help button home button
AJRCCM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published ahead of print on June 23, 2004, doi:10.1164/rccm.200312-1686OC
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200312-1686OCv1
170/8/911    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bull, T. M.
Right arrow Articles by Geraci, M. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bull, T. M.
Right arrow Articles by Geraci, M. W.
American Journal of Respiratory and Critical Care Medicine Vol 170. pp. 911-919, (2004)
© 2004 American Thoracic Society
doi: 10.1164/rccm.200312-1686OC

Gene Microarray Analysis of Peripheral Blood Cells in Pulmonary Arterial Hypertension

Todd M. Bull, Christopher D. Coldren, Mark Moore, Sylk M. Sotto-Santiago, David V. Pham, S. Patrick Nana-Sinkam, Norbert F. Voelkel and Mark W. Geraci

Pulmonary Hypertension Center, Division of Pulmonary Sciences and Critical Care Medicine, University of Colorado Health Sciences Center, Denver, Colorado

Correspondence and requests for reprints should be addressed to Todd M. Bull, M.D., Division of Pulmonary Sciences and Critical Care Medicine, 4200 East 9th Avenue, Denver, CO 80262. Email: todd.bull{at}uchsc.edu


   ABSTRACT
TOP
 ABSTRACT
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The importance of genetic predisposition, inflammation, and autoimmune mechanisms in the development of pulmonary arterial hypertension (PAH) is becoming increasingly clear. We hypothesized that the analysis of gene expression profiles from peripheral blood mononuclear cells would distinguish patients with PAH from normal volunteers. We also hypothesized that a subset of genes would discriminate between patients with idiopathic PAH and pulmonary hypertension related to secondary causes. Mononuclear cells were isolated from 15 patients diagnosed with PAH and 6 normal control subjects. Microarray expression was performed, and the expression profiles were analyzed for consistent and predictive differences in gene expression. We identified a signature set of 106 genes that discriminated with high certainty (p <= 0.002) between patients with PAH and normal individuals. The results of the microarray analysis were retrospectively and prospectively confirmed by quantitative polymerase chain reaction for 2 of the 106 genes. Supervised clustering analysis generated a list of differentially expressed genes between patients with idiopathic and secondary causes of pulmonary hypertension. Microarray expression profiling of peripheral blood cells can discriminate between patients with PAH and normal volunteers. These findings may have important implications toward diagnosis, screening, and pathogenesis of this disease.

Key Words: biomarker • gene microarray • inflammation • mononuclear cells • pulmonary arterial hypertension

It is hypothesized that the development of pulmonary arterial hypertension (PAH) requires first a genetic susceptibility followed by one or several secondary trigger factors such as a viral infection or drug exposure (1, 2). Although a number of promoting events are currently recognized, the individual's genetic susceptibility and the interaction of the genotype with the promoting factor or factors remain areas of active research.

Inflammation may play an important role in the development of some or all forms of severe pulmonary hypertension (3–5). PAH is a recognized complication of a number of systemic inflammatory conditions such as scleroderma and systemic lupus erythematosus (6). Mononuclear inflammatory cells surround the plexiform lesions in patients with scleroderma-related PAH and idiopathic PAH (IPAH) (7, 8) and plasma levels of inflammatory markers are elevated in patients with IPAH compared with normal control subjects (9, 10).

In concert with potential inflammatory mechanisms, there exists evidence that immunologic abnormalities may be associated with the development of PAH. Patients with human immunodeficiency virus-1 infection or with the POEMS syndrome (polyneuropathy, organomegaly, endocrinopathy, M protein, and skin changes) are known to develop severe pulmonary hypertension (1, 11). A significant number of patients diagnosed with IPAH have evidence of an autoimmune disorder and systemic inflammation. These abnormalities include the presence of antinuclear antibodies, an increased incidence of autoimmune thyroiditis, increased serum levels of proinflammatory cytokines such as interleukin-1 and interleukin-6, increased incidence of specific major histocompatibility complex class II molecules, and increased pulmonary expression of platelet-derived growth factor, macrophage inflammatory protein-{alpha}, and RANTES (regulated upon activation, normal T cell expressed and secreted) (3, 12–16).

The pathogenesis of severe PAH is complex, and it is likely that multiple modulating genes and environmental factors are involved. Such complexity lends itself to the use of microarray technology, which allows the efficient and accurate simultaneous expression measurement of thousands of genes (17). This technology has been most successfully employed in the investigation of cancer, including hematologic malignancies and in the classification of histologically indistinct tumor types with different natural histories (18–22). A significant challenge to the study of gene expression is the collection of biological material of sufficient homogeneity, quantity, and quality for microarray study. Biopsy specimens from patients with early-stage disease tend to be small, and routine histologic preservation (formalin fixation) generally prohibits quality ribonucleic acid extraction. Furthermore, in PAH lung biopsy is relatively contraindicated because of the high associated morbidity and mortality of the procedure.

As discussed previously here, PAH is associated with an altered cytokine, chemokine, and growth factor milieu of circulating cells (12, 14, 23, 24). These changes, combined with the genetic susceptibility to the development of PAH and the possibility of autoimmune and inflammatory etiologies of the disease, led us to hypothesize that peripheral blood mononuclear cells (PBMCs) from patients with PAH would have an altered gene expression pattern compared with normal individuals. Here we used PBMCs as a surrogate marker to distinguish patients with PAH from normal individuals by gene expression profiling. We demonstrate, for the first time, that global expression profiling of peripheral blood cells can distinguish patients with a chronic pulmonary disease.


