Enhanced Insulin-Like Growth Factor Binding Protein–Related Protein 2 (Connective Tissue Growth Factor) Expression in Patients with Idiopathic Pulmonary Fibrosis and Pulmonary Sarcoidosis Jeremy T. Allen, Richard A. Knight, Claire A. Bloor, and Monica A. Spiteri Lung Injury and Inflammation Research Group, Department of Respiratory Medicine, North Staffordshire Hospital, Stoke-on-Trent, United Kingdom
Connective tissue growth factor is a recently described chemoattractant and fibroblast mitogen which, because of sequence homology and weak binding to insulin-like growth factor (IGF)-1, has been proposed as the eighth member of the IGF binding protein (IGFBP) superfamily, named IGFBP-related protein 2 (IGFBP-rP2). Previous studies have implicated IGFBP-rP2 in a number of heterogeneous fibrotic pathologies, including renal fibrosis, dermal scleroderma, and bleomycin-induced pulmonary fibrosis in mice. Because profibrogenic cytokines may be produced by inflammatory cells, we developed a multiplex competitive reverse transcription/polymerase chain reaction to quantify IGFBP-rP2 transcripts in bronchoalveolar lavage cells from healthy subjects and patients with idiopathic pulmonary fibrosis (IPF) and pulmonary sarcoidosis. IGFBP-rP2 messenger RNA expression was enhanced . 10-fold (P , 0.003) in patients with IPF; . 40-fold (P , 0.006) in stage I/II sarcoidosis patients, and . 90-fold (P , 0.005) in stage III/IV sarcoidosis patients by comparison with healthy nonsmoking control subjects. We suggest these increases are predominantly associated with lymphocyte- and neutrophil-driven IGFBP-rP2 production. These findings, together with previous reports implicating other IGFBPs in the pathogenesis of pulmonary fibrosis, suggest that the complex network of IGFBPs within the human lung is an important determinant of the outcome of the fibroproliferative response to injury. Allen, J. T., R. A. Knight, C. A. Bloor, and M. A. Spiteri. 1999. Enhanced insulin-like growth factor binding protein–related protein 2 (connective tissue growth factor) expression in patients with idiopathic pulmonary fibrosis and pulmonary sarcoidosis. Am. J. Respir. Cell Mol. Biol. 21:693–700.
Connective tissue growth factor (CTGF) is a recently described, 38-kD, cysteine-rich peptide that is a potent enhancer of fibroblast proliferation and collagen deposition. CTGF is a member of a closely related family of genes encoding immediate early gene products, including nov, an avian proto-oncogene, and cyr61/cef10 nephroblastoma overexpressed gene (1, 2). On the basis of sequence homology toward the amino terminus of CTGF with insulinlike growth factor binding proteins (IGFBPs), together
(Received in original form March 8, 1999 and in revised form June 9, 1999 ) Address correspondence to: Jeremy T. Allen, Dept. of Respiratory Medicine, North Staffordshire Hospital, Newcastle Road, Stoke-on-Trent ST4 6QG, UK. E-mail:
[email protected] Abbreviations: bronchoalveolar lavage, BAL; BAL cell, BALC; BAL fluid, BALF; base pair(s), bp; complementary DNA, cDNA; competitive template, CT; connective tissue growth factor, CTGF; glyceraldehyde-3phosphate dehydrogenase, G3PDH; insulin-like growth factor, IGF; IGF binding protein, IGFBP; IGFBP-related protein, IGFBP-rP; idiopathic pulmonary fibrosis, IPF; messenger RNA, mRNA; reverse transcription/ polymerase chain reaction, RT-PCR; transforming growth factor, TGF. Am. J. Respir. Cell Mol. Biol. Vol. 21, pp. 693–700, 1999 Internet address: www.atsjournals.org
with its low-affinity binding of insulin-like growth factor (IGF)-1, CTGFs with nov and cyr61 have therefore been proposed as members of the IGFBP superfamily (3). Provisionally named IGFBP-8, CTGF and IGFBP-7 have now been designated as IGFBP-related proteins (IGFBP-rP) pending further investigations, with CTGF named IGFBPrP2 (4). IGFBP-rPs appear to have predominantly IGFindependent actions in contrast to the classic IGFBP-1– IGFBP-6 (5). In fibroblasts, IGFBP-rP2 is expressed as both a cellassociated glycosylated 38-kD form and as a smaller 10- to 12-kD secreted product that retains biologic activity (6). Potential proteolytic 18- and 24-kD degradation products have also been found in a variety of biologic fluids (7). The mitogenic effects of transforming growth factor (TGF)-b on fibroblasts appear to be at least partially mediated by IGFBPrP2 (7), and the IGFBP-rP2 promoter contains a consensus TGF-b response element (8). Glucocorticoids such as dexamethasone also induce IGFBP-rP2 in fibroblastoid cell lines, an effect that is independent of TGF-b but which could be blocked by tumor necrosis factor (TNF)-a (9). Pulmonary fibrosis in humans has been associated with increased expression of TGF-b and it is therefore possible
694
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
that IGFBP-rP2 is a final fibrogenic mediator. Increased TGF-b and IGFBP-rP2 have been demonstrated in bleomycin-induced pulmonary fibrosis in a murine model (10). A number of recent studies have shown increased IGFBPrP2 to be a common feature in the pathogenesis of fibrosis in a heterogeneous group of disorders, including inflammatory bowel disease (11), skin lesions from scleroderma, and systemic sclerosis (12), and in patients with renal fibrosis (13). Other IGFBPs have been implicated in the fibroproliferative response associated with the development of pulmonary fibrosis. We have previously demonstrated increased 29-kD IGFBP-3 in pulmonary sarcoidosis patients with fibrosis (14). IGFBP-3 expression is increased in idiopathic pulmonary fibrosis (IPF) (15), and IGFBP-2 expression is enhanced in children with pulmonary sarcoidosis and IPF (16). Inflammatory cells in the lung are crucial in the deployment of fibrogenic mediators such as IGF-1 and TGF- b. Accordingly, we hypothesized that any potential increase in IGFBP-rP2 expression would be reflected within the bronchoalveolar lavage cells (BALCs) of patients with pulmonary fibrosis.To evaluate this we developed a quantitative multiplex reverse transcription/polymerase chain reaction (RT-PCR) for both IGFBP-rP2 and the housekeeper gene glyceraldehyde-3-phosphate dehydrogenase (G3PDH). We compared BALC samples from healthy nonsmokers, patients with IPF, and patients having pulmonary sarcoidosis, both with and without evidence of fibrotic lung disease.
