPD or AECOPD. Briefly, for the assessment of severity, each component was assigned a score of 0, 1, 2 or 4. The score of 4 as the maximum value indicates far above normal range or much severer condition, and 0 as the minimum value indicates within normal physiological range. After compiling patient data, we added the points for each variable. The DESS scores ranged from 0 to 256 points, with a higher score indicating a severer condition. Patients were scored on the day when their blood samples were collected.

Isolation of PBMC RNA

PBMCs were isolated by using BD Vacutainer CPT cell preparation tubes (Becton Dickinson, Franklin Lakes, NJ, USA) according to the manufacturer’s instructions. Approximately 4 ml of whole blood was collected from each subject. Following centrifugation, cells were lysed for RNA isolation. DNase-free total RNA preparation was performed using TRIzol reagent (Life Technologies, Carlsbad, CA, USA) and the RNeasy kit (QIAGEN, Valencia, CA, USA) according to the manufacturers’ recommendations. RNA concentrations were determined by using a NanoDrop ND-1000 spectrophotometer (NanoDrop, Wilmington, DE, USA). RNA quality was assessed on an Agilent 2100 Bioanalyzer (Agilent Technologies, Santa Clara, CA, USA), and samples with an RNA integrity number >6.0 were used.

Microarray analysis

The Human 12×135K Gene Expression Array (Roche NimbleGen Systems, Madison, WI, USA), with about 45,000+ human genes and transcripts represented with public domain annotations, was applied for this study. Sample labeling and array hybridization were performed according to the one-color microarray-based gene expression analysis protocol (Roche NimbleGen Systems). Double-stranded cDNA (ds-cDNA) was synthesized from 5 μg of total RNA using an Invitrogen SuperScript reverse transcriptase ds-cDNA synthesis kit (Life Technologies) in the presence of 100 pmol oligo(dT) primers. ds-cDNA was cleaned and labeled in accordance with the NimbleGen gene expression analysis protocol. Briefly, ds-cDNA was incubated with 4 μg of RNase A at 37°C for 10 minutes and cleaned using phenol:chloroform:isoamyl alcohol, followed by ice-cold absolute ethanol precipitation. The purified cDNA was quantified using the NanoDrop ND-1000 spectrophotometer. For Cy3 labeling of cDNA, the NimbleGen one-color DNA labeling kit was used according to the manufacturer’s guidelines as detailed in its gene expression analysis protocol. One microgram of ds-cDNA was incubated for 10 minutes at 98°C with 1 optical density of Cy3-9mer primer. Next, 100 pmol of deoxynucleoside triphosphates and 100 U of the Klenow fragment (New England Biolabs, Ipswich, MA, USA) were added, and the mix was incubated at 37°C for 2 hours. The reaction was stopped by adding 0.1 vol of 0.5 M ethylenediaminetetraacetic acid, and the labeled ds-cDNA was purified by isopropanol/ethanol precipitation. Microarrays were hybridized at 42°C for 16 to 20 hours with 4 μg of Cy3-labeled ds-cDNA in NimbleGen hybridization buffer/hybridization component A in a hybridization chamber. Following hybridization, washing was performed using the NimbleGen wash buffer kit. After being washed in an ozone-free environment, the slides were scanned using an Axon GenePix 4000B microarray scanner (Molecular Devices, Sunnyvale, CA, USA).

Data analysis

For clinical data, all values were expressed as mean ± SE. Analyses were performed using SPSS software (SPSS 18.0; SPSS, Chicago, IL, USA). For microarray analysis, slides were scanned at 5 μm/pixel resolution using the Axon GenePix 4000B microarray scanner piloted by GenePix Pro 6.0 software (Molecular Devices). Scanned images (in TIFF file format) were then imported into NimbleScan software (version 2.5) files for grid alignment and expression data analysis. Expression data were normalized through quantile normalization and the Robust Multi-array Average (RMA) algorithm included in the NimbleScan software. The probe-level (*_norm_RMA.pair) files and gene-level (*_RMA.calls) files were generated after normalization. All gene-level files were imported into GeneSpring GX software (version 11.5.1; Agilent Technologies) for further analysis. Differentially expressed genes between two samples were identified by fold change filtering. Hierarchical clustering was performed using the GeneSpring GX software. Gene Ontology (GO) database analysis and pathway analysis were performed using the standard enrichment computation method. The GO database covers three domains: biological process, cellular component and molecular function. Fisher’s exact test was used to find more overlaps between the descriptive list and the GO annotation list than would be expected by chance. The P-value denoted the significance of GO term enrichment in the descriptive genes. The gene expression data are publicly available in the Gene Expression Omnibus database [GEO:GSE60399] [21].

Results

Clinical informatics analysis

Clinical phenotypes are described in Table 1, including age, sex, smoking status, lung function test results and emphysema scores of the subjects. Control subjects were nonsmokers, and patients with stable COPD or AECOPD were ex-smokers. Because of the severity of disease, lung function tests were not performed at the onset of AECOPD; however, the baseline FEV1/forced vital capacity (FVC%) and FEV1/predicted percentage of patients with AECOPD were similar to those of patients with stable COPD. In addition, there was no significant difference in the extent of emphysema between patients with stable COPD and those with AECOPD (P = 0.47). DESS scores of subjects from each group are shown in Additional file 1. DESS values of patients with stable COPD or AECOPD were significantly higher than those of control subjects (P < 0.01), as shown in Table 2. DESS scores represented the severity of COPD and declined as the patient’s condition improved. DESS values of patients with AECOPD on day 1 of hospital admission (AE-1) were significantly higher than those on day 3 (AE-3) and day 10 (AE-10) (P < 0.05 and P < 0.01, respectively) (Table 2).

Table 1

Clinical phenotypes of healthy controls, patients with stable chronic obstructive pulmonary disease and patients with acute exacerbation of chronic obstructive pulmonary disease

Groups	Subject no.	Age (yr)	Smoking status	FEV 1 /FVC%	FEV 1 /pred%	Goddard emphysema score
Control	1	56	Nonsmoker	75	85	0
	2	53	Nonsmoker	80	87	0
	3	62	Nonsmoker	77	91	0
	4	68	Nonsmoker	81	83	0
	5	58	Nonsmoker	79	81	0
	6	67	Nonsmoker	76	90	0
Mean ± SE		60.7 ± 2.5		78.0 ± 1.0	86.2 ± 1.6	0.0 ± 0.0
Stable COPD	1	71	Ex-smoker	57	47	10
	2	75	Ex-smoker	46	66	6
	3	61	Ex-smoker	46	47	8
	4	57	Ex-smoker	38	29	12
	5	59	Ex-smoker	67	66	7
	6	53	Ex-smoker	29	36	11
Mean ± SE		62.7 ± 3.5		47.2 ± 5.5	48.5 ± 6.2	9.0 ± 1.0
AECOPD	1	77	Ex-smoker	40	42	10
	2	72	Ex-smoker	36	27	11
	3	65	Ex-smoker	28	33	16
	4	56	Ex-smoker	48	61	6
	5	61	Ex-smoker	69	55	4
	6	67	Ex-smoker	56	60	8
Mean ± SE		66.3 ± 3.1		46.2 ± 6.0	46.3 ± 5.9	9.2 ± 1.7

aAECOPD, Acute exacerbation of chronic obstructive pulmonary disease; COPD, Chronic obstructive pulmonary disease; FEV1, Forced expiratory volume in 1 second; FVC, Forced vital capacity; pred, Prediction. Data represent information gathered on days 1, 3 and 10 of hospital admission.

Table 2

Digital evaluation score system scores

	DESS scores
Patient no.	Control	Stable COPD	AE-1	AE-3	AE-10
1	0	30	100	78	43
2	4	27	81	66	46
3	8	35	86	76	36
4	4	55	70	51	30
5	3	38	80	71	35
6	0	47	97	81	30
Mean ± SE	3.2 ± 1.2	38.7 ± 4.3	85.7 ± 4.6	70.5 ± 4.5	36.7 ± 2.7

aAE-1, Day 1 of hospital admission; AE-3, Day 3 of hospital admission; AE-10, Day 10 of hospital admission; COPD, Chronic obstructive pulmonary disease; DESS, Digital evaluation score system.

Gene expression profiles

The quality of the genetic data obtained after filtering and the distribution of data sets were assessed and visualized by creating box plots, which showed that there were no significant differences in the distributions of log2 ratios among the groups (see Additional file 2: Figure S1). The variation or reproducibility of gene expression between arrays of different groups was visualized and assessed by creating scatterplots, which are shown in Figure 2. There was a significant variation in gene arrays between healthy controls and patients with stable COPD or AECOPD (Figures 2A to 2D) and between patients with stable COPD and AECOPD (Figures 2E to 2G). The variation in gene array data at AE-1 and AE-3 was significantly different from that at AE-10 (Figures 2I and 2J), whereas there was no difference between AE-1 and AE-3 (Figure 2H). The results of hierarchical clustering showed gene expression profiles similar to those revealed by the scatterplots shown in Figure S2 of Additional file 2.

Figure 2

Scatterplots showing variations in gene expression profiles. Scatterplots of peripheral blood monocytes between patients with stable chronic obstructive pulmonary disease (Stable COPD) (A), acute exacerbation of chronic obstructive pulmonary disease at day 1 of hospital admission (AECOPD-D1) (B), AECOPD at day 3 of hospital admission (AECOPD-D3) (C) or AECOPD at day 10 of hospital admission (AECOPD-D10) (D) compared with healthy controls. Scatterplots also illustrate variations between AECOPD-D1 (E), AECOPD-D3 (F) or AECOPD-D10 (G) and stable COPD; between AECOPD-D3 (H) or AECOPD-D10 (I) with AECOPD-D1; and between AECOPD-D3 and AECOPD-D10 (J).