   METHODS
TOP
 ABSTRACT
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Subjects and Blood Collection
Blood donors were volunteers, as approved by the institutional human–subjects review board (Colorado Multiple Institution Review Board [COMIRB] protocol number 00–605). All subjects gave informed consent. The "microarray cohort" of subjects was composed of 15 patients diagnosed with PAH (seven patients diagnosed with IPAH and eight diagnosed with PAH related to a secondary cause [s-PAH]) and 6 normal volunteers (Table 1). In addition to the microarray cohort, we prospectively studied a second group of patients. This group was composed of 14 patients with PAH (4 patients with IPAH and 10 patients with s-PAH) and 6 normal volunteers and is referred to as the "prospective cohort" (Table 2). The diagnosis of IPAH was established using the algorithm developed by the Primary Pulmonary Hypertension (PPH) National Institutes of Health registry (25). Patients were excluded from the study if there was evidence of other active disease processes unrelated to PAH within the preceding 30 days (i.e., systemic infection, bleeding diathesis) Blood recovered from a peripheral venipuncture was collected between 1:00 and 4:00 P.M. into vacutainer tubes containing ethylenediaminetetraacetic acid. All blood was drawn from a peripheral vein through a 21-gauge needle. All specimens were processed within 2 hours of collection. The enrolled patients had previously undergone right heart catheterization to confirm the diagnosis of severe pulmonary hypertension. The peripheral blood draws were performed at least 14 days after cardiac catheterization, and in the majority of cases, the blood draw occurred more then 60 days after the catheterization. Echocardiography was performed within 90 days of the blood draw to confirm the persistent elevation in pulmonary arterial pressure. The blood analysis from these patients was compared with that from six normal volunteers. The patients and control subjects were age matched. The patients with IPAH compared with s-PAH were matched by disease severity (mean pulmonary artery pressure, cardiac output, and pulmonary vascular resistance).


View this table:
[in this window]
[in a new window]
  		TABLE 1. Characteristics of patients included in the microarray cohort

 

View this table:
[in this window]
[in a new window]
  		TABLE 2. Characteristics of patients included in the prospective cohort for quantitative-polymerase chain reaction analysis

 
The results from the gene microarrays were confirmed by quantitative polymerase chain reaction (q-PCR) retrospectively on a subset of patients from the microarray cohort (five patients with PAH and three normal control subjects) and prospectively on a second group of PAH patients and normal control subjects (14 patients with PAH and 6 normal control subjects) (Table 2).

Isolation of PBMCs
Four milliliters of peripheral blood were collected in tubes containing ethylenediaminetetraacetic acid. The blood was diluted in three volumes of phosphate-buffered saline + 2-mM ethylenediaminetetraacetic acid + 0.5% bovine serum albumin. The mononuclear cell layer (which includes monocytes/macrophages, B and T lymphocytes, and natural killer cells) was isolated via density gradient centrifugation (Hisotpaque 1077, 1,200 rpm for 30 minutes). The purity of the mononuclear cell layer was assessed by a Coulter counter on a representative population of the patients (eight patients with PAH and six normal control subjects) and determined to comprise greater than 90% mononuclear cells.

Microarray Data Generation
Sample preparation, RNA isolation, and high-density oligonucleotide array hybridization and scanning were performed as described previously (26). Fluorescence intensities were quantified using the Affymetrix Microarray Suite 5.0 statistical algorithm with default parameters for the array type used in this study (Affymetrix, Santa Clara, CA). Tabular gene expression data are published online in Gene Expression Omnibus (www.ncbi.nlm.nih.gov/geo/), submission GSE703.

q-PCR
RNA was extracted from PBMCs using the RNAeasy kit (Qiagen, Valencia, CA). Primers and probes were obtained from Applied Biosystems Assays on Demand (Foster City, CA). All reactions were performed on a Gene Amp 5,700 sequence detector (Applied Biosystems) using the conditions recommended by the manufacturer. Standard curves were created using cloned PCR products in concentrations ranging from 0.1 to 0.0001 ng/µl. We confirmed the absence of nonspecific amplification by examining PCR products by agarose gel electrophoresis. Standard and experimental samples were run in triplicate, and the results were averaged. All results were standardized to expression of ß-actin. Genes assayed by q-PCR included endothelial cell growth factor-1 (ECGF-1), adrenomedullin (ADM), and tumor necrosis factor receptor super family 14 also referred to as herpesvirus entry mediator (HVEM). The primers and probes for q-PCR were purchased from Assay on Demand (Applied Biosystems).

Data Analysis
Patient demographics.
Statistical analysis of the demographic data was performed with the Prism version 3.0 for windows (GraphPad Software, San Diego, CA). One-way analysis of variance was used to compare patient ages. Unpaired t tests were used to compare mean pulmonary artery (PA) pressures, cardiac output, and pulmonary vascular resistance (PVR) between patients with PAH. Unpaired t tests with a Mann-Whitney correction were used to analyze q-PCR results.

Microarray analysis.
Samples were analyzed by the use of Affymetrix HuFL GeneChips. Each gene on the chip is represented by 10 to 20 oligonucleotides referred to as a "probe set." The level of expression of a particular gene is directly associated with the intensity of hybridization of labeled messenger RNA to these probe sets. The HuFL GeneChip is designed to assess the expression level of approximately 6,000 different genes.