these patients had any documented evidence of previous respiratory disease. In each patient group, individuals were categorized as having never smoked or as having stopped smoking. In all individuals described as ex-smokers, the period since they had last smoked was . 5 yr and as much as 25 yr in some cases. Thus, for the purposes of analysis, the ex-smokers were combined with those having never smoked. None of the patients or normal subjects were current smokers at the time of this study. The control group comprised seven healthy volunteers (5 males, 2 females, mean age 30 yr, range 21–42 yr). Five of the seven were lifelong nonsmokers; the remaining two had smoked socially (approximately 10 cigarettes/wk) in the past but stopped . 5 years before enrollment. All had normal pulmonary function. None had suffered any respiratory symptoms in the month preceding the study. Prior approval for this study was obtained from the Research Ethics Committee of the North Staffordshire Hospital. All subjects gave informed written consent.
Materials and Methods
Sample Processing BALCs were centrifuged at 400 3 g/48C/5 min, homogenized in Trizol (Life Technologies, Paisley, UK) and stored at 2808C. RNA extraction and complementary DNA (cDNA) synthesis were performed according to previously described methods (14).
Subject Group A total of 13 patients (5 males, 8 females, mean age 45 yr, range 29–63 yr) with biopsy-proven sarcoidosis, presented with acute symptoms of cough, dyspnea, and/or chest pain. Radiographic studies (conventional chest X-ray and highresolution computed tomography [HRCT] scan) confirmed that eight sarcoid patients had radiologic stages I/II with no evidence of parenchymal fibrosis, and that five sarcoid patients had stages III/IV with bilateral nodular interstitial infiltrates and areas of fibrosis, honeycombing, and traction bronchiectasis. Three patients with stage I sarcoid had normal lung function; the remaining 10 patients had significant reductions in their lung volumes and carbon monoxide transfer factor. All 13 patients had high serum angiotensin converting enzyme levels, mean 95 6 21 IU/liter (normal range 18–55). A total of 18 patients (10 males, 8 females, mean age 64 yr, range 46–81 yr) with IPF presented with cough and/or dyspnea (duration 7 6 4 mo), bilateral basal to midzone crackles, and radiographic bilateral interstitial fibrosis, supported by pathognomonic HRCT scan patterns of varying degrees of reticular infiltrates, diffuse honeycombing, and sparse ground-glass pacification in a predominantly subpleural and basal distribution. Of the 18 patients, 14 had lung biopsies, with histologic confirmation of usual interstitial pneumonia. All 18 IPF patients showed significant reduction in total lung capacity and transfer factor. None of the sarcoidosis and IPF patients had received prior steriod or immunosuppressive therapy. None of
BAL BAL was performed on all subjects during fiberoptic bronchoscopy under local anesthesia. Briefly, the right middle lobe was instilled with successive aliquots of sterile 0.9% isotonic saline, to a volume of 180 ml. The lavage fluid aliquots were immediately gently aspirated by suction, keeping the dwell time to a minimum; collected into sterile siliconized glass bottles; and kept at 48C for processing (14).