To identify differentially expressed genes, a fold change filtering between each group pair was performed with a threshold fold change ≥2.0. There were ten comparison pairs with information for fold changes and regulation (that is, SEQ-ID, log fold change, log or absolute fold change, or regulation), normalized intensities or annotations (that is, GENE_NAME, synonyms, description, NCBI_GENE_ID, chromosome, GO, UniGene ID, The Institute of Genomic Research Database-TDB (TIGRID) or Ensembl ID), as shown in Additional file 3. Table 3 shows the number of genes overexpressed more than twofold, (for example, 4,508, 3,899, 4,167 and 3,488 genes of stable, AE-1, AE-3 and AE-10, respectively, above controls; 4,067, 5,063 or 5,451 genes of AE-1, AE-3 and AE-10, respectively, above stable COPD; 586 genes of AE-3 above AE-1; and 1,735 and 1,706 genes of AE-10, respectively, above AE-1 and AE-3). Tables 4, 5 and 6, respectively, list the genes overexpressed (above controls) in PBMCs from patients with stable COPD, AE-1, AE-3 or AE-10 by more than 30-fold (Table 4), between 20- and 30-fold (Table 5) and between 15- and 20-fold (Table 6). Tables 7, 8 and 9 list the genes overexpressed (above patients with stable COPD) in PBMCs from patients with AE-1, AE-3 or AE-10 by more than 30-fold (Table 7), between 20- and 30-fold (Table 8) and between 15- and 20-fold. Table 10 presents upregulated genes in PBMCs of patients at AE-1, AE-3 or AE-10.

Table 3

Genes upregulated in peripheral blood mononuclear cells

	Fold changes in upregulated genes ( n )
Comparisons	>2	>5	>8	>10	>15	>20	>30	>50	>100
Stable vs Con	4,508	671	217	145	49	27	9	1	0
AE-1 vs Con	3,899	734	334	221	136	86	40	18	3
AE-3 vs Con	4,167	742	358	259	149	97	51	17	5
AE-10 vs Con	3,488	677	331	238	116	74	35	10	1
AE-1 vs Stable	4,067	389	135	80	36	21	9	3	1
AE-3 vs Stable	5,063	620	221	146	56	24	10	1	0
AE-10 vs Stable	5,451	534	178	117	56	33	14	1	0
AE-3 vs AE-1	586	8	2	2	0	0	0	0	0
AE-10 vs AE-1	1,735	164	55	26	10	4	1	0	0
AE-10 vs AE-3	1,706	156	49	29	2	2	1	0	0

aData are number of upregulated genes expressed in peripheral blood mononuclear cells of healthy controls (Con) or of patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on hospital admission day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10).

Table 4

Genes upregulated >30-fold in peripheral blood mononuclear cells

Stable vs control		AE-1 vs control		AE-3 vs control		AE-10 vs control
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
31.7	REXO1L2P	30.3	HP	30.1	FOS	30.8	EMP2
33.0	DEFA1	30.5	LOC152573	30.6	BPIL1	31.0	SEPP1
33.3	DUB3	31.2	INHBA	31.0	ARG1	31.0	FOLR1
37.2	LOC402207	31.4	COL6A3	31.6	N/A	31.1	GPX3
37.3	DUB3	32.4	MPO	31.9	LOC152573	31.2	SFTPB
40.5	LOC402110	32.6	ELF3	32.5	COL6A3	31.4	S100A14
43.1	LOC653600	34.4	CLDN4	32.9	TIMP3	33.1	FOLR1
43.5	N/A	34.9	DCN	33.5	FOS	33.4	CDH5
50.7	MGC45438	35.7	CTGF	34.4	KRT19	34.9	CAV1
		35.7	MMP2	34.7	INHBA	35.4	DLC1
		36.2	MFAP4	35.2	HP	35.6	FOSB
		37.1	EPB42	35.6	CD177	36.1	KRT19
		37.2	H19	36.5	LCN2	36.4	SUSD2
		37.3	ATP1B1	36.9	CTGF	36.9	FN1
		37.5	INHBA	37.9	MMP8	37.2	ADH1C
		38.0	AZU1	38.3	ORM1	37.2	RNASE1
		38.5	LCN2	38.8	ELF3	37.3	IL1RL1
		39.6	CEACAM8	38.9	DCN	41.1	FOLR1
		40.3	CALCA	39.0	CTSG	41.3	DHCR24
		41.4	LOC387763	39.1	CLDN4	41.3	LOC387763
		42.2	CEACAM3	39.3	CALCA	42.0	ADH1B
		45.9	UNQ473	40.0	DCN	43.6	LAMA3
		54.0	BPIL1	40.1	FOSB	45.0	GPX3
		56.2	FN1	41.1	ATP1B1	47.9	DCN
		56.7	CEACAM5	41.6	MFAP4	49.1	EPAS1
		58.4	MMP8	41.8	FN1	50.9	CNN3
		65.0	CALCA	42.0	MMP2	51.5	DCN
		66.3	BPI	42.0	GPR97	54.5	LOC653509
		68.7	DEFA1	42.2	INHBA	56.2	CXCL2
		72.3	COL1A2	45.5	AZU1	58.2	MGC45438
		77.2	CA1	46.0	BPI	58.5	CYP4B1
		80.2	PLUNC	46.4	LOC387763	59.3	CTGF
		83.0	CEACAM1	46.6	MPO	75.8	GPRC5A
		83.9	DEFA4	50.0	HP	88.9	TIMP3
		85.0	COL3A1	50.7	ORM2	149.5	MFAP4
		96.1	DEFA1	53.1	UNQ473
		99.4	CEACAM5	57.8	AQP9
		101.2	CEACAM1	59.6	CEACAM5
		115.8	LOC653600	59.6	BPIL1
		140.3	DEFA4	61.0	CEACAM1
				62.8	DEFA1
				66.5	CEACAM1
				72.6	DEFA4
				82.5	PLUNC
				86.7	DEFA1
				92.9	COL1A2
				100.8	CEACAM5
				101.1	CALCA
				109.4	LOC653600
				111.5	COL3A1
				165.7	DEFA4

aData are from patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission, as compared to healthy controls.

Table 5

Genes upregulated between 20- and 30-fold in peripheral blood mononuclear cells of patients with stable COPD or AECOPD compared with healthy control subjects

Stable vs control		AE-1 vs control		AE-3 vs control		AE-10 vs control
Fold changes	Genes	Fold changes	Gene	Fold changes	Genes	Fold changes	Genes
20.1	P8	20.3	PLAU	20.0	ALPL	20.3	SCNN1A
20.1	REXO1L5P	21.0	COL6A3	20.1	MUC1	20.4	MGC45438
20.2	UNQ473	21.1	SLC25A37	20.2	SPDEF	20.7	FBLN1
20.2	DEFA1	21.1	HIG2	20.3	HIG2	20.7	CLDN4
20.5	LOC440015	21.2	GPRC5A	20.4	KLK11	20.9	SFTPA2
21.1	LOC391749	21.2	CFB	20.4	MGP	21.0	FKBP9
21.3	MGC45438	21.3	LTF	20.4	GPR109A	21.1	FAM107A
21.7	RP11-146D12.2	21.4	VSIG4	21.0	LOC653342	21.3	N/A
22.0	LOC399839	21.7	FOSB	21.1	CFB	21.4	C10orf10
22.9	SPDEF	21.9	SLC25A37	21.3	P8	21.5	SELENBP1
23.0	CLDN4	22.0	ARG1	21.8	PBEF1	21.6	ANXA3
24.7	LOC349196	22.0	SPDEF	21.9	S100P	21.6	IFI27
25.3	STAC2	22.2	LTF	21.9	MS4A3	21.8	C1QC
25.8	REXO1L3P	22.3	FOS	22.4	COL6A3	21.9	SEPP1
26.3	SCGB3A1	22.6	FAM46C	23.1	MANSC1	22.0	KLK11
26.9	RNASE1	22.6	ISLR	23.2	COL1A2	22.1	P8
27.0	AZGP1	22.6	COL1A2	23.2	GCA	22.1	LOC653723
29.5	H19	22.8	ATP1B1	23.3	LTBP2	22.5	LOC391359
		23.8	SCNN1A	23.9	CHI3L1	22.7	LAMB2
		23.8	SERPINE1	24.0	TMC5	22.8	AQP1
		23.8	EPB42	24.2	CD24	24.0	C9orf61
		23.8	C1QC	24.2	HP	24.1	C4BPA
		23.9	RGS1	24.3	ISLR	24.2	LTBP2
		23.9	ORM2	24.3	SIX1	24.3	UNQ473
		24.1	COL5A1	24.5	APOE	24.5	TMEM139
		24.5	MS4A3	24.6	COL3A1	24.6	N/A
		25.6	CD177	24.6	LOC646309	25.7	OLFML3
		25.6	APOE	24.7	CEACAM3	25.9	SNF1LK
		26.4	C20orf114	24.9	AATK	25.9	A2M
		26.6	BPIL1	25.3	LTF	26.4	FXYD3
		27.1	CTSG	25.4	ALPL	27.0	HP
		27.4	FOS	25.6	ACSL1	27.1	N/A
		27.6	ALAS2	26.2	CEACAM6	27.4	LOC653509
		28.0	INHBA	26.3	COL5A1	28.0	LDB2
		28.0	TIMP3	26.4	KLK11	28.0	OLFML3
		28.1	COL3A1	26.7	PRTN3	28.5	SFTPA1
		28.1	SLC4A1	26.9	RGS1	28.6	MUC1
		28.2	KLK11	27.3	KCNJ15	29.6	HSPA12B
		28.2	LOC653492	27.4	CAMP	29.8	MFAP4
		28.5	LOC203510	27.6	PLAU
		28.7	CEACAM3	27.8	LTF
		28.8	DCN	27.9	ANXA3
		28.9	CEACAM1	28.0	H19
		29.0	CEACAM6	28.0	SERPINE1
		29.3	SELENBP1	28.1	LTF
		29.7	KRT19	28.3	INHBA

aData are from patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission, as compared to healthy controls.