The array data were analyzed with BRB ArrayTools v3.0.2E developed by Dr. Richard Simon and Amy Tan. The initial data set consisted of 6,086 gene measurements for each of the 21 samples. Genes whose expression was not reliably detected (i.e., had an "absolute call" of "present" or "marginal," details in the Affymetrix Signal Algorithm Description Document white paper) in at least 11 of the 21 samples were excluded, resulting in the inclusion of 2,906 gene expression measurements in 21 samples. These measurements were mean centered and analyzed using the clustering, class comparison, and class prediction functions of BRB ArrayTools. The Gene Ontology (GO) analysis was conducted using GenMAPP and MAPPFinder (27). The z score assigned to each category by MAPPFinder reflects the degree to which the differential expression of genes in that category was greater than that expected by chance. A high, positive z score indicates that a large number of genes in that category are differentially expressed between the compared conditions.


   RESULTS
TOP
 ABSTRACT
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Patient Characteristics
Table 1 displays the characteristics of the patients whose PBMCs underwent microarray analysis. The patients with PAH and the normal volunteers were similar in age (50.2 ± 3.5 years, n = 15, vs. 39.0 ± 1.9 years, n = 6, mean ± SEM, p >= 0.05). The patients with IPAH and s-PAH were not significantly different in terms of mean PA pressure (60.3 ± 6.0 mm Hg, n = 7, vs. 49.8 ± 1.7 mm Hg, n = 8, mean ± SEM, p >= 0.09), cardiac output (3.77 ± 0.78 L/min, n = 7, vs. 4.46 ± 0.5 L/min, n = 7, mean ± SEM, p >= 0.4) or PVR (18.2 ± 4.0 wood units, n = 5, vs. 10.1 ± 2.5 wood units, n = 5, mean ± SEM, p >= 0.13).

Table 2 displays the characteristics of patients who underwent quantitative PCR for prospective confirmation of the microarray data (prospective cohort). The patients with PAH and the normal volunteers were matched by age (53.6 ± 3.6 years, n = 14, vs. 47.0 ± 5.3 years, n = 6, mean ± SEM, p >= 0.3). The patients with IPAH and s-PAH were not significantly different in terms of mean PA pressure (53.0 ± 7.3 mm Hg, n = 4, versus 60.3 ± 6.1 mm Hg, n = 8, mean ± SEM, p >= 0.4), cardiac output (3.5 ± 0.41 L/min, n = 4, vs. 2.7 ± 0.54 L/min, n = 8, mean ± SEM, p >= 0.3), or PVR (10.5 ± 3.6 wood units, n = 4, vs. 15.0 ± 1.7 wood units, n = 8, mean ± SEM, p >= 0.23).

Microarray Analysis
Microarray data were examined first in an unsupervised mode, using expression values for all 2,906 probe sets present in the majority of samples. Clustering of these data is shown in Figure 1A, which demonstrates that the non-PAH samples are more closely related to one another than to the PAH samples. The robustness of this grouping is high (0.931) when perturbed data are reclustered. A 106-gene expression signature (Table E1 in the online supplement), which supported this segregation of normal and PAH samples, was generated using a two-sample t test and a conservative p value cutoff (0.001) for expression differences supporting the class assignment. Perturbation testing of the resulting 106-gene signature suggests that this list contains fewer than two false discoveries. Supervised clustering of the samples using this signature results in a more robust (0.971) grouping of the normal volunteers (Figure 1B). Most (96 of 106) of the gene expression values in this signature have a higher mean value in the PAH samples relative to the non-PAH samples (Figure 2).



View larger version (44K):
[in this window]
[in a new window]
  		Figure 1. Dendrograms of pulmonary arterial hypertension (PAH) and normal samples, clustered using centered correlation and average linkage. (A) Unsupervised clustering based on 2,906 genes. (B) Supervised clustering based on 106 genes.

 


View larger version (64K):
[in this window]
[in a new window]
  		Figure 2. Color display of the 106-gene signature discriminating between PAH and normal samples. The color display was generated using the GeneSpring program (Silicon Genetics, Redwood, CA). The green color represents relatively high expression, and the red color indicates relatively low expression.

 
The utility of PBMC gene expression data for sample discrimination was evaluated in two separate prediction protocols. In the first protocol, a gene expression profile was developed from the first 14 patient and volunteer samples evaluated. This profile was used to predict the class membership of seven subsequent samples (05 Nor, 06 Nor, 04 PAH, 06 PAH, 07 PAH, 08 PAH, 13 PAH). In each case, the correct prediction was made with all of the available prediction algorithms (see Table E2 in the online supplement). Because of the limited size of the patient group, a second "leave-one-out" protocol was employed for prediction accuracy cross-validation; this has the advantage of using the data more efficiently. In our leave-one-out cross-validation, each patient or volunteer sample was excluded from the data set sequentially, and the remaining 20 samples were used to build a gene expression profile discriminating between the two classes, and the resulting profile was used to predict the class of the left-out specimen. In each of the 21 iterations, an independent expression profile was computed, and the prediction of the left-out specimen was completed with 95% or 100% accuracy using the available algorithms (see Table E3 in the online supplement). The p values for each of the predictors are estimated to be 0.002 or less based on 2,000 random permutations.