PCR Primers Specific PCR primers were developed with computer assistance (Primer, version 0.5; Whitehead Institute for Biomedical Research, Cambridge, MA) for IGFBP-rP2 and G3PDH genes. Oligonucleotides (Table 1) were synthesized commercially (Cruachem Ltd., Glasgow, UK). Competitive template (CT) primers were prepared using the protocol of Celi and colleagues (17). Primers for both the native and CT amplifications were optimized for use under the same PCR conditions, and were designed such that when native sense and CT primers were used in a PCR, they would give rise to shorter CT PCR products that could be separated sufficiently from their respective native products on a 2% agarose gel. CT and native sense primers were the same, whereas the CT antisense primers were specifically synthesized to produce CT products. They were a hybrid of the native antisense primer sequence at the 59 end joined with an internal antisense sequence (Table 1). Thus, the CT primer can bind with either the 3 9 or the internal sequence of the cDNA but amplification can occur only from the internal site. The resultant CT PCR product, therefore, is shorter than native but contains the native antisense primer sequence, allowing subsequent am-
Allen, Knight, Bloor, et al.: Increased IGFBP-rP2 in IPF and Sarcoidosis
695
TABLE 1
Primers for multiplex PCR amplification Gene
IGFBP-rP2*
G3PDH†
Primer
Sequence
Product
Sense Antisense CT Sense Antisense CT
59-GGC CTC TTC TGT GAC TTC GG-39 59-TGC AGG AGG CGT TGT CAT T-39 59-AS1GAC CAG GCA GTT GGC TCT AAT C-39 59-TGA AGG TCG GAG TCA ACG GAT TTG GT-39 59-CAT GTG GGC CAT GAG GTC CAC CAC-39 59-AS1CGT CAA AGG TGG AGG AGT GGG TGT-39
N 451 bp S 388 bp N 983 bp S 903 bp
AS: antisense sequence; N: native; S: synthetic. *Genbank accession J04038. † Genbank accession M92934.
plification with the native primer pair. With 2 ml undiluted cDNA, a minimum of 10218 M target gene CT could be coamplified and visualized on an ethidium bromide–stained agarose gel using the selected primers. Competitive Template Preparation CT PCR reactions consisted of 25 pmol primers (sense plus CT), 200 mM deoxynucleotide triphosphates, 1.5 mM MgCl2, 5 ml 103 PCR buffer, and 2.5 U Amplitaq Gold (Perkin-Elmer, Warrington, UK) in 50-ml total. A “Hot Start” PCR was employed in which cycling conditions were: 958C enzyme activation for 12 min; then 1 min 15 s at 948C, 1 min at 558C, and 1 min at 728C for 40 cycles; then a final extension of 10 min at 728C. cDNA from a bronchial epithelial cell line (BEAS-2B) known to express IGFBPrP2 was included. Two separate reactions were combined and electrophoresed through a 2% Seakem agarose gel (Flowgen Instruments, Lichfield, UK) and stained with 2 mg/ml ethidium bromide. Predicted PCR products were excised in a gel slice and extracted using Microspin S-300 HR columns (Pharmacia Biotech, St. Albans, UK). The concentration of purified CTs (in distilled H2O) was quantified by electrophoresis, as described previously, of 2 or 4 ml of CT adjacent to a 15 to 50-ng range of a size standard, and subsequent visualization and densitometric analysis of bands performed using the Gel-Doc1000 system and Molecular Analyst software (Bio-Rad, Hemel Hempstead, UK). Size standards were prepared from pBR322 (Life Technologies) cut with restriction enzymes (Roche Diagnostics, Lewes, UK) HindIII to give 909 base pairs (bp) (for 903-bp G3PDH CT) or BamHI and EcoRI to give 375 bp (for 388-bp IGFBP-rP2 CT). CT mixtures and dilutions thereof, containing known amounts of both IGFBP-rP2 and G3PDH CTs, were prepared from the same original stock solutions according to the method employed by Willey and coworkers (18). After preliminary experiments, CT mixtures were prepared to contain G3PDH (G) and CTGF/IGFBP-rP2 (C) at the following concentrations [M]: G 211/C212; G211/C213; G211/ C214; G211/C215; G211/C216; G212/C213; G212/C214; G212/ C215; G212/C216. A total of 5 ml of the required CT mix was included in each 50-ml PCR; thus, the final CT molarity was 10-fold lower than the above concentrations. There was no interference between the CTs because primers for
each gene amplified only a single band from the CT mix when no native cDNA was present. Quantitative Multiplex RT-PCR All PCRs were performed in duplicate. Each reaction was 50-ml volume and contained 25 pmol native primers for both G3PDH and IGFBP-rP2, 5 ml of the appropriate CT molarity, 2 ml cDNA, and an aliquot of a master mix containing PCR reagents as described in C OMPETITIVE TEMPLATE PREPARATION. PCR cycling conditions and electrophoresis were performed as described previously. Selection of the appropriate CT mixture was based on preliminary experiments using the strategy of Willey and associates (19) to establish the relative cDNA concentration in a sample and on the relative expression of IGFBP-rP2 to G3PDH. In samples used in this study, 2 ml cDNA would compete equally with either 104 or 105 molecules of G3PDH in the PCR. For IGFBP-rP2, the appropriate amount of CT ranged from 101 to 105 molecules in the PCR. PCR products were quantified using digital image analysis. Bands analyzed were nonsaturating and therefore only within the dynamic linear range of the system. They were quantified on the basis of total ethidium bromide staining. Ethidium bromide staining intensity was dependent on both the number and size of molecules present in each band. Thus, it was necessary to employ a correction factor to account for the differences in size between native and CT products. Staining intensity of the native band was therefore corrected to the the size of the CT band. Gene expression was then calculated according to equation (1), in which iN is the initial number of native molecules; ND is the native band density; CTD is the CT band density; CTs is the CT size in bp; Ns is the native size in bp; and iCT is the initial number of CT molecules. ND × CTs ⁄ Ns iN = -----------------------------------CTD × iCT