Table 6

Genes upregulated between 15- and 20-fold in peripheral blood mononuclear cells of patients with stable COPD or AECOPD compared with healthy control subjects

Stable vs control		AE-1 vs control		AE-3 vs control		AE-10 vs control
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
15.2	LOC645558	15.0	CNN3	15.0	GPR109B	15.0	USP54
15.7	N/A	15.0	GPT2	15.0	LOC653492	15.0	MGC45438
15.9	LOC653455	15.1	ORM1	15.2	RNASE1	15.2	SLCO2A1
15.9	DUX4	15.1	LOC402110	15.3	FN1	15.3	AGER
16.1	LOC653768	15.1	MDK	15.5	ACSL1	15.3	FLJ11259
16.5	RAB17	15.2	ELF3	15.5	CDH5	15.5	CLEC3B
16.6	LOC653541	15.2	PSG8	15.6	FOLR3	15.8	ADCY4
16.6	LOC391763	15.3	SLC25A37	15.8	PVRL2	16.0	FN1
16.7	LOC642286	15.4	FKBP9	15.9	KRT19	16.1	HP
16.7	S100A14	15.5	C1QB	15.9	MDK	16.1	CKB
16.7	NBPF9	15.6	BPGM	16.0	APOC1	16.1	CYP4B1
16.9	PSG8	15.7	AQP9	16.3	NOL3	16.2	RARRES2
17.0	REXO1L6P	15.7	LOC402207	16.3	ATP1B1	16.3	TSPAN1
17.0	MLPH	15.7	PSG11	16.4	TMC4	16.6	SDC4
17.1	FAM90A7	16.0	KLK11	16.4	VEGF	16.7	ERG
17.4	LOC401650	16.2	KIAA0703	16.6	SPAG4	16.8	LOC653107
17.8	DUB3	16.2	IGFBP5	16.8	LIF	17.2	RAB25
17.9	MGC45438	16.2	IGFBP3	16.8	CCDC80	17.2	COL1A2
18.9	COL3A1	16.2	N/A	16.9	CEACAM3	17.3	DCN
19.1	LOC645732	16.2	SLC25A37	16.9	IGFBP3	17.5	TSPAN13
19.8	LOC392188	16.3	SIX1	17.1	CXCL2	17.6	HSD17B6
20.0	MUC1	16.3	LOC645009	17.2	FKBP9	17.8	RHOB
		16.4	C1QA	17.2	CEACAM1	17.9	KRT19
		16.5	UBD	17.7	ELF3	18.0	AQP9
		16.6	LOC653342	17.7	CNN3	18.2	FOLR1
		17.0	GPR97	17.8	PGLYRP1	18.2	IL1RL1
		17.1	COL1A1	17.9	KRT23	18.2	SERPING1
		17.3	ALPL	18.1	SLC44A4	18.3	MGC35295
		17.4	FBLN1	18.1	SCNN1A	18.4	FLJ43663
		17.5	HIG2	18.4	FBLN1	18.6	TGM2
		17.7	COL8A1	18.5	HPR	18.6	ADH1C
		17.9	TMC5	18.6	SYT7	18.7	KIAA1026
		18.1	LTBP2	18.6	CEACAM8	19.1	DKFZP686A01247
		18.4	SLC25A37	18.8	C1R	19.2	CCDC48
		18.7	CEACAM3	18.8	COL1A1	19.2	ANKRD25
		18.9	MPO	18.9	COL8A1	19.3	DMBT1
		19.0	CD24	18.9	C1QC	19.4	MALL
		19.0	CHI3L1	18.9	SFRP2	19.5	ANXA8
		19.0	DCN	19.0	HIG2	19.5	SPRY4
		19.1	P8	19.2	C1QB	19.7	ELF3
		19.1	CEACAM6	19.2	GPRC5A	19.9	EHD2
		19.1	ACSL1	19.3	MMP25	20.0	DCN
		19.5	PRTN3	19.3	UBD
		19.5	LIF	19.3	GADD45A
		19.6	LTF	19.4	ISLR
		19.7	ANXA3	19.5	ORM1
		19.7	C1R	19.5	C20orf114
		19.7	MUC1	19.5	LOC203510
		19.8	PSG4	19.6	DCN
		19.9	HP	19.7	FN1
				19.8	DAAM2
				19.9	FOLR3

aData are from patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission, as compared to healthy controls.

Table 7

Genes upregulated >30-fold in peripheral blood mononuclear cells of patients with AECOPD compared to patients with stable COPD

AE-1 vs stable		AE-3 vs stable		AE-10 vs stable
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
37.3	MMP8	33.2	LOC646309	30.0	CCDC48
37.6	CEACAM5	34.7	SERPINE1	31.9	LOC653509
38.6	PLUNC	34.9	FOS	32.0	EPAS1
39.4	BPIL1	37.6	CYR61	32.2	CDH5
40.3	CYR61	39.5	CEACAM5	34.4	CLDN5
45.4	CEACAM5	39.6	PLUNC	36.3	SEPP1
55.2	CALCA	40.1	ARG1	38.7	CAV1
56.0	VSIG4	43.5	BPIL1	39.2	CYR61
103.9	CA1	46.0	CEACAM5	42.1	ADH1B
		85.9	CALCA	44.2	CTGF
				44.9	CAV1
				45.1	GPRC5A
				49.8	SEPP1
				81.4	GPX3

aData are from patients with patients with acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission, as compared to patients with stable chronic obstructive pulmonary disease (Stable).

Table 8

Genes upregulated between 20- and 30-fold in peripheral blood mononuclear cells of patients with AECOPD compared to patients with stable COPD

AE-1 vs stable		AE-3 vs stable		AE-10 vs stable
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
20.1	MS4A3	20.5	GPR97	20.3	TIMP3
21.0	CEACAM6	20.6	ALPL	20.3	SLC6A4
21.1	SLC25A37	20.7	MTHFS	20.4	SFTPA2
21.2	DCN	21.0	FLJ32028	20.6	AKAP2
22.4	SPP1	21.4	ADM	20.7	DST
24.0	TCN1	23.3	ACSL1	21.2	TCF21
24.7	BPIL1	23.3	DCN	21.5	ADH1C
26.4	SLC25A37	24.3	MMP8	21.6	SLIT3
26.6	CTGF	24.5	TCN1	21.7	C9orf61
28.5	ARG1	25.3	FOS	22.5	FOSB
28.6	FOS	25.3	FOSB	25.5	MFAP4
29.5	SERPINE1	27.5	CTGF	26.0	GPX3
		28.3	BPIL1	26.5	DCN
		28.4	VSIG4	26.9	SFTPB
				27.6	FBLN5
				28.1	LOC653509
				28.5	ADH1C
				28.7	SFTPA1
				28.7	TIMP3

aData are from patients with patients with acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission, as compared to patients with stable chronic obstructive pulmonary disease (Stable).

Table 9

Genes upregulated between 15- and 20-fold in peripheral blood mononuclear cells of patients with AECOPD compared to patients with stable COPD

AE-1 vs stable		AE-3 vs stable		AE-10 vs stable
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
15.6	ADM	15.2	LOC387763	15.0	VSIG4
15.8	DEFA4	15.2	MMP25	15.6	IL1RL1
16.2	DEFA4	15.3	USP15	15.8	PZP
16.2	C1R	15.5	C1R	16.0	LDB2
16.9	GPNMB	15.8	KCNJ15	16.2	FLJ43663
17.2	DCN	15.9	GADD45A	16.5	N/A
17.3	FAM46C	15.9	LRRC4	16.6	CD55
17.6	ALAS2	16.3	GLT1D1	16.8	CXCL2
17.6	CALCA	16.4	CD55	16.9	IL1RL1
17.9	GPNMB	16.5	CEACAM6	17.0	RHOB
18.2	DUSP1	16.6	SPP1	17.1	DLC1
18.2	CEACAM6	16.7	SLC25A37	17.2	VIPR1
18.2	SLC25A37	17.1	ORM1	17.2	CRYAB
18.7	FOS	17.2	CALCA	17.8	CNN3
18.9	SLC25A37	17.3	DUSP1	18.1	DCN
		17.5	CD177	18.1	IFI27
		17.6	GPNMB	18.2	SLIT2
		17.7	MS4A3	18.3	RASIP1
		17.8	DCN	18.8	MFAP4
		17.8	GPR109A	19.0	CAMK2N1
		17.9	BASP1	19.0	CD55
		17.9	IL8RB	19.5	AGER
		18.4	AQP9	19.9	DKFZP686A01247
		18.7	DEFA4
		18.8	QPCT
		19.0	PBEF1
		19.0	BASP1
		19.0	CEACAM6
		19.2	GNG10
		19.7	GPNMB
		19.7	GCA
		20.0	RNASE3

aData are from patients with patients with acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission, as compared to patients with stable chronic obstructive pulmonary disease (Stable).

Table 10

Genes upregulated more than fivefold in peripheral blood mononuclear cells of patients with AECOPD

AE-3 vs AE-1		AE-10 vs AE-1		AE-10 vs AE-3
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
5.1	TMEM50A	10.3	SUSD2	10.1	SLCO2A1
5.2	BCL2A1	10.6	TCF21	10.1	OAS3
5.3	C6orf32	10.6	FOLR1	10.1	C4BPA
6.0	PI3	10.7	C9orf61	10.2	DMBT1
7.0	KCNJ15	10.9	LOC653107	10.4	VSIG2
7.6	CISH	11.3	AGER	10.4	LOC653107
10.4	CISH	12.0	SLIT2	10.5	ITLN2
10.7	CISH	12.7	ITLN2	10.7	CX3CR1
		12.9	FLRT3	10.7	MSLN
		13.1	VIPR1	10.8	SOCS2
		13.2	SOCS2	10.9	LOC653107
		13.3	IL1RL1	11.7	FOLR1
		13.4	LOC653107	11.7	GPX3
		13.8	C4BPA	11.8	CLIC5
		14.4	CYP4B1	11.8	SLIT2
		14.4	LAMA3	11.9	LOC653107
		15.1	CYP4B1	12.1	AQP1
		15.2	ADH1C	12.6	LOC653509
		15.7	MGC35295	12.6	ADH1C
		15.8	GPX3	12.7	ADH1C
		17.0	IL1RL1	12.8	ADH1B
		17.9	MSLN	12.9	LAMA3
		20.0	ADH1C	13.6	IL1RL1
		22.4	ADH1B	13.6	CYP4B1
		24.5	SLC6A4	13.9	FAM107A
		35.3	FOLR1	14.2	LOC653107
				14.9	CYP4B1
				22.0	MGC35295
				31.2	SLC6A4

aData are from day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission.