In addition to the comparisons of normal and PAH PBMC gene expression, we conducted a comparison of PBMC gene expression within the PAH patient population of seven IPAH and eight s-PAH samples. A comparison of these two groups with the class comparison protocol did not reveal a statistically significant pattern of gene expression discriminating between these two groups. Some individual genes did attain a statistically significant difference in this class comparison: 28 genes are significant at the {alpha} = 0.01 level, and 178 are significant at the {alpha} = 0.05 level. However, both sets of statistically significant gene lists failed to sustain significance on permutation testing, resulting in an estimated 20% probability that either of these sets would have little predictive value in a larger cohort.

q-PCR Results
PAH versus normal.
Two genes identified through microarray analysis to distinguish patients with PAH versus normal individuals were selected for further investigation using q-PCR. The genes were selected for both their ability to discriminate between groups by microarray and their perceived biologic interest. These genes were ECGF-1 (p = 0.0008) and ADM (p = 0.0008). q-PCR was performed on the PBMC samples of a subset of the patients who had undergone microarray analysis (patients PAH 1, PAH 2, PAH 5, PAH 6, PAH 10, PAH 11, PAH 15, normal 2, normal 3, normal 6). As predicted by the microarray data, there was a significant difference in the expression of these genes between patients with PAH compared with normal control subjects (ECGF-1, 14,930 ± 4,912, n = 5, vs. 2,175 ± 963, n = 3, mean ± SEM, p <= 0.04) (ADM, 35.4 ± 25, n = 5, vs. 1.1 ± 0.5 n = 3, mean ± SEM, p <= 0.04) (Figures 3A and 4A). q-PCR for these two genes was then performed on a second prospective cohort of normal individuals and patients with PAH. Again, a significant difference of expression of ECGF-1 and ADM was detected in PAH versus normal control subjects in the direction predicted by the microarray (ECGF-1, 14,930 ± 4,912, n = 14, vs. 2,175 ± 963, n = 6, mean ± SEM, p <= 0.05) (ADM, 61.5 ± 15.8, n = 14, vs. 17.5 ± 10.4, n = 6 mean ± SEM, p <= 0.03) (Figures 3B and 4B).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 3. Quantitative PCR (q-PCR) measurements of gene expression of endothelial cell growth factor-1 (ECGF-1) from peripheral blood mononuclear cell (PBMC) samples of patients with PAH and normal volunteers. (A) Gene expression of ECGF-1 in patients with PAH compared with normal volunteers from the microarray cohort. (B) ECGF-1 expression in patients with PAH compared with normal volunteers from the prospective cohort.

 


View larger version (21K):
[in this window]
[in a new window]
  		Figure 4. q-PCR measurements of gene expression of adrenomedullin (ADM) from PBMC samples of patients with PAH. (A) Gene expression of ADM in patients with PAH compared with normal volunteers from the microarray cohort. (B) ADM expression in patients with PAH compared with normal volunteers from the prospective cohort.

 
IPAH versus s-PAH.
Reanalysis of the microarray data using a supervised class comparison algorithm identified a list of 28 genes, which were differentially expressed in patients with IPAH versus s-PAH. One of these genes, tumor necrosis factor receptor superfamily, member 14, also called HVEM (p = 0.007), was selected for analysis on our prospective cohort of patients by q-PCR.

q-PCR confirmed a significant difference in the expression of HVEM in the PBMCs of patients with IPAH compared with patients diagnosed with s-PAH (5,157 ± 1,248, n = 9, vs. 10,410 ± 1,412, n = 7, mean ± SEM, p <= 0.05) using this method (Figure 5).



View larger version (23K):
[in this window]
[in a new window]
  		Figure 5. q-PCR measurements of gene expression of herpesvirus entry mediator (HVEM) from PBMC samples of patients with idiopathic PAH (IPAH) and PAH related to a secondary cause (s-PAH) measured in the prospective cohort.

 
GO Analysis
To discover classes of genes that were involved in PAH, we used GenMAPP and MAPPfinder, and we employed less stringent statistical criteria for the identification of groups of genes than had been used for the identification of individual genes. This software calculates a standardized difference score (z score) for each gene category, comparing the number of observed changes in a category to the number expected in that category by chance. Table 3 lists discriminating genes by their described GO category. To be included in this list the genes had to have a z score of more than 3.0 and a minimum of six genes had to be differentially expressed in the involved GO category.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Gene ontology (go) categories with an abundance of differentially expressed genes in individuals with pulmonary arterial hypertension vs. normal individuals

 
A significant number of genes in the GO categories of inflammatory response, stress response, cytochrome c oxidase, lysosome, and intracellular signaling cascade were identified as being differentially expressed.


   DISCUSSION
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Methods for large-scale molecular profiling of diseased tissues are well established with proven efficacy both diagnostically and prognostically (28–31). Recently, microarray analysis of surrogate tissues has been used to gain disease specific information. A number of investigators have employed this technology to document differences in gene expression of peripheral blood cells in a variety of disease states (21, 22, 32–36). Here we attempted to define the gene expression of PBMCs to explore alternative means of defining and diagnosing severe PAH and to gain further insight into the pathobiology of the disease process.