(1)
Results were then expressed as the number of molecules of IGFBP-rP2 messenger RNA (mRNA) per 10 6 molecules of G3PDH mRNA. PCR Product Analysis Specificity of both native IGFBP-rP2 and G3PDH PCR products was verified by their predicted sizes of 451 and 983 bp, respectively. In addition, IGFBP-rP2 product was
696
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
restricted at the unique internal SphI (Roche Diagnostics) site, giving predicted 283 and 168 bp fragments. Detection of genomic DNA contamination in RT-PCR has been extensively discussed elsewhere (19). A number of measures were therefore taken to exclude this possibility. Because the genomic sequence of G3PDH was known, it was possible to utilize primers that spanned introns in the genomic template, and any amplified PCR product deriving from genomic DNA could be distinguished by its increased size. Absence of such a band would suggest that there was no significant contamination of the RNA sample with genomic DNA. The genomic structure of IGFBP-rP2 remains unknown, therefore it was not possible to ensure that IGFBP-rP2 primers spanned the intronic sequence. Thus, it was possible that the genomic and cDNA templates were the same and, in circumstances of low expression of the gene, some interference from genomic DNA might occur. To check this, the size of the genomic IGFBP-rP2 template was determined. Genomic DNA was isolated from BEAS-2B cells using a Rapidprep Kit (Pharmacia Biotech) according to the supplier’s instructions and 1 or 2 ml (of 50 ml) was used in a PCR, performed as already described with native IGFBP-rP2 primers. Data Analysis Statistical analyses were carried out using Stata software (Intercooled Stata, version 6; Timberlake Consultants, West Wickham, UK). Analysis of variance was performed using the nonparametric Kruskal–Wallis test for differences between the four groups (Normal, Sarcoid I/II, Sarcoid III/IV, and IPF). The effect on IGFBP-rP2 expression of age, sex, cell type (alveolar macrophages, neutrophils, lymphocytes, eosinophils), and cell number (cells per milliliter of BAL fluid [BALF]) were evaluated. When this test indicated a significant difference (P , 0.05), each pairing was examined using the Mann-Whitney U test. Where a significant difference (P , 0.05) was found, a test for correlation was performed using Spearman’s rank correlation coefficient.
Results BAL Findings The cellular characteristics of each of the subject groups are presented in Table 2. In comparison with normal nonsmokers, patient groups showed significantly increased cellularity. IPF patients were characterized by significant increases in neutrophil (20% of total BALCs) and eosinophil numbers (P , 0.0007 and P , 0.002, respectively). Stage I/II sarcoidosis patients showed a classic predominant lymphocytic alveolitis (P , 0.02) (26% of total BALCs), with no evidence of neutrophils. A significant increase in eosinophil numbers was observed (P , 0.01), largely due to only one of the eight patients in this group. In contrast, stage III/IV sarcoidosis patients were characterized by a significant influx of neutrophils (P , 0.02) (24% of total BALCs), possibly reflecting the emergence of fibrosis in their lungs, in addition to a significant lymphocytosis (P , 0.02) (36% of total BALCs). The percents of BAL returns were diminished in patients, particularly in the IPF group, although these did not reach significance. The viability of recovered BALCs before RNA extraction was . 90% as assessed by trypan blue dye exclusion. IGFBP-rP2 expression could not be correlated with sex or age in any of the subject groups. In addition, both the cell type, number (cells/ml BALF), and proportion (% of total BALF) were examined for any effect on IGFBP-rP2 expression. IGFBP-rP2 expression could not be correlated directly with alveolar macrophage, lymphocyte, neutrophil, or eosinophil counts, either absolute or proportional. Multiplex RT-PCR The results of the multiplex RT-PCR amplification obtained for each sample are presented in Figure 1. The optimal CT molarity used and the resultant gel picture are shown, which were then used for the quantification. Each multiplex PCR typically produced the pattern of bands shown in more detail in Figure 2. For G3PDH, bands of 983 and 903 bp were identified; and for IGFBP-rP2, bands
TABLE 2
Summary of BALC findings* Subject Group (n)
Normal (7) IPF (18) Sarcoid [I 1 II] (8) Sarcoid [III 1 IV] (5)
% (bold type) and Absolute Number† (normal type) of Total BALCs
BAL Return (%)
BAL Cs/ml (3 103)
66 6 5
66 6 14
52 6 3
207 6 47‡
54 6 5
112 6 23‡
59 6 5
149 6 56‡
AM
95.6 6 0.9 63.1 6 0.6 63.8 6 6.1‡ 128.1 6 39 62.0 6 6.6‡ 69.1 6 17 39.5 6 11‡ 37.7 6 16
N
0.6 6 0.2 0.4 6 0.1 19.6 6 5.9‡ 52.4 6 21‡ 3.1 6 2.2 2.4 6 1.6 23.8 6 19 66.6 6 62‡
Definition of abbreviations: AM, alveolar macrophages; E, eosinophils; L, lymphocytes; N, neutrophils. * Mean 6 standard error of the mean. † Nucleated cells/ml 3103. ‡ Significant difference (P , 0.05) from normal subjects.