Table 11 lists the number of genes downregulated more than twofold, including 4,516, 2,975, 3,426 and 2,798 genes of PBMCs from patients with stable COPD on AE-1, AE-3 and AE-10, respectively, below controls; 3,207, 4,510 and 5288 genes on AE-1, AE-3 and AE-10, respectively, below stable COPD; 598 genes from AE-3 below AE-1; and 2,162 and 1,918 genes from AE-10 below those from AE-1 and AE-3, respectively. Downregulated genes of PBMCs from patients with stable COPD, AE-1, AE-3 or AE-10 greater than tenfold, between 10- and 8-fold or between 8- and 6-fold below healthy control subjects are listed in Tables 12, 13 and 14, respectively. Downregulated genes of PBMCs from patients at AE-1, AE-3 or AE-10 compared to stable COPD, or among patients with AECOPD, are shown in Tables 15 and 16.

Table 11

Number of downregulated genes in peripheral blood mononuclear cells of healthy control subjects, patients with stable COPD and patients with AECOPD

	Fold changes in upregulated genes ( n )
Compared pairs	>2	>5	>6	>8	>10	>15	>20	>30	>50	>100
Stable vs Con	4,516	135	55	9	4	2	1	0	0	0
AE-1 vs Con	2,975	182	107	47	22	7	4	1	0	0
AE-3 vs Con	3,426	225	149	65	35	11	5	2	0	0
AE-10 vs Con	2,798	124	73	31	16	2	1	1	0	0
AE-1 vs Stable	3,207	33	16	4	4	2	0	0	0	0
AE-3 vs Stable	4,510	125	71	21	8	3	1	0	0	0
AE-10 vs Stable	5,288	445	236	97	49	20	8	3	0	0
AE-3 vs AE-1	598	32	23	17	5	3	2	0	0	0
AE-10 vs AE-1	2,162	261	168	82	43	21	14	10	5	1
AE-10 vs AE-3	1,918	192	130	66	36	15	9	6	4	0

aData are from controls (Con) or patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of the hospital admission.

Table 12

Genes downregulated more than tenfold in peripheral blood mononuclear cells of patients with stable COPD or AECOPD compared to healthy control subjects

Stable vs Con		AE-1 vs Con		AE-3 vs Con		AE-10 vs Con
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
10.7	EIF3S6	10.3	HAND1	10.2	GZMK	10.0	C21orf7
10.7	YLPM1	10.3	CD8B	10.5	CXCR3	10.0	NELL2
16.1	TFCP2L1	10.4	UBASH3A	10.6	AK5	10.4	C21orf7
21.0	SCP2	10.8	TRA@	10.7	TRA@	10.4	GFI1B
		10.9	TRBV3-1	10.7	IL24	10.5	LOC129293
		11.2	CD8B	10.9	CD6	10.5	LOC123876
		11.4	MAL	10.9	N/A	10.7	HIST1H3H
		11.4	LOC643514	11.2	KIAA0748	11.1	IL24
		11.5	NELL2	11.4	LCK	11.4	GFI1B
		11.7	TTC24	11.5	CD8B	11.9	CRTAC1
		12.7	CD8B	12.3	APBB1	11.9	OR10A4
		13.1	LEF1	12.3	IL12RB1	11.9	SAA3P
		13.8	TCF7	12.5	TTC24	12.7	TTC24
		14.2	LOC129293	12.5	GFI1B	14.9	TFCP2L1
		14.5	LOC129293	12.5	CRTAC1	18.6	SCP2
		15.6	TCF7	12.6	TRBV3-1	32.3	UNQ470
		16.1	TCF7	12.6	ATG9B
		16.8	CD8B	12.9	ABLIM1
		21.8	TFCP2L1	12.9	LOC129293
		25.4	CRTAC1	13.0	CD8B
		27.9	SCP2	13.1	CD28
		44.1	UNQ470	13.1	GRAP2
				14.3	UBASH3A
				14.4	CCR7
				15.0	LOC129293
				16.0	CD8B
				18.1	UNQ470
				18.7	SCP2
				18.8	LEF1
				19.3	LEF1
				23.5	CD8B
				24.3	TCF7
				25.1	TCF7
				30.4	TCF7
				32.0	TFCP2L1

aData are from patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of the hospital admission, as compared to healthy controls (Con).

Table 13

Genes downregulated between eight- and tenfold in peripheral blood mononuclear cells of patients with stable COPD or AECOPD compared to healthy control subjects

Stable vs Con		AE-1 vs Con		AE-3 vs Con		AE-10 vs Con
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
8.2	AK5	8.1	CD3G	8.0	TRBV19	8.1	HFE2
8.6	TRA@	8.2	LY9	8.0	OTOA	8.2	TRA@
9.1	ZC3HAV1	8.2	AK5	8.1	CD7	8.6	UNQ470
9.3	MAL	8.2	C21orf7	8.1	GRAP2	8.6	CD248
9.7	TMEM50B	8.2	TRBC1	8.1	TNFRSF25	8.6	XG
		8.3	ANKDD1A	8.2	C21orf7	8.7	ATG9B
		8.4	CD6	8.2	EPHA6	8.8	LOC339778
		8.4	RPS6KB1	8.2	GIMAP5	8.9	TCF7
		8.5	TMEM50B	8.3	1-Sep	8.9	CCR7
		8.7	YLPM1	8.3	UBASH3A	9.2	LOC644663
		8.7	TRBV19	8.4	GIMAP7	9.4	LOC129293
		8.8	FLT3LG	8.5	MGC23244	9.5	MGC39606
		8.9	N/A	8.6	LOC645852	9.7	GZMK
		9.1	LEF1	8.7	SCAP1	9.9	AK5
		9.1	GZMK	9.0	HIST1H3H	9.9	TCF7
		9.1	KIAA0748	9.0	HFE2
		9.2	ABLIM1	9.2	GFI1B
		9.5	C21orf7	9.2	TMEM50B
		9.5	ATG9B	9.5	N/A
		9.6	LCK	9.5	C21orf7
		9.6	LOC647353	9.6	GATA3
		9.8	CCR7	9.7	C21orf7
		9.8	UNQ470	9.7	CD247
		9.9	OR10A4	9.8	LCK
		9.9	IL12RB1	9.8	KSP37
				9.9	FAIM3
				9.9	SPOCK2
				9.9	TRA@
				9.9	SH2D1B
				10.0	GRAP2

aData are from patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of the hospital admission, as compared to healthy controls (Con).

Table 14

Genes downregulated between six- and eightfold in peripheral blood mononuclear cells of patients with stable COPD or AECOPD compared to healthy control subjects