PAH constitutes a wide spectrum of diseases that result in similar histopathologic and clinical phenotypes. Although it is increasingly accepted that patients who develop PAH have a genetic predisposition, the exact nature of these genetic anomalies and how the genotype interacts with various environmental factors remain unclear (37–40). Circulating blood cells may carry disease-specific information either because of inherent genomic alterations or because of alterations in their local environment. Furthermore, the gene expression of PBMCs may provide disease-relevant information because of inflammatory and autoimmune mechanisms that likely play important roles in the development of PAH (3, 4, 10, 41–44). We have identified a gene expression pattern that accurately distinguishes patients with PAH from normal volunteers. We have also identified a number of novel genes that may be associated with the pathobiology of pulmonary hypertension.

The genes and GO categories identified as being differentially expressed in PAH versus normal control subjects may be of significant biologic interest. We selected 2 genes from the 106-gene list that distinguishes PAH from normal individuals and 1 gene from the 28-gene list that was differentially expressed between patients with IPAH and s-PAH for confirmation by q-PCR. These genes were selected both because of their ability to discriminate between groups and because of their perceived biologic interest. ADM (increased in PAH vs. normal) (Figure 4) is a potent pulmonary vasodilator (45). Plasma levels of ADM are elevated in idiopathic and secondary forms of pulmonary hypertension, and its use as a therapeutic inhalational agent in the treatment of PAH is currently under investigation (46–49). ECGF-1 (increased in PAH versus normal) (Figure 3) is an angiogenic factor, and its expression is increased in a number of malignancies (50–52). As the plexiform lesion of PAH is composed of abnormally proliferating endothelial cells, an association with increased expression of ECGF-1 is of significant interest. The differential expression of HVEM between patients with IPAH and s-PAH (increased in s-PAH vs. IPAH) (Table 4 and Figure 5) is potentially relevant in the context of our recent reports of an association between IPAH and human herpesvirus-8 (HHV-8) infection (43, 44). The change of HVEM gene expression in PBMCs could be consistent with a herpesvirus infection, an activated immune system, or both. Further investigation is required to define the relevance of these finding to the disease process of PAH.


View this table:
[in this window]
[in a new window]
  		TABLE 4. List of 28 genes found to be differentially expressed between patients with ipah vs. s-pah by supervised class comparison

 
Table 3 lists GO categories and specific genes within these categories that are differentially expressed in the PBMC samples of patients with PAH compared with normal volunteers. The "inflammatory response" and "response to stress" GO categories identified are intriguing and fit well with concept of an inflammatory phenotype previously described in PAH. Genes encoding chemokines and their receptors (CXCL1, CCR1, CCR7) and Toll-like receptors may be overexpressed as a result of a yet to be defined immune response. Increased expression of the genes for arachidonate-5 lipoxygenase and the arylhydrocarbon receptor are of interest as both proteins have been shown to be expressed in the plexiform lesions of severe PAH (53). Ultimately, determining whether these genes are in fact involved in the pathogenesis of PAH will require further investigation.

Beyond the potential biologic and diagnostic importance of this work, we also emphasize several important features regarding its execution. This study was designed to emphasize the importance of measure validation. Because of the high measurement-to-sample ratio inherent in microarray studies, steps must be taken to avoid "overfitting" any predictive model (54). For this reason, we employed both independent and "leave-one-out" cross-validation of our predictor. This was accomplished using a training set of two-thirds of the total number of arrays. The remaining one-third of the samples was used as a test set to determine, in a blinded fashion, the predicted value of the class prediction analysis. The blinded, prospectively analyzed, one-third contingent was accurately classified in all cases. This independent blinded data set represents the most stringent manner of analyzing the quality of a prediction rule. We used both cross-validation and permutation testing to assess the quality of the prediction rule. A further validation was performed using an independent measure of gene expression magnitude.

The differences in gene expression were confirmed using an independent measure of messenger RNA abundance. The power of the microarray analysis to determine biomarkers, which discriminate between normal and pulmonary hypertension, is made evident by the fact that the prospective cohort also demonstrated the differences in gene expression profiles predicted by the microarray subset of diseased and normal patients.

We attempted in this study to define a signature set of genes that could reproducibly distinguish patients with IPAH from those with s-PAH. However, we were unable to identify a gene set that reliably defined class prediction when scrutinized by random permutation validation. However, a number of differentially expressed genes exist between patients with IPAH and s-PAH, and we prospectively validated one of these genes (HVEM) by q-PCR. There are several potential reasons why we were unable to define a gene profile that discriminated between patients with IPAH and s-PAH. These include our relatively small sample size of IPAH patients and the heterogeneous nature of the s-PAH cohort we used, the use of microarrays with a limited number of available gene probes, or the possibility that these disease phenotypes are not in fact distinguishable by microarray evaluation of PBMCs.

Several other potential limitations to this study merit discussion. First, it is not possible to determine from our findings whether the changes in PBMC gene expression are cause or consequence of the pressure elevation in the pulmonary vasculature. The observation that unsupervised clustering does not separate patients with IPAH from those with s-PAH might suggest that the pressure elevation in the pulmonary vasculature was the more important determinant of peripheral blood cell gene expression. An analysis of the gene expression in IPAH and s-PAH patients with less severe pulmonary hypertension may clarify this point. Second, it is also likely that some of the changes in gene expression observed are related to the therapy received rather then the underlying disease process. It is relevant to note, however, that treatment differences existed between our initial microarray cohort and the prospective cohort of patients, yet the differences in gene expression investigated by q-PCR remained consistent. Future studies need to examine the gene expression of patients with mild disease and attempt to obtain a cohort of previously untreated patients with severe disease. Finally, although there was not a statistically significant difference in the ages of patients with PAH as compared with normal volunteers in the microarray cohort, the mean age of the PAH patients was higher then that of the normal individuals (50.2 vs. 39.7 years), which theoretically could have affected our results.