L
E
3.8 6 0.8 2.5 6 0.5 10.5 6 2.5 12.3 6 3.1 25.9 6 6.9‡ 27.6 6 11‡ 36.2 6 10 43.8 6 26‡
0 0 6.0 6 2.3‡ 14.3 6 8.3‡ 9.0 6 8.7‡ 13.0 6 12.8‡ 0.5 6 0.3‡ 0.9 6 0.9
Allen, Knight, Bloor, et al.: Increased IGFBP-rP2 in IPF and Sarcoidosis
697
Figure 2. Multiplex RT-PCR products. Ethidium bromide– stained agarose gel showing the results of a multiplex experiment for a single sample. Lane 3 contains a 100-bp molecular size marker, running top to bottom, with the positions of the 400- and 1,000-bp bands indicated. In lanes 1 and 2 the bands for G3PDH and IGFBP-rP2 are arrowed. The native PCR product is denoted by N and the CT-PCR product by S. For G3PDH the sizes are N 5 983 bp and S 5 903 bp, and for IGFBP-rP2 N 5 451 bp and S 5 388 bp. Lanes 1 and 2 both contain G3PDH CT at 10212 M, but lane 1 contains IGFBP-rP2 CT at 10217 M and lane 2 IGFBPrP2 CT at 10216 M.
ucts, respectively. IGFBP-rP2 native PCR product was subjected to a restriction analysis with SphI; these results are shown in Figure 3. The digested 451-bp IGFBP-rP2 product gave fragments corresponding with the predicted 283- and 168-bp sizes, thereby confirming the identity of the PCR product. Quantitative Analysis of Gene Expression IGFBP-rP2 expression in each sample relative to 10 6 G3PDH molecules is shown, grouped by subject, in Figure 4. IGFBP-rP2 expression could be quantified in all samples. However, there was considerable inter-individual variation within the subject groups. Among the normal subjects (Figure 4a) this variation was relatively small, Figure 1. Summary of results from quantitative multiplex RT-PCR experiments. The first column lists the subject code number for each sample, with molecular size markers (MW) at the top and bottom of the figure. The second column denotes the subject group: normal nonsmokers (N); sarcoidosis, stages I to IV (S1 to S4); or IPF (F). The third column lists the optimal CT molarity used to perform the quantitative analysis. A range of CT mixtures was used to optimize the native/CT ratio of PCR products for IGFBPrP2 and G3PDH. The indicated molarity represents the molarity of the CT in the PCR (G denotes G3PDH and C denotes CTGF/IGFBP-rP2). Amplicons were electrophoresed through 2% Seakem agarose gels and digital image analysis was performed on the resultant bands pictured in the figure, as described in MATERIALS AND METHODS. The top and bottom images depict the 100-bp molecular size markers for these gels, running from left to right, with the 600bp marker visible as the darkest band.
of 451 and 388 bp can be seen. These correspond with the predicted sizes calculated from the published sequences and given in Table 1 for their native and CT PCR prod-
Figure 3. IGFBP-rP2 PCR product specificity. Ethidium bromide– stained agarose gel showing the results for native-only IGFBP-rP2 PCR and restriction analysis of the product. Lane 1 shows the single 451-bp band obtained for amplification of sample cDNA with native IGFBP-rP2 PCR primers and no CT. Lane 2 is a negative control. Lane 3 shows the restriction digest of the amplicon from lane 1 with SphI that has a unique cutting site in the amplified sequence. The predicted sizes of the resultant fragments are 283 and 168 bp. Lane 4 contains 100-bp molecular size marker, running bottom to top, with 100- and 600-bp bands indicated.
698
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
only 7-fold. Inter-individual variations were very much higher among patients, those with IPF (Figure 4c) reaching 100-fold and in those having sarcoidosis (Figure 4b), 2,000-fold. Replicate measurements were taken of G3PDH expression in five different samples to evaluate reproducibility. The standard deviation was found to be 10 copies IGFBPrP2 in a 50-ml PCR. The mean levels of IGFBP-rP2 expression in each of the subject groups are provided in Table 3. In comparison with normal nonsmokers, the mean IGFBP-rP2 expression levels were significantly increased in patient groups. IGFBP-rP2 was upregulated . 10-fold in IPF patients (P , 0.003), and in sarcoidosis patients was increased . 40-fold in stage I/II patients (P , 0.006) and . 90-fold in stage III/ IV patients (P , 0.005). Genomic DNA Contamination No band corresponding to the genomic DNA template was observed in any of the samples for the G3PDH gene. Purified genomic DNA isolated from human bronchial epithelial cells was used to determine the size of the IGFBPrP2 genomic DNA template. After PCR with native IGFBP-rP2 primers and agarose gel electrophoresis a single band was observed, estimated to be 808 bp in size (gel not shown) and corresponding to the IGFBP-rP2 genomic template, compared with the 451-bp cDNA amplicon. Thus, the IGFBP-rP2 primers flank one or more introns in the genomic sequence, allowing us to clearly identify any contaminating genomic DNA. Although there was no evidence of genomic DNA in any samples, these results confirmed that the presence of genomic DNA would not interfere with the competitive RT-PCR.