Stable vs Con		AE-1 vs Con		AE-3 vs Con		AE-10 vs Con
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
6.0	NDUFV3	6.0	MAL	6.0	IL7R	6.0	HKDC1
6.0	C17orf45	6.0	ARHGAP12	6.0	CD28	6.1	TANC2
6.1	MAL	6.0	TRAPPC4	6.0	KIR2DS1	6.1	FAM5B
6.1	CXCR6	6.0	GNLY	6.0	FLJ20647	6.2	KIAA0748
6.1	SUCLA2	6.0	N/A	6.1	N/A	6.3	CD40LG
6.1	C21orf7	6.0	LOC642376	6.1	CLDN1	6.3	PCDH10
6.2	TNPO1	6.0	MYOZ3	6.1	TRBV5-4	6.3	LOC644273
6.2	LOC643514	6.1	FLJ20647	6.1	CARD11	6.3	CD96
6.2	ALS2CR13	6.1	CD96	6.1	LOC441320	6.3	TRA@
6.2	CREB1	6.2	MAL	6.1	ACADSB	6.3	TRBV3-1
6.2	C17orf45	6.2	GIMAP5	6.1	NXPH4	6.4	TRA@
6.3	NELL2	6.2	CLDN1	6.2	SCNN1D	6.4	LOC642483
6.3	C6orf32	6.2	CD3D	6.2	MTMR1	6.5	ANKDD1A
6.3	LOC642455	6.2	LY9	6.2	MAL	6.5	N/A
6.4	GMDS	6.3	LOC123876	6.2	ZAP70	6.5	N/A
6.4	ABHD6	6.3	TNFRSF25	6.3	MAL	6.5	LY9
6.4	DAPP1	6.3	C21orf7	6.3	IL2RB	6.6	CD8B
6.4	SH3BGRL	6.3	LOC645885	6.3	EDG8	6.6	MGC26597
6.5	IL7R	6.3	BLOC1S3	6.3	HKDC1	6.7	TRBV19
6.6	LOC441601	6.3	LOC644727	6.3	SCAP1	6.7	LOC145783
6.6	GPR18	6.4	CCDC45	6.3	LOC440455	6.8	CD8B
6.7	P2RX5	6.4	C21orf7	6.3	CD300E	6.9	C21orf7
6.7	LY9	6.5	CD28	6.4	LY9	6.9	UBASH3A
6.8	GGPS1	6.5	LOC440455	6.4	KIR2DS2	7.0	LOC400768
6.8	EIF3S6	6.5	IL24	6.4	SLAMF6	7.1	CD8B
6.8	ARHGAP15	6.5	GHRL	6.4	SAA3P	7.1	HAND1
6.8	SF3B1	6.5	FAM113B	6.4	SF3A2	7.2	LOC126075
6.8	GPR89A	6.5	LOC644663	6.5	UNQ470	7.2	TNFRSF7
6.9	LOC129293	6.5	C15orf37	6.5	C6orf21	7.3	LEF1
6.9	CPNE3	6.5	MAL	6.6	CD96	7.3	HLA-DOA
6.9	LY9	6.5	LOC644445	6.6	CD244	7.4	LOC646279
7.0	PIP3-E	6.6	LOC126075	6.6	N/A	7.4	YLPM1
7.0	TAF9	6.6	1-Sep	6.6	KLRK1	7.4	LOC643514
7.0	N/A	6.6	UBASH3A	6.6	C16orf5	7.5	MTMR1
7.0	KIAA0748	6.7	SAA3P	6.6	TRBC1	7.6	NOG
7.1	CD55	6.8	CD6	6.6	LOC339778	7.7	TCF7
7.2	EIF3S6	6.8	TRBV5-4	6.7	GNLY	7.7	KIAA0748
7.2	PGRMC2	6.9	1-Sep	6.7	LDLRAP1	7.7	C21orf7
7.3	C21orf7	6.9	LOC129293	6.8	HAND1	7.7	PRDM9
7.4	PSMD6	7.0	SCNN1D	6.8	CD3D	7.7	FCER2
7.5	ABLIM1	7.0	SIT1	6.8	FLJ45825	7.9	CD8B
7.6	STAG2	7.1	GATA3	6.8	SF3A2	8.0	LEF1
7.8	CCDC45	7.1	CD7	6.8	CXCR3
7.8	UNQ470	7.1	CDKN3	6.8	KIR3DL3
7.9	LY9	7.2	SCAP1	6.8	LAT
8.0	CD40LG	7.3	TRA@	6.9	CD52
		7.3	LY9	6.9	TNFRSF7
		7.3	DDAH1	6.9	LOC442726
		7.3	TRA@	6.9	3-Sep
		7.5	TNFRSF7	6.9	KIAA0748
		7.5	KIAA0748	6.9	XG
		7.6	ITM2A	6.9	KIAA1549
		7.6	CD5	7.0	RNF157
		7.6	D4S234E	7.0	SIT1
		7.6	CD300E	7.0	CD1C
		7.7	APBB1	7.0	SLC16A10
		7.8	CD3D	7.0	CD3G
		7.8	LCK	7.1	CD6
		7.8	UBASH3A	7.1	LY9
		7.9	XG	7.1	FLT3LG
				7.1	LOC647353
				7.2	LOC123876
				7.2	CX3CR1
				7.2	LOC126075
				7.3	NELL2
				7.4	LY9
				7.4	MAL
				7.4	KIR2DS2
				7.4	CHIA
				7.4	BIN1
				7.5	CCDC78
				7.5	MAL
				7.5	C21orf7
				7.5	KIR2DL4
				7.6	CD6
				7.6	CD3D
				7.7	1-Sep
				7.7	LCK
				7.8	ITM2A
				7.8	TRA@
				7.9	SIT1
				7.9	CD5
				8.0	CD8A
				8.0	LOC129293

aData are from patients with stable chronic obstructive pulmonary disease (Stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) after the hospital admission, as compared to healthy controls (Con).

Table 15

Genes downregulated more than fivefold in peripheral blood mononuclear cells of patients with AECOPD compared to patients with stable COPD

AE-1 vs Stable		AE-3 vs Stable		AE-10 vs Stable
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
5.0	PRODH	10.2	KSP37	10.1	LOC646781
5.1	MT1F	10.3	DUB3	10.1	LOC389634
5.1	OR2A7	10.6	DUB3	10.1	LOC441056
5.3	CD8B	10.8	TCF7	10.1	LOC340243
5.4	CGI-38	11.2	CX3CR1	10.2	C1QL2
5.4	DMBT1	17.6	MGC35295	10.2	LOC653541
5.4	N/A	19.9	STAC2	10.2	LOC158318
5.4	GNLY	25.0	AZGP1	10.3	N/A
5.5	LCK			10.4	LOC644373
5.5	DZIP1			10.6	SPDEF
5.6	TCF7			10.7	DUX1
5.6	MGC45438			10.9	LOC643001
5.6	UNQ470			11.1	LOC391767
5.8	MGLL			11.2	LOC645509
5.8	B4GALNT3			11.7	FLJ36131
5.9	CGI-38			11.8	LOC441323
5.9	CGI-38			11.9	LOC440015
6.1	LOC388886			11.9	LOC441812
6.1	GNLY			12.0	TCEB3C
6.2	N/A			12.1	SPDEF
6.4	CD8B			12.3	DUX4
6.4	AEBP2			12.5	LOC285697
6.4	EDG8			12.9	LOC646066
6.5	PRDM16			13.3	LOC441873
6.8	CX3CR1			13.6	LOC645402
7.0	MGC45438			13.7	LOC285563
7.3	MST1			13.9	LOC391763
7.4	LOC644088			14.4	DUB3
7.5	EDG8			14.7	LOC391766
10.1	MGC45438			15.0	LOC392197
12.6	MGC35295			15.0	REXO1L2P
15.5	STAC2			15.2	DUB3
19.1	AZGP1			15.2	LOC402199
				15.7	LOC653442
				15.8	LOC653455
				16.0	LOC402207
				16.5	LOC391745
				16.7	LOC392188
				18.1	REXO1L6P
				19.1	LOC391764
				19.4	DUB3
				20.6	LOC645836
				21.0	LOC391749
				23.8	LOC402110
				24.2	REXO1L7P
				29.6	REXO1L1
				30.0	STAC2
				33.5	REXO1L3P
				39.7	REXO1L5P

aData are from patients acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) after the hospital admission, as compared to patients with stable chronic obstructive pulmonary disease (Stable).

Table 16

Genes downregulated more than fivefold in peripheral blood mononuclear cells of patients with AECOPD

AE-3 vs AE-1		AE-10 vs AE-1		AE-10 vs AE-3
Fold changes	Genes	Fold changes	Genes	Fold changes	Genes
5.0	ITGB3	10.2	MPO	10.3	MOXD1
5.1	CGI-69	10.4	LOC653492	10.3	LOC152573
5.2	SPTB	10.5	SPP1	10.7	SPDEF
5.2	BCL2L1	10.6	ANK1	10.8	CCDC80
5.2	GATA1	11.0	DEFA4	11.0	CTSG
5.3	FBXO7	11.0	MOXD1	11.0	CAMP
5.6	SELENBP1	11.0	HIG2	11.3	PLA2G2D
5.8	OSBP2	11.1	OSBP2	11.4	SPP1
5.9	LOC643855	11.2	REXO1L3P	11.6	S100P
6.1	ERAF	11.6	SPDEF	11.7	SLC4A11
6.2	EPB49	12.0	COL1A1	11.8	COL3A1
6.2	MYH9	12.2	BPI	11.8	SPAG4
6.4	ALAS2	12.3	SNCA	12.5	THBS2
7.4	LOC644462	12.3	SLC4A11	12.7	MPO
7.8	GMPR	12.5	COL1A1	13.0	PRTN3
8.1	ANK1	12.6	AZU1	13.2	COL1A1
8.9	BPGM	12.6	ARG1	13.3	ELA2
9.1	FAM46C	13.2	GREM1	14.3	LIF
9.2	LOC643497	13.5	DEFA4	14.4	CEACAM5
9.4	TRIM58	13.5	ELA2	14.6	RNF183
9.4	MBNL3	14.2	CEACAM5	14.9	B3Gn-T6
9.5	EPB49	14.5	ITGA11	15.1	AZU1
9.6	EPB49	15.0	CEACAM8	15.4	ITGA11
9.6	EPB42	15.3	SPTB	15.9	DEFA4
9.7	EPB41	15.6	CEACAM5	16.4	CEACAM5
9.7	SLC14A1	16.6	LIF	17.5	MS4A3
9.9	EPB42	17.1	TRIM58	17.8	ARG1
10.1	SNCA	19.2	THY1	20.4	THY1
13.5	TRIM58	19.5	MS4A3	21.1	MS4A3
19.7	SLC4A1	23.1	TRIM58	22.5	SPP1
20.7	EPB41	24.1	MS4A3	34.4	SFRP2
21.6	CA1	27.1	SFRP2	49.9	PLUNC
		29.9	EPB42	57.1	CALCA
		30.4	SPP1	68.9	CALCA
		41.9	ALAS2	80.4	BPIL1
		43.8	EPB42	93.1	BPIL1
		44.2	CALCA
		48.5	PLUNC
		55.5	SLC4A1
		58.6	CALCA
		70.0	BPIL1
		84.3	BPIL1
		109.9	CA1

aData are from patients with acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) after the hospital admission.

COPD-specific genes

To search for COPD-specific genes, co–differentially expressed genes of PBMCs from patients with stable COPD or AECOPD were compared with those from control subjects (listed in Additional file 4). There were five groups and four comparison pairs with information regarding fold changes and regulation (that is, SEQ-ID, fold change, log or absolute fold change, or regulation), normalized intensities or annotations (that is, GENE_NAME, synonyms, description, NCBI_GENE_ID, chromosome, GO, UniGene ID, TIGRID or Ensembl ID). Seventy-nine genes were upregulated and 23 genes downregulated in PBMCs from patients with COPD, including both stable COPD and AECOPD, as compared to the healthy control subjects, as shown in Table 17. Of them, 14 genes were upregulated and 2 were downregulated more than tenfold, as compared to control subjects, including carcinoembryonic antigen–related cell adhesion molecule 1, collagen type VIα3(VI), collagen type I(α)2(I), nucleolar protein 3 (apoptosis repressor with CARD domain), melanophilin, cell surface–associated mucin 1, nuclear protein 1, chemokine (C-X-C motif) ligand 17, claudin 4, ribonuclease 1, imprinted maternally expressed transcript, defensin α1, transcription factor CP2-like 1 and sterol carrier protein 2 (SCP2).