Our work represents a novel approach to the identification and classification of PAH. A lack of access to the site of pathology—the lung vasculature—early in the patient's course has severely limited the study of PAH disease progression at both histologic and molecular levels. The ability to distinguish patients with PAH by examining peripheral blood has significant implications in terms of diagnosis and patient screening. Elucidation of the pathologic mechanisms involved in these changes in gene expression will require further investigation.


   FOOTNOTES
 
Supported by a grant from the Pulmonary Hypertension Association administered through the American Heart Association (T.M.B.), R01 HL 72,340 (M.W.G. and C.D.C.), and P01 HL66254 (N.F.V. and M.W.G.).

Microarray data has been deposited in Gene Expression Omnibus, accession GSE703.

This article has an online supplement, which is accessible from this issue's table of contents online at www.atsjournals.org

Conflict of Interest Statement: T.M.B. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; C.D.C. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.M. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.M.S.-S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; D.V.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; S.P.N.-S. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; N.F.V. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript; M.W.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript.

Received in original form December 10, 2003; accepted in final form June 19, 2004


   REFERENCES
TOP
 ABSTRACT
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Humbert M, Nunes H, Sitbon O, Parent F, Herve P, Simonneau G. Risk factors for pulmonary arterial hypertension. Clin Chest Med 2001;22:459–475.[CrossRef][Medline]
  2. Voelkel NF. Appetite suppressants and pulmonary hypertension. Thorax 1997;52:S63–S67.[Free Full Text]
  3. Dorfmuller P, Perros F, Balabanian K, Humbert M. Inflammation in pulmonary arterial hypertension. Eur Respir J 2003;22:358–363.[Abstract/Free Full Text]
  4. Tuder RM, Voelkel NF. Pulmonary hypertension and inflammation. J Lab Clin Med 1998;132:16–24.[CrossRef][Medline]
  5. Voelkel NF, Tuder R. Interleukin-1 receptor antagonist inhibits pulmonary hypertension induced by inflammation. Ann N Y Acad Sci 1994;725:104–109.[Medline]
  6. Fagan KA, Badesch DB. Pulmonary hypertension associated with connective tissue disease. Prog Cardiovasc Dis 2002;45:225–234.[CrossRef][Medline]
  7. Cool CD, Kennedy D, Voelkel NF, Tuder RM. Pathogenesis and evolution of plexiform lesions in pulmonary hypertension associated with scleroderma and human immunodeficiency virus infection. Hum Pathol 1997;28:434–442.[CrossRef][Medline]
  8. Tuder RM, Groves B, Badesch DB, Voelkel NF. Exuberant endothelial cell growth and elements of inflammation are present in plexiform lesions of pulmonary hypertension. Am J Pathol 1994;144:275–285.[Abstract]
  9. Balabanian K, Foussat A, Dorfmuller P, Durand-Gasselin I, Capel F, Bouchet-Delbos L, Portier A, Marfaing-Koka A, Krzysiek R, Rimaniol AC, et al. CX(3)C chemokine fractalkine in pulmonary arterial hypertension. Am J Respir Crit Care Med 2002;165:1419–1425.[Abstract/Free Full Text]
  10. Dorfmuller P, Zarka V, Durand-Gasselin I, Monti G, Balabanian K, Garcia G, Capron F, Coulomb-Lhermine A, Marfaing-Koka A, Simonneau G, et al. Chemokine RANTES in severe pulmonary arterial hypertension. Am J Respir Crit Care Med 2002;165:534–539.[Abstract/Free Full Text]
  11. Lesprit P, Godeau B, Authier FJ, Soubrier M, Zuber M, Larroche C, Viard JP, Wechsler B, Gherardi R. Pulmonary hypertension in POEMS syndrome: a new feature mediated by cytokines. Am J Respir Crit Care Med 1998;157:907–911.[Abstract/Free Full Text]
  12. Humbert M, Monti G, Brenot F, Sitbon O, Portier A, Grangeot-Keros L, Duroux P, Galanaud P, Simonneau G, Emilie D. Increased interleukin-1 and interleukin-6 serum concentrations in severe primary pulmonary hypertension. Am J Respir Crit Care Med 1995;151:1628–1631.[Abstract]
  13. Isern RA, Yaneva M, Weiner E, Parke A, Rothfield N, Dantzker D, Rich S, Arnett FC. Autoantibodies in patients with primary pulmonary hypertension: association with anti-Ku. Am J Med 1992;93:307–312.[CrossRef][Medline]