Discussion The data presented in this report are the first to show dramatic increases in IGFBP-rP2 mRNA in BALCs of patients with IPF and pulmonary sarcoidosis. Using a sensitive quantitative multiplex RT-PCR method, we have shown significant increases of IGFBP-rP2 transcripts of . 10-fold in patients with IPF and of . 40-fold for patients with stage I/II pulmonary sarcoidosis or of . 90-fold for patients with stage III/IV pulmonary sarcoidosis. Cultured normal human and mouse lung fibroblasts constitutively express IGFBP-rP2 and this expression is in-
Figure 4. Quantitative analysis of IGFBP-rP2 gene expression. IGFBP-rP2 expression data in samples from normal subjects (a), sarcoidosis subjects (b), and IPF subjects (c) are presented subdivided according to whether they have ever smoked. Open bars represent ex-smokers and hatched bars represent subjects who
have never smoked. Sarcoidosis patient disease status is indicated above the bars, graded I to IV accordingly. The bars represent the average of duplicate PCR, and the error bars the range. Individuals’ code numbers are given along the x-axis and the IGFBP-rP2 expression as mRNA transcripts per 10 6 G3PDH mRNA transcripts along the log scale y-axis. Data were obtained from image analysis of PCR products shown in Figure 1, according to the method described in MATERIALS AND METHODS.
Allen, Knight, Bloor, et al.: Increased IGFBP-rP2 in IPF and Sarcoidosis
TABLE 3
Summary of IGFBP-rP2 gene expression Subject Group (n)
mRNA Transcripts/106 G3PDH Molecules*
Normal (7) IPF (18) Sarcoid [I 1 II] (8) Sarcoid [III 1 IV] (5)
512 (6 97) 5,455 (6 2,560) 20,932 (6 11,714) 47,910 (6 24,505)
*Mean (6 standard error of the mean).
creased approximately 2-fold with TGF-b treatment. IGFBP-rP2 expression by fibroblasts from mice with bleomycin-induced lung fibrosis is similarly increased by TGF-b (10). Spontaneous IGFBP-rP2 expression is increased in a number of nonpulmonary fibroproliferative diseases, such as systemic sclerosis, inflammatory bowel disease, and renal fibrosis where TGF-b levels are also increased (11–13). However, in BALCs the upregulation in IGFBP-rP2 appears much more striking than in the mesenchymal cell types from these previous studies. Thus, it is possible that at least some of the fibrogenic effects attributed to TGF-b are mediated by IGFBP-rP2. In addition, evidence suggests that IGFBP-rP2 can be upregulated through a TGFb–independent mechanism as well (9). The binding and activation of fibroblasts by IGFBP-rP2 suggests that fibroblasts express a specific IGFBP-rP2 receptor similar to that described on chondrocytes (20) and that the fibrogenic effects of BALC-derived IGFBP-rP2 on fibroblasts is exerted in a paracrine manner. The cell type(s) within the BALC population responsible for the IGFBP-rP2 expression are unknown and cannot be explained by simple correlation with cell counts. However, an examination of the tabulated BALC findings (Table 2) with clinical data leads us to propose at least two possible sources. First, IPF patient BALCs are characterized by a dramatic neutrophil influx, suggesting that these cells may represent a source of IGFBP-rP2. This increase in neutrophil numbers is even more striking in the stage III/IV pulmonary sarcoidosis patients with fibrosis. Second, the predominant lymphocytic alveolitis in the stage I/II pulmonary sarcoidosis patients, in which IGFBP-rP2 expression is enhanced beyond that seen for IPF patients, suggests lymphocyte-driven IGFBP-rP2 production. This is supported by the large lymphocyte presence in the stage III/IV pulmonary sarcoidosis patients as well, and suggests a cumulative effect on IGFBP-rP2 expression from lymphocytes and neutrophils in these patients. Circulating lymphocytes have been shown to express IGFBP-2 and IGFBP-3, and, after activation, IGFBP-5 as well (21). Further investigations will be necessary to identify the source(s) of IGFBP-rP2 more precisely. Although some of the patients had smoked at some time in their lives, all these had ceased at least 5 yr before this investigation. Cigarette smoking is one of the few epidemiologic risk factors for the development of pulmonary fibrosis, and therefore if smoking, through increased IGFBP-rP2 expression, were an initiating factor in fibrotic lung disease, it is clearly unnecessary for the sustained increases in IGFBP-rP2 observed in patients.