Table 17

Number and details of co–differentially up- or downregulated genes in peripheral blood mononuclear cells of patients with stable COPD or AECOPD compared to healthy control subjects

Fold change	>5		>10
Upregulated	79		14
Downregulated	23		2
Unexpressed genes (>10)
SEQ-ID	Gene name	Full name of gene	Stable vs Con	AE-1 vs Con	AE-3 vs Con	AE-10 vs Con
D12502	CEACAM1	Carcinoembryonic antigen-related cell adhesion molecule 1	10.1	83.0	66.5	10.5
NM_004369	COL6A3	Collagen, type VI, α3	10.4	21.0	22.4	10.8
AF064599	NOL3	Nucleolar protein 3 (apoptosis repressor with CARD domain)	12.1	13.6	16.3	11.5
BC042586	COL1A2	Collagen, type I, α2	13.1	72.3	92.9	17.2
BC014473	CEACAM1	Carcinoembryonic antigen-related cell adhesion molecule 1	14.7	101.2	61.0	11.8
AY358857	MLPH	Melanophilin	17.0	10.3	12.8	12.2
AF348143	MUC1	Mucin 1, cell surface-associated	20.0	19.7	20.1	28.6
NM_012385	P8	p8 protein (candidate of metastasis 1)	20.1	19.1	21.3	22.1
BC093946	UNQ473	DMC	20.2	45.9	53.1	24.3
NM_001305	CLDN4	Claudin 4	23.0	34.4	39.1	20.7
NM_002933	RNASE1	Ribonuclease, RNase A family, 1 (pancreatic)	26.9	12.5	15.2	37.2
BC053636	H19	H19, imprinted maternally expressed untranslated mRNA	29.5	37.2	28.0	11.8
BC069423	DEFA1	Defensin, α1	33.0	96.1	86.7	10.2
XM_928349	LOC653600	Similar to neutrophil defensin 1 precursor (HNP-1) (HP-1) (HP1) (defensin, α1)	43.1	115.8	109.4	12.8
Downregulated genes (>5)
SEQ-ID	Gene name	Full name of genes	Stable vs Con	AE-1 vs Con	AE-3 vs Con	AE-10 vs Con
M38056	HLA-DOA	Major histocompatibility complex, class II, DOα	5.3	5.9	5.6	7.3
AY209188	SAA3P	Serum amyloid A3 pseudogene	5.3	6.7	6.4	11.9
BC069511	UBASH3A	Ubiquitin-associated and SH3 domain-containing, A	5.5	10.4	14.3	6.9
AJ421515	CRTAC1	Cartilage acidic protein 1	5.6	25.4	12.5	11.9
AL133666	EPHA6	EPH receptor A6	5.6	5.8	8.2	5.3
NM_020152	C21orf7	Chromosome 21 open reading frame 7	5.7	8.2	9.7	10.4
XM_089384	TTC24	Tetratricopeptide repeat domain 24	5.8	11.7	12.5	12.7
NM_006850	IL24	Interleukin 24	6.0	6.5	10.7	11.1
AL713701	C21orf7	Chromosome 21 open reading frame 7	6.1	9.5	9.5	10.0
XM_931594	LOC643514	Hypothetical protein LOC643514	6.2	11.4	5.7	7.4
NM_006159	NELL2	NEL-like 2 (chicken)	6.3	11.5	7.3	10.0
NM_002348	LY9	Lymphocyte antigen 9	6.7	8.2	7.4	6.5
XM_934852	LOC129293	Hypothetical protein LOC129293	6.9	14.5	12.9	9.4
BC062589	LY9	Lymphocyte antigen 9	6.9	7.3	7.1	5.5
XM_934149	KIAA0748	KIAA0748	7.0	7.5	11.2	6.2
BC008567	C21orf7	Chromosome 21 open reading frame 7	7.3	6.3	7.5	7.7
NM_138363	CCDC45	Coiled-coil domain containing 45	7.8	6.4	5.9	5.2
BC022101	UNQ470	GAAI470	7.8	44.1	18.1	32.3
BC027920	LY9	Lymphocyte antigen 9	7.9	6.2	5.8	5.3
BC033896	AK5	Adenylate kinase 5	8.2	8.2	10.6	9.9
XM_085151	YLPM1	YLP motif containing 1	10.7	8.7	5.1	7.4
NM_014553	TFCP2L1	Transcription factor CP2-like 1	16.1	21.8	32.0	14.9
NM_001007098	SCP2	Sterol carrier protein 2	21.0	27.9	18.7	18.6

aData are from patients with stable chronic obstructive pulmonary disease (stable) or acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of the hospital admission, as compared to healthy controls (Con).

AECOPD-specific genes

To search for AECOPD-specific genes, co–differentially expressed genes of PBMCs from patients with AECOPD on days 1, 3 and 10 were compared to those from either patients with stable COPD or healthy control subjects (listed in Additional file 4). There were five groups and six comparison pairs with information regarding fold changes and regulation (that is, SEQ-ID, fold change, log or absolute fold change, or regulation), normalized intensities or annotations (that is, GENE_NAME, synonyms, description, NCBI_GENE_ID, chromosome, GO, UniGene ID, TIGRID or Ensembl ID). As compared with both patients with stable COPD and healthy control subjects, 58 genes were upregulated more than fivefold and 238 downregulated more than twofold in patients with AECOPD. Of them, eight upregulated (more than tenfold) and eight downregulated (more than threefold) genes are listed in Table 18. These genes include FBJ murine osteosarcoma viral oncogene homologue (FOS); interferon α-inducible protein 27 (IFI27); cysteine-rich angiogenic inducer 61 (CYR61), connective tissue growth factor (CTGF); G protein–coupled receptor family C group 5 member A (GPRC5A); FBJ murine osteosarcoma viral oncogene homologue B (FOSB); decorin (DCN); hypothetical LOC387763 (LOC387763); killer cell immunoglobulin-like receptor, two domains, short cytoplasmic tail, 2 (KIR2DS2); SH2 domain containing 1B (SH2D1B); CD8b molecule (CD8B); olfactory receptor family 2, subfamily W, member 5 (OR2W5); fibroblast growth factor binding protein 2 (FGF2); and transcription factor 7 (TCF7).

Table 18

Number of co–differentially up- or downregulated genes in peripheral blood mononuclear cells of patients with AECOPD compared to patients with stable COPD and healthy control subjects

Fold change		>5			>10
Upregulated		58			8
Fold change		>2			>3
Downregulated		238			8
Selected co–differentially upregulated genes (>10-fold)
SEQ_ID	Gene name	AE-1		AE-3		AE-10
		AE-1 vs Con	AE-1 vs Stable	AE-3 vs Con	AE-3 vs Stable	AE-10 vs Con	AE-10 vs Stable
BC004490	FOS	27.4	28.6	33.5	34.9	13.2	13.7
BC015492	IFI27	12.3	10.3	13.1	11.0	21.6	18.1
NM_001554	CYR61	12.0	40.3	11.2	37.6	11.7	39.2
NM_001901	CTGF	35.7	26.6	36.9	27.5	59.3	44.2
NM_003979	GPRC5A	21.2	12.6	19.2	11.4	75.8	45.1
NM_006732	FOSB	21.7	13.7	40.1	25.3	35.6	22.5
NM_133504	DCN	19.0	17.2	19.6	17.8	20.0	18.1
XM_373497	LOC387763	41.4	13.5	46.4	15.2	41.3	13.5
Selected co–differentially downregulated genes (>3-fold)
SEQ_ID	Gene names	AE-1		AE-3		AE-10
		AE-1 vs Con	AE-1 vs Stable	AE-3 vs Con	AE-3 vs Stable	AE-10 vs Con	AE-10 vs Stable
AJ002102	KIR2DS2	3.7	3.8	7.4	7.6	4.2	4.4
BC022407	SH2D1B	3.0	3.7	4.8	5.9	3.1	3.8
BC066595	SH2D1B	3.6	3.2	9.9	8.9	3.6	3.2
BC100911	CD8B	11.2	4.4	16.0	6.3	7.9	3.1
NM_001004698	OR2W5	3.7	3.1	4.7	4.0	3.7	3.1
NM_004931	CD8B	10.3	5.3	11.5	5.9	6.6	3.4
NM_031950	KSP37	4.8	5.0	9.8	10.2	3.0	3.1
NM_201633	TCF7	15.6	5.6	30.4	10.8	8.9	3.2

aData are from acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of the hospital admission, as compared to patients with stable COPD (Stable) or healthy controls (Con).

Dynamic change in gene expression in patients with AECOPD

Dynamic changes (down–down, down–up, up–down and up–up) of co–differentially expressed genes of PBMCs from patients with AECOPD are listed in Additional file 4, including fold changes and regulation (that is, SEQ-ID, fold change, log or absolute fold change, or regulation), normalized intensities or annotations (that is, GENE_NAME, synonyms, description, NCBI_GENE_ID, chromosome, GO, UniGene ID, TIGRID or Ensembl ID). Table 19 shows the dynamic changes in the patterns of down–down (52 genes), down–up (131 genes), up–down (238 genes) and up–up (8 genes) more than twofold, as compared with the gene expression on the previous day. The major genes of PBMCs from patients with AECOPD were aminolevulinate, delta-, synthase 2 (ALAS2), erythrocyte membrane protein band 4.2 (EPB42) and carbonic anhydrase I (CA1) in a down–down pattern; selenium-binding protein 1 (SELENBP1) and myosin heavy chain 9, non-muscle (MYH9), in a down–up pattern; HLA complex group 27 (HCG27), BCL2-related protein A1 (BCL2A1), G protein–coupled receptors 109A and 109B (GPR109A and GPR109B) in an up–down pattern; and zeta protein kinase C (PRKCZ), ATP-binding cassette, subfamily A, member 8 (ABCA8), and folate receptor 1 (adult) (FOLR1) in an up–up pattern (Table 19). Levels of genes from patients with AECOPD were also compared with those from patients with stable COPD, as shown in Figure 3, where positive or negative values indicate up- or downregulation as compared with those from patients with stable COPD. When correlated with DESS, ALAS2 and CA1 had similar patterns of change with DESS.