  14. Humbert M, Monti G, Fartoukh M, Magnan A, Brenot F, Rain B, Capron F, Galanaud P, Duroux P, Simonneau G, et al. Platelet-derived growth factor expression in primary pulmonary hypertension: comparison of HIV seropositive and HIV seronegative patients. Eur Respir J 1998;11:554–559.[Abstract]
  15. Chu JW, Kao PN, Faul JL, Doyle RL. High prevalence of autoimmune thyroid disease in pulmonary arterial hypertension. Chest 2002;122:1668–1673.[Abstract/Free Full Text]
  16. Morse JH, Barst RJ, Fotino M, Zhang Y, Flaster E, Gharavi AE, Fritzler MJ, Dominguez M, Angles-Cano E. Primary pulmonary hypertension, tissue plasminogen activator antibodies, and HLA-DQ7. Am J Respir Crit Care Med 1997;155:274–278.[Abstract]
  17. Schena M, Heller RA, Theriault TP, Konrad K, Lachenmeier E, Davis RW. Microarrays: biotechnology's discovery platform for functional genomics. Trends Biotechnol 1998;16:301–306.[CrossRef][Medline]
  18. Khan J, Wei JS, Ringner M, Saal LH, Ladanyi M, Westermann F, Berthold F, Schwab M, Antonescu CR, Peterson C, et al. Classification and diagnostic prediction of cancers using gene expression profiling and artificial neural networks. Nat Med 2001;7:673–679.[CrossRef][Medline]
  19. Bittner M, Meltzer P, Chen Y, Jiang Y, Seftor E, Hendrix M, Radmacher M, Simon R, Yakhini Z, Ben Dor A, et al. Molecular classification of cutaneous malignant melanoma by gene expression profiling. Nature 2000;406:536–540.[CrossRef][Medline]
  20. Bhattacharjee A, Richards WG, Staunton J, Li C, Monti S, Vasa P, Ladd C, Beheshti J, Bueno R, Gillette M, et al. Classification of human lung carcinomas by mRNA expression profiling reveals distinct adenocarcinoma subclasses. Proc Natl Acad Sci USA 2001;98:13790–13795.[Abstract/Free Full Text]
  21. Bullinger L, Dohner K, Bair E, Frohling S, Schlenk RF, Tibshirani R, Dohner H, Pollack JR. Use of gene-expression profiling to identify prognostic subclasses in adult acute myeloid leukemia. N Engl J Med 2004;350:1605–1616.[Abstract/Free Full Text]
  22. Valk PJ, Verhaak RG, Beijen MA, Erpelinck CA, Barjesteh van Waalwijk van Doorn-Khosrovani S, Boer JM, Beverloo HB, Moorhouse MJ, van der Spek PJ, Lowenberg B, et al. Prognostically useful gene-expression profiles in acute myeloid leukemia. N Engl J Med 2004;350:1617–1628.[Abstract/Free Full Text]
  23. Voelkel NF, Hoeper M, Maloney J, Tuder RM. Vascular endothelial growth factor in pulmonary hypertension. Ann N Y Acad Sci 1996;796:186–193.[Medline]
  24. Eddahibi S, Humbert M, Sediame S, Chouaid C, Partovian C, Maitre B, Teiger E, Rideau D, Simonneau G, Sitbon O, et al. Imbalance between platelet vascular endothelial growth factor and platelet-derived growth factor in pulmonary hypertension: effect of prostacyclin therapy. Am J Respir Crit Care Med 2000;162:1493–1499.[Abstract/Free Full Text]
  25. D'Alonzo GE, Barst RJ, Ayres SM, Bergofsky EH, Brundage BH, Detre KM, Fishman AP, Goldring RM, Groves BM, Kernis JT, et al. Survival in patients with primary pulmonary hypertension: results from a national prospective registry. Ann Intern Med 1991;115:343–349.
  26. Geraci MW, Moore M, Gesell T, Yeager ME, Alger L, Golpon H, Gao B, Loyd JE, Tuder RM, Voelkel NF. Gene expression patterns in the lungs of patients with primary pulmonary hypertension: a gene microarray analysis. Circ Res 2001;88:555–562.[Abstract/Free Full Text]
  27. Doniger SW, Salomonis N, Dahlquist KD, Vranizan K, Lawlor SC, Conklin BR. MAPPFinder: using Gene Ontology and GenMAPP to create a global gene-expression profile from microarray data. Genome Biol 2003;4:R7.[CrossRef][Medline]
  28. Golub TR, Slonim DK, Tamayo P, Huard C, Gaasenbeek M, Mesirov JP, Coller H, Loh ML, Downing JR, Caligiuri MA, et al. Molecular classification of cancer: class discovery and class prediction by gene expression monitoring. Science 1999;286:531–537.[Abstract/Free Full Text]
  29. Brown PO, Botstein D. Exploring the new world of the genome with DNA microarrays. Nat Genet 1999;21:33–37.[CrossRef][Medline]
  30. Alizadeh AA, Eisen MB, Davis RE, Ma C, Lossos IS, Rosenwald A, Boldrick JC, Sabet H, Tran T, Yu X, et al. Distinct types of diffuse large B-cell lymphoma identified by gene expression profiling. Nature 2000;403:503–511.[CrossRef][Medline]
  31. van ‘t Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao M, Peterse HL, van der Kooy K, Marton MJ, Witteveen AT, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002;415:530–536.[CrossRef][Medline]