699
This study extends previous data implicating the IGF/IGFBP system in fibrotic disease. Thus, we have shown degradation of IGFBP-3 to a 29-kD fragment that no longer binds IGF-1 in the BALF of sarcoidosis patients, an observation that may account for the increased IGF-1– dependent fibroblast mitogenicity of sarcoidosis BALF (14). Other work has shown increased expression of IGFBP-2 in children with fibrotic lung diseases (16). IGFBP-3 and IGFBP-4 are differentially expressed in normal skin fibroblasts and the hypertrophic scar fibroblasts associated with wound-healing. In these cells, IGFBP-3 production mediated by TGF-b promotes excessive collagen deposition but can be inhibited by TNF-a (22), a clear parallel with IGFBP-rP2. In contrast, in epithelial cells TGF-b–mediated inhibition of proliferation is dependent upon IGFBP-3 induction (23), similar to the reduced epithelial cell proliferation that follows TGF-b induction of IGFBP-rP2 (3). IGFBP-rP1, which, like IGFBP-rP2, binds IGF-1 with low affinity, has been shown to be highly expressed in nondividing epithelial cells, suggesting it has an antiproliferative role (24). In contrast, in proliferating myogenic cells IGFBP-rP1 expression was also high, declining with differentiation, suggesting it has an antidifferentiation function (25). This raises the question of whether IGFBP-rP1 and IGFBP-rP2 are differentially regulated in the lung epithelium during the pathogenesis of fibrosis, where proliferation of type 2 cells and type 1 cell differentiation is crucial to re-epithelialization during lung repair, and in fibroblast proliferation and subsequent differentiation into myofibroblasts. If so, it may be possible to utilize appropriate blocking strategies to ameliorate expression of IGFBP-rP2 and other IGFBP in BALCs and other cell types involved. The present data demonstrate enhanced IGFBP-rP2 expression in individuals with inflammatory lung injury from early-stage pulmonary sarcoidosis, and in patients with pulmonary fibrosis, either IPF or late-stage pulmonary sarcoidosis. Thus, it appears that expression is associated with inflammation as well as fibrosis, and therefore potentially reversible disease. IGFBP-rP2 expression may therefore have value as a clinical marker. These findings support the hypothesis that local upregulation of IGFBPrP2 expression in the lungs reflects proportional changes in the local immune-effector cell population associated with the onset of inflammation and progression to fibrosis. The differential expression of IGFBP-rP2 by different cell types, reflecting inter-individual heterogeneity, may also help to explain the wide variation in phenotype of fibrotic lung diseases. It is therefore likely that IGFBP-rP2 is an important factor influencing the outcome of the fibroproliferative response to lung injury at the level of both the epithelium and the fibroblast. Our findings also suggest that the IGFBP superfamily, acting both independently and with their IGF-1 ligand, forms a complex molecular network that modulates the response of the lung to pathogenic insults. Acknowledgments: The authors gratefully recognize the support of the School of Postgraduate Medicine, Keele University, in providing a research fellowship to one author (J.T.A.).
References 1. Bradham, D. M., A. Igarashi, R. L. Potter, and G. R. Grotendorst. 1991.
700
2.
3.
4.
5. 6.
7.
8.
9.
10.
11.
12.
13.
AMERICAN JOURNAL OF RESPIRATORY CELL AND MOLECULAR BIOLOGY VOL. 21 1999
Connective tissue growth factor: a cysteine-rich mitogen secreted by human vascular endothelial cells is related to the SRC-induced immediate early gene product. J. Cell Biol. 114:1285–1294. Frazier, K., S. Williams, D. Kothapalli, H. Klapper, and G. R. Grotendorst. 1996. Stimulation of fibroblast cell growth, matrix production, and granulation tissue formation by connective tissue growth factor. J. Invest. Dermatol. 107:404–411. Kim, H.-S., S. R. Nagalla, Y. Oh, E. Wilson, C. T. Roberts, Jr., and R. G. Rosenfeld. 1997. Identification of a family of low-affinity insulin-like growth factor binding proteins (IGFBPs): characterization of connective tissue growth factor as a member of the IGFBP superfamily. Proc. Natl. Acad. Sci. USA 94:12981–12986. Baxter, R. C., M. A. Binoux, D. R. Clemmons, C. A. Conover, S. L. S. Drop, J. M. P. Holly, S. Mohan, Y. Oh, and R. G. Rosenfeld. 1998. Recommendations for nomenclature of the insulin-like growth factor binding protein superfamily. Endocrinology 139:4036. Clemmons, D. R. 1998. Role of insulin-like growth factor binding proteins in controlling IGF actions. Mol. Cell Endocrinol. 140:19–24. Steffen, C. L., D. K. Ball-Mirth, P. A. Harding, N. Bhattacharyya, S. Pillai, and D. R. Brigstock. 1998. Characterization of cell-associated and soluble forms of connective tissue growth factor (CTGF) produced by fibroblasts in vitro. Growth Factors 15:199–213. Yang, D.-H., H.-S. Kim, E. M. Wilson, R. G. Rosenfeld, and Y. Oh. 1998. Identification of glycosylated 38-kDa connective tissue growth factor (IGFBP-related protein 2) and proteolytic fragments in human biological fluids, and up-regulation of IGFBP-rP2 expression by TGF-b in Hs578T human breast cancer cells. J. Clin. Endocrinol. Metab. 83:2593–2596. Kothapalli, D., K. S. Frazier, A. Welply, P. R. Segarini, and G. R. Grotendorst. 1996. Transforming growth factor b induces anchorage-independent growth of NRK fibroblasts via a connective tissue growth factor-dependent signaling pathway. Cell Growth Differ. 8:61–68. Dammeier, J., H. D. Beer, M. Brauchle, and S. Werner. 1998. Dexamethasone is a novel potent inducer of connective tissue growth factor expression: implications for glucocorticoid therapy. J. Biol. Chem. 273:18185– 18190. Lasky, J. A., L. A. Ortiz, B. Tonthat, G. W. Hoyle, M. Corti, G. Athas, G. Lungarella, A. Brody, and M. Friedman. 1998. Connective tissue growth factor mRNA expression is upregulated in bleomycin-induced lung fibrosis. Am. J. Physiol. 275:L365–L371. Dammeier, J., M. Brauchle, W. Falk, G. R. Grotendorst, and S. Werner. 1998. Connective tissue growth factor: a novel regulator of mucosal repair and fibrosis in inflammatory bowel disease? Int. J. Biochem. Cell Biol. 30: 909–922. Igarashi, A., K. Nashiro, K. Kikuchi, S. Sato, H. Ihn, G. R. Grotendorst, and K. Takehara. 1995. Significant correlation between connective tissue growth factor gene expression and skin sclerosis in tissue sections from patients with systemic sclerosis. J. Invest. Dermatol. 105:280–284. Ito, Y., J. Aten, R. J. Bende, B. S. Oemar, T. J. Rabelink, J. J. Weening, and
14.