Table 19

Number of genes in peripheral blood mononuclear cells of patients with AECOPD

	Down–down		Down–up	Up–down	Up–up
Total	353		784	1,005	127
>2-fold	52		131	238	8
>4-fold	3		3	7	0
>5-fold	2		0	0	0
Selected co–differentially expressed genes at the down–down pattern (>4-fold)
SEQ-ID	Gene name		Full name of gene	AE-3 vs AE-1	AE-10 vs AE-3
NM_000032	ALAS2		Aminolevulinate, delta-, synthase 2	6.4	6.5
BC099627	EPB42		Erythrocyte membrane protein band 4.2	9.9	4.4
BC027890	CA1		Carbonic anhydrase I	21.6	5.1
Selected co–differentially expressed genes at the down–up pattern (>4-fold)
SEQ-ID	Gene name		Full name of gene	AE-3 vs AE-1	AE-10 vs AE-3
AK127453	N/A		Homo sapiens cDNA FLJ45545 fis, clone BRTHA2034281.	4.7	5.7
NM_003944	SELENBP1		Selenium-binding protein 1	5.6	4.1
BC090921	MYH9		Myosin, heavy chain 9, non-muscle	6.2	4.1
Selected co–differentially expressed genes at the up–down pattern (>4-fold)
SEQ-ID	Gene name		Full name of gene	AE-3 vs AE-1	AE-10 vs AE-3
NM_181717	HCG27		HLA complex group 27	4.1	7.3
NM_177551	GPR109A		G protein-coupled receptor 109A	4.3	7.5
NM_006018	GPR109B		G protein-coupled receptor 109B	4.4	5.1
AF249277	MTHFS		5,10-methenyltetrahydrofolate synthetase (5-formyltetrahydrofolate cyclo-ligase)	4.6	5.3
AY234180	BCL2A1		BCL2-related protein A1	5.2	4.0
BC010952	PI3		Peptidase inhibitor 3, skin-derived (SKALP)	6.0	4.4
NM_002243	KCNJ15		Potassium inwardly rectifying channel, subfamily J, member 15	7.0	4.8
Selected co–differentially expressed genes at the up–up pattern (>2-fold)
SEQ-ID	Gene name		Full name of gene	AE-3 vs AE-1	AE-10 vs AE-3
Z15108	PRKCZ		Protein kinase C, zeta	2.0	2.8
BC037798	CGI-38		Brain-specific protein	2.0	2.4
NM_001033581	PRKCZ		Protein kinase C, zeta	2.1	2.8
NM_007168	ABCA8		ATP-binding cassette, subfamily A, member 8	2.1	4.0
AK022468	SORBS1		Sorbin and SH3 domain containing 1	2.3	3.5
NM_006403	NEDD9		Neural precursor cell expressed, developmentally downregulated 9	2.3	2.2
NM_023037	FRY		Furry homologue (Drosophila)	2.3	2.1
NM_016730	FOLR1		Folate receptor 1 (adult)	3.0	11.7
Down–down	GENE_NAME	SEQ_ID	AE-1 vs Stable	AE-3 vs Stable	AE-10 vs Stable
	ALAS2	NM_000032	17.64	2.76	−2.37
	EPB42	BC099627	10.02	1.01	−4.37
	CA1	BC027890	103.93	4.81	−1.06
Down–up	GENE_NAME	SEQ_ID	AE-1 vs Stable	AE-3 vs Stable	AE-10 vs Stable
	N/A	AK127453	−1.69	−7.90	−1.38
	SELENBP1	NM_003944	3.97	−1.41	2.92
	MYH9	BC090921	−1.36	−8.40	−2.04
Up–down	GENE_NAME	SEQ_ID	AE-1 vs Stable	AE-3 vs Stable	AE-10 vs Stable
	HCG27	NM_181717	1.09	4.47	−1.63
	GPR109A	NM_177551	4.12	17.79	2.36
	GPR109B	NM_006018	2.64	11.64	2.28
	MTHFS	AF249277	4.51	20.75	3.95
	BCL2A1	AY234180	2.38	12.45	3.11
	PI3	BC010952	1.03	6.20	1.42
	KCNJ15	NM_002243	2.25	15.78	3.26
Up–up	GENE_NAME	SEQ_ID	AE-1 vs Stable	AE-3 vs Stable	AE-10 vs Stable
	PRKCZ	Z15108	−1.25	1.61	4.46
	CGI-38	BC037798	−5.87	−2.86	−1.18
	PRKCZ	NM_001033581	−1.61	1.30	3.64
	ABCA8	NM_007168	−1.27	1.68	6.69
	SORBS1	AK022468	1.28	2.92	10.30
	NEDD9	NM_006403	2.43	5.57	12.15
	FRY	NM_023037	−1.11	2.08	4.34
	FOLR1	NM_016730	−4.20	−1.39	8.39

aData are from acute exacerbation of chronic obstructive pulmonary disease on day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of the hospital admission. Comparisons are between AE-1 and AE-3 or between AE-3 and AE-10.

Figure 3

Dynamic patterns of changes of gene expression of peripheral blood monocytes. Consistent decrease (A) or consistent increase (B), followed by a decrease (C), or a decrease followed by a recovery (D), in patients with acute exacerbation of chronic obstructive pulmonary disease (AECOPD) at day 1 (AE-1), day 3 (AE-3) and day 10 (AE-10) of hospital admission as compared with changes seen in patients with stable COPD.

Gene ontology analysis and pathway analysis

Within ten comparison pairs, up- or downregulated genes mainly involved in the biological process are shown in Figures S3 and S4 of Additional file 2, those in cellular components are shown in Figures S5 and S6 of Additional file 2 and those in molecular functions are shown in Figures S7 and S8 of Additional file 2. Additional file 5 lists gene numbers for ten comparison pairs with certain GO terms and different ranges of enrichment scores.

In the biological process, COPD-specific upregulated genes were involved mainly in peptide cross-linking, blood vessel development, biological adhesion or cell adhesion (Figure 4A). COPD-specific downregulated genes were involved mainly in T cell receptor signaling pathways, antigen receptor–mediated signaling pathways, immune response–activating cell surface receptor signaling pathways or steroid biosynthetic process (Figure 4B). AECOPD-specific genes upregulated in response to organic substance, response to wounding, multicellular organismal process or response to chemical stimulus are shown in Figure 4C. AECOPD-specific downregulated genes were involved mainly in the regulation of immune response and the immune system process or in the immune response and immune system process themselves (Figure 4D). In the cellular component, COPD-specific upregulated genes were involved mainly in the extracellular region, the extracellular matrix part, the proteinaceous extracellular matrix or the extracellular matrix (Figure 5A). COPD-specific downregulated genes were involved mainly in the major histocompatibility complex class II (MHC II) protein complex, microbody lumen, peroxisomal matrix or MHC II protein complex (Figure 5B). AECOPD-specific upregulated genes were involved mainly in the extracellular region part, the extracellular matrix, the extracellular space or the extracellular region (Figure 5C). AECOPD-specific downregulated genes were involved mainly in the cell periphery and the plasma membrane and were integral to the plasma membrane and intrinsic to the plasma membrane (Figure 5D). In molecular function, COPD-specific upregulated genes participated mainly in extracellular matrix structural constituent, platelet-derived growth factor binding, serine-type endopeptidase activity and protein binding (Figure 6A). COPD-specific downregulated genes were involved mainly in nucleoside kinase activity, MHC class II receptor activity, C-acyltransferase activity and ephrin receptor activity (Figure 6B). AECOPD-specific upregulated genes were involved mainly in protein binding, growth factor binding, calcium ion binding and polysaccharide binding (Figure 6C). AECOPD-specific downregulated genes were involved mainly in receptor activity, signaling receptor activity, molecular transducer activity and signal transducer activity (Figure 6D).

Figure 4

Gene expression profile comparisons regarding the biological process. Graphs describe co–differentially upregulated genes (A) and downregulated genes (B) in the biological process of peripheral blood mononuclear cells from patients with chronic obstructive pulmonary disease (COPD), including those with stable COPD and acute exacerbation of COPD (AECOPD), as compared to healthy control subjects. Also shown are co–differentially expressed upregulated genes (C) and downregulated genes (D) from patients with AECOPD, as compared to patients with stable COPD and healthy control subjects.

Figure 5

Gene expression profile comparisons regarding the cellular component. Graphs describe co–differentially upregulated genes (A) or downregulated genes (B) in the cellular component of peripheral blood mononuclear cells from patients with chronic obstructive pulmonary disease (COPD), including stable COPD and acute exacerbation of COPD (AECOPD), as compared to healthy control subjects. Also shown are co–differentially expressed upregulated genes (C) or downregulated genes (D) from patients with AECOPD, as compared to both patients with stable COPD and healthy control subjects. MHC, Major histocompatibility complex.

Figure 6

Gene expression profile comparisons regarding molecular function. Graphs describe co–differentially upregulated genes (A) or downregulated genes (B) in the molecular function of peripheral blood mononuclear cells from patients with chronic obstructive pulmonary disease (COPD), including stable COPD and acute exacerbation of COPD (AECOPD), as compared to healthy control subjects. Also shown are co–differentially expressed upregulated genes (C) or downregulated genes (D) from patients with AECOPD, as compared to both patients with stable COPD and healthy control subjects. MHC, Major histocompatibility complex.

COPD-specific upregulated genes also participated in extracellular matrix receptor interaction, protein digestion and absorption, focal adhesion and the phosphatidylinositol 3-kinase-Akt signaling pathway (Figure 7A). AECOPD-specific upregulated genes participated in Chagas disease, complement and coagulation cascades, pertussis and Staphylococcus aureus infection (Figure 7B). AECOPD-specific downregulated genes participated in antigen processing and presentation, natural killer cell–mediated cytotoxicity, graft-versus-host disease and thyroid cancer (Figure 7C).