15. 16.
17. 18.
19.
20.
21. 22. 23.
24.
25.
R. Goldschmeding. 1998. Expression of connective tissue growth factor in human renal fibrosis. Kidney Int. 53:853–861. Allen, J. T., C. A. Bloor, R. A. Knight, and M. A. Spiteri. 1998. Expression of insulin-like growth factor binding proteins in bronchoalveolar lavage fluid (BALF) of patients with pulmonary sarcoidosis. Am. J. Respir. Cell Mol. Biol. 19:250–258. Aston, C., J. Jagirdir, T. C. Lee, T. Hur, R. L. Hintz, and W. N. Rom. 1995. Enhanced insulin-like growth factor molecules in idiopathic pulmonary fibrosis. Am. J. Respir. Crit. Care Med. 151:1597–1603. Chadelat, K., M. Boule, S. Corroyer, B. Fauroux, B. Delaisi, G. Tournier, and A. Clement. 1998. Expression of insulin-like growth factors and their binding proteins by bronchoalveolar cells from children with and without interstitial lung disease. Eur. Respir. J. 11:1329–1336. Celi, F. S., M. E. Zenilman, and A. R. Schuldiner. 1993. A rapid and versatile method to synthesize internal standards for competitive PCR. Nucleic Acids Res. 21:1047. Willey, J. C., E. L. Crawford, C. M. Jackson, D. A. Weaver, J. C. Hoban, S. A. Khuder, and J. P. DeMuth. 1998. Expression measurement of many genes simultaneously by quantitative RT-PCR using standardized mixtures of competitive templates. Am. J. Respir. Cell Mol. Biol. 19:6–17. Willey, J. C., E. L. Coy, M. W. Frampton, A. Torres, M. J. Apostolakos, G. Hoehn, W. H. Schuermann, W. G. Thilly, D. E. Olsen, J. R. Hammersley, C. L. Crespi, and M. J. Utell. 1997. Quantitative RT-PCR measurement of cytochromes p450 1A1, 1B1, and 2B7, microsomal epoxide hydrolase, and NADPH oxidoreductase expression in lung cells of smokers and nonsmokers. Am. J. Respir. Cell Mol. Biol. 17:114–124. Nishida, T., T. Nakanishi, T. Shimo, M. Asano, T. Hattori, T. Tamatani, K. Tezuka, and M. Takigawa. 1998. Demonstration of receptors specific for connective tissue growth factor on a human chondrocytic cell line (HCS2/8). Biochem. Biophys. Res. Commun. 247:905–909. Nyman, T., and F. Pekonen. 1993. The expression of insulin-like growth factors and their binding proteins in normal human lymphocytes. Acta Endocrinologica 128:168–172. Hathaway, C. L., A. M. Arnold, R. P. Rand, L. H. Engrav, and L. S. Quinn. 1996. Differential expression of IGFBPs by normal and hypertrophic scar fibroblasts. J. Surg. Res. 60:156–162. Oh, Y., H. L. Muller, L. Ng, and R. G. Rosenfeld. 1995. Transforming growth factor b-induced cell growth inhibition in human breast cancer cells is mediated through insulin-like growth factor-binding protein-3 action. J. Biol. Chem. 270:13589–13592. Swisshelm, K., K. Ryan, K. Tsuchiya, and R. Sager. 1995. Enhanced expression of an insulin growth factor-like binding protein (mac25) in senescent human mammary epithelial cells and induced expression with retinoic acid. Proc. Natl. Acad. Sci. USA 92:4472–4476. Damon, S. E., K. L. Haugk, K. Swisshelm, and L. S. Quinn. 1997. Developmental regulation of mac25/insulin-like growth factor binding protein-7 expression in skeletal myogenesis. Exp. Cell Res. 237:192–195.