Figure 7

Gene expression profile comparisons regarding signaling pathways. Graphs describe co–differentially upregulated genes (A) in different pathways of peripheral blood mononuclear cells from patients with chronic obstructive pulmonary disease (COPD), including patients with stable COPD and acute exacerbation of COPD (AECOPD), as compared to healthy control subjects. Also shown are co–differentially expressed upregulated genes (B) or downregulated genes (C) from patients with AECOPD, as compared to patients with stable COPD and healthy control subjects. ECM, Extracellular matrix; MHC, Major histocompatibility complex; Pi3K, Phosphatidylinositol 3-kinase.

Discussion

PBMCs play a critical and important role in the occurrence of AECOPD, owing to less capacity for balancing the proinflammatory immune response caused by infection and for secreting adequate amounts of anti-inflammatory cytokines [22]. The fact that patients with COPD are more susceptible to acute exacerbation has been suggested to be associated with PBMC dysfunction and failure of adaptation to infection, stimuli or hypoxia, although there have been not yet studies on the phenotypes of PBMCs in AECOPD. For example, PBMCs from patients with COPD could not induce hypoxia-inducible factor 1 and vascular endothelial growth factor, owing to a reduction in histone deacetylase 7 under hypoxic condition [23]. It was suggested that overproduction of proinflammatory cytokines (CXCL6 and interleukin 6 (IL-6)) from human PBMCs could be stimulated by the infection through activation of Toll-like receptor 4, nicotinamide adenine dinucleotide phosphate oxidase phosphatidylinositol 3-kinase and nuclear factor κB [24], at least as partial mechanisms by which PBMCs may be involved in the occurrence of AECOPD. The present study provides initial evidence that dynamic alterations of PBMC genetic phenotypes occurred in patients with AECOPD after their hospital admission and during their hospital stay.

Gene expression profiles of PBMCs were investigated in patients with COPD, compared with healthy controls and correlated with lung function measurement [12]. Differential expression of 45 known genes was identified, of which 16 markers had significant correlation with quantitative traits and differential expression between cases and controls and 2 genes, RP9 and NAPE-PLD, were identified as decreased in patients with COPD, as compared to controls, in both lung tissue and blood. Gene expression profiles of PBMCs were recently identified and validated in smokers with and without COPD and corrected with clinical phenotypes such as sex, age, body mass index, family history, smoking status and pack-years of smoking [25]. Of them, 16 candidate genes were found to be associated with airflow obstruction and secondary clinical phenotypes, 12 with emphysema, 13 with gas trapping and 8 with distance walked. Both previous studies demonstrated the gene expression profiles of PBMCs from patients with stable COPD and addressed the potential significance of smoking. In the present study, we selected healthy control subjects and patients who were not current smokers and demonstrated gene expression profiles of PBMCs from patients with COPD, including stable COPD and AECOPD. We addressed COPD-specific gene expression profiles that should appear in both stable COPD and COPD exacerbation conditions and found COPD-specific 79 genes were upregulated and 23 genes down-regulated more than fivefold as compared with gene expression in controls. In the present study, we selected consistent up- or downregulated gene expression on days 1, 3 and 10 of AECOPD-specific as compared with gene expression in both healthy controls and patients with stable COPD, as AECOPD-specific gene expression profiles. We found that 58 AECOPD-specific genes were upregulated more than fivefold and 238 genes were downregulated more than twofold, as compared to both control subjects and patients with stable COPD.

Variation of gene expression profiles is dependent upon multiple uncontrollable factors, such as study population, age, history, genetic background and treatment. In addition, gene expression profiles vary between harvested sample types, such as sputum, bronchoalveolar lavage fluid, blood or lung tissues. For example, 102 genes were identified to distinguish between non- or mild emphysema and severe emphysema in lung tissue [15] and to distinguish 70 microRNAs and 2,667 mRNAs between smoking patients with or without COPD [26]. In the present study, we investigated gene expression profiles of PBMCs from control subjects, patients with stable COPD, and patients with AECOPD on day 1, day 3 and day 10 of hospital admission, and we found about 3,000 overexpressed genes and 2,000 downregulated genes in patients with stable COPD or AECOPD, as compared with control subjects. These findings indicate that those COPD-specific genes exist in the stable COPD condition and during acute exacerbations of COPD.

Of the COPD-specific genes we studied, CEACAM1, COL6A3, NOL3, COL1A2, MLPH, MUC1, P8, UNQ473, CLDN4, RNASE1, H19, DEFA1 and LOC653600 were upregulated more than tenfold, mainly related to nuclear proteins, collagens or molecular structure. We noted that transcription factor CP2 (TFCP2L1) and SCP2 were downregulated more than tenfold. In previous studies, these genes, including CEACAM1, TFCP2L1 and SCP2, were not found to be associated with COPD. The SCP2 gene is located within chromosome 1 and encodes the nonspecific lipid transfer protein SCP2, which is involved in organellar fatty acid metabolism [27,28] and the translocation of cytoplasmic free cholesterol to the mitochondria [29]. Our results indicate that PBMCs from patients with stable COPD or AECOPD had downregulated SCP2, which might point to severe metabolic disorder and thus that SCP2 downregulation might contribute to one of the common comorbidities of COPD [30]. TFCP2 is a member of a family of transcription factors that regulate genes involved in events from early development to terminal differentiation [31]. PBMCs with downregulated TFCP2 of patients with COPD might have less capacity of the transcriptional switch of globin gene promoters, many other cellular and viral gene promoters, or interaction with certain inflammatory response factors, although the exact mechanism and pathological role remain unclear.

AECOPD-specific gene expression profiles were selected by comparing them with both healthy control subjects and patients with stable COPD, including 647 upregulated genes and 238 downregulated genes (greater than twofold upregulation). Of them, FOS, IFI27, CYR61, CTGF, GPRC5A, FOSB, DCN and LOC387763 were upregulated more than tenfold and KIR2DS2, SH2D1B, CD8B, OR2W5, KSP37 and TCF7 were downregulated more than threefold.

We noticed that some genes, such as FOS, CYR61 and CTGF, were upregulated in PBMCs from patients with either stable COPD or AECOPD, consistent with the lung tissue gene expression profiles of patients with COPD or smokers, in whom the genes were expressed mainly in alveolar epithelial cells, airway epithelial cells and stromal and inflammatory cells [14]. Other genes, including GPRC5A, LOC387763 and KIR2DS2, were not found to be associated with AECOPD in previous publications. CTGF is a cysteine-rich peptide implicated in several biological processes, such as cell proliferation, survival and migration, and involved in pulmonary vascular remodeling and hypertension in COPD. It was evidenced by the experimental finding that CTGF short-hairpin RNA could significantly prevent CTGF and cyclin D1 expression, arrest cell cycle at the G0/G1 phase, suppress cell proliferation in smoking-exposed pulmonary smooth muscle cells and ameliorate pulmonary vascular remodeling [32]. Another study demonstrated that some inflammatory genes (IL-1β, IL-6, IL-8, CCL2 and CCL8) were upregulated, whereas some growth factor receptor genes (BMPR2, CTGF, FGF1, KDR and TEK) were downregulated in lung tissue samples from patients who were current smokers or had moderate COPD [33].

Downregulation of TCF7 was found in PBMCs of patients with COPD and current smoking and was correlated with some clinical phenotypes, such as emphysema, gas trapping and distance walked [25]. In the present study, we also found that TCF7 was downregulated in ex-smokers with COPD by about an absolute threefold compared with control subjects, and, in patients with AECOPD, TCF7 was downregulated by about an absolute tenfold compared with both control subjects and patients with stable COPD. These findings indicate that TCF7 not only is a COPD-specific gene but also is associated with the severity of the disease. TCF7 is a member of a family of HMG box containing factors associated with β-catenin to mediate Wnt signaling, controls the switch between cell self-renewal and differentiation and plays a role in B cell and T cell development. TCF7 was found to be the most downregulated transcription factor when CD34+ cells switched into CD34− cells through a coordinated regulation of the binding between TCF7 and the short isoforms of RUNX1 [34]. It is possible the downregulation of TCF7 and associated regulation may be one part of molecular mechanism of PBMC incapacity during AECOPD.

Dynamic alterations of gene expression profiles in patients with AECOPD were evaluated with dynamic DESS scores. ALAS2, EPB42 and CA1 were co–differentially expressed with a down–down type in patients with AECOPD. Among these three genes, the CA1 gene encodes a protein which is important in respiratory function, fluid secretion and maintenance of cellular acid–base homeostasis [35]. The genes with a down–up type included SELENBP1, MYH9 and an unnamed gene in chromosome 19, both of which are associated with psychotic disorders [36,37]. One limitation of the present study is the small sample size, which detracts from the generalizability of the results presented.

Conclusions

Dynamic alterations of PBMC gene expression profiles were initially investigated in patients with AECOPD, as compared with healthy control subjects or patients with stable COPD. A panel of genes, including eight that were upregulated and eight that were downregulated, were recommended as AECOPD-specific dynamic biomarkers. AECOPD-specific up- or downregulated genes in the biological process, cellular components or molecular function were defined and participated in complement and coagulation cascades, infection, antigen processing and presentation, natural killer cell–mediated cytotoxicity, and/or cancer-causing potential. The integration of dynamic bioinformatics with clinical phenotypes helped us to identify and validate AECOPD-specific biomarkers to help define the severity, duration and response of the disease to therapies.

Key messages

Circulating dynamic biomarkers were identified for the specificity and severity of AECOPD.

A panel of 16 genes were selected as AECOPD-specific biomarkers.

This is an initial study designed to examine gene expression profiles of peripheral blood mononuclear cells and identify dynamic changes of AECOPD-specific biomarkers.