  32. Gu J, Marker-Hermann E, Baeten D, Tsai WC, Gladman D, Xiong M, Deister H, Kuipers JG, Huang F, Song YW, et al. A 588-gene microarray analysis of the peripheral blood mononuclear cells of spondyloarthropathy patients. Rheumatology 2002;41:759–766.[Abstract/Free Full Text]
  33. Huang F, Gu J, Zhao W, Zhu J, Zhang J, Yu DT. One-year open-label trial of thalidomide in ankylosing spondylitis. Arthritis Rheum 2002;47:249–254.[CrossRef][Medline]
  34. Rus V, Atamas SP, Shustova V, Luzina IG, Selaru F, Magder LS, Via CS. Expression of cytokine- and chemokine-related genes in peripheral blood mononuclear cells from lupus patients by cDNA array. Clin Immunol 2002;102:283–290.[CrossRef][Medline]
  35. Ramanathan M, Weinstock-Guttman B, Nguyen LT, Badgett D, Miller C, Patrick K, Brownscheidle C, Jacobs L. In vivo gene expression revealed by cDNA arrays: the pattern in relapsing-remitting multiple sclerosis patients compared with normal subjects. J Neuroimmunol 2001;116:213–219.[CrossRef][Medline]
  36. Jison ML, Munson PJ, Barb JJ, Suffredini AF, Talwar S, Logun C, Raghavachari N, Beigel JH, Shelhamer JH, Danner RL, et al. Blood mononuclear cell gene expression profiles characterize the oxidant, hemolytic and inflammatory stress of sickle cell disease. Blood 2004;104:270–280.[Abstract/Free Full Text]
  37. Thompson P, McRae C. Familial pulmonary hypertension: evidence of autosomal dominant inheritance. Br Heart J 1970;32:758–760.[Abstract/Free Full Text]
  38. Loyd JE, Butler MG, Foroud TM, Conneally PM, Phillips JA III, Newman JH. Genetic anticipation and abnormal gender ratio at birth in familial primary pulmonary hypertension. Am J Respir Crit Care Med 1995;152:93–97.[Abstract]
  39. Deng Z, Morse JH, Slager SL, Cuervo N, Moore KJ, Venetos G, Kalachikov S, Cayanis E, Fischer SG, Barst RJ, et al. Familial primary pulmonary hypertension (gene PPH1) is caused by mutations in the bone morphogenetic protein receptor-II gene. Am J Hum Genet 2000;67:737–744.[CrossRef][Medline]
  40. De Caestecker M, Meyrick B. Bone morphogenetic proteins, genetics and the pathophysiology of primary pulmonary hypertension. Respir Res 2001;2:193–197.[CrossRef][Medline]
  41. Balabanian K, Foussat A, Dorfmuller P, Durand-Gasselin I, Capel F, Bouchet-Delbos L, Portier A, Marfaing-Koka A, Krzysiek R, Rimaniol AC, et al. CX(3)C chemokine fractalkine in pulmonary arterial hypertension. Am J Respir Crit Care Med 2002;165:1419–1425.
  42. Tuder RM, Groves B, Badesch DB, Voelkel NF. Exuberant endothelial cell growth and elements of inflammation are present in plexiform lesions of pulmonary hypertension. Am J Pathol 1994;144:275–285.
  43. Bull TM, Cool CD, Serls AE, Rai PR, Parr J, Neid JM, Geraci MW, Campbell TB, Voelkel NF, Badesch DB. Primary pulmonary hypertension, Castleman's disease and human herpesvirus-8. Eur Respir J 2003;22:403–407.[Abstract/Free Full Text]
  44. Cool CD, Rai PR, Yeager ME, Hernandez-Saavedra D, Serls AE, Bull TM, Geraci MW, Brown KK, Routes JM, Tuder RM, et al. Expression of human herpesvirus 8 in primary pulmonary hypertension. N Engl J Med 2003;349:1113–1122.[Abstract/Free Full Text]
  45. Lippton H, Chang JK, Hao Q, Summer W, Hyman AL. Adrenomedullin dilates the pulmonary vascular bed in vivo. J Appl Physiol 1994;76:2154–2156.[Abstract/Free Full Text]
  46. Kakishita M, Nishikimi T, Okano Y, Satoh T, Kyotani S, Nagaya N, Fukushima K, Nakanishi N, Takishita S, Miyata A, et al. Increased plasma levels of adrenomedullin in patients with pulmonary hypertension. Clin Sci (Lond) 1999;96:33–39.[Medline]
  47. Yoshibayashi M, Kamiya T, Kitamura K, Saito Y, Kangawa K, Nishikimi T, Matsuoka H, Eto T, Matsuo H. Plasma levels of adrenomedullin in primary and secondary pulmonary hypertension in patients < 20 years of age. Am J Cardiol 1997;79:1556–1558.[CrossRef][Medline]
  48. Kakishita M, Nishikimi T, Okano Y, Satoh T, Kyotani S, Nagaya N, Fukushima K, Nakanishi N, Takishita S, Miyata A, et al. Increased plasma levels of adrenomedullin in patients with pulmonary hypertension. Clin Sci (Lond) 1999;96:33–39.
  49. Kandler MA, Von Der Hardt K, Mahfoud S, Chada M, Schoof E, Papadopoulos T, Rascher W, Dotsch J. Pilot intervention: aerosolized adrenomedullin reduces pulmonary hypertension. J Pharmacol Exp Ther 2003;306:1021–1026.[Abstract/Free Full Text]
  50. Ishikawa F, Miyazono K, Hellman U, Drexler H, Wernstedt C, Hagiwara K, Usuki K, Takaku F, Risau W, Heldin CH. Identification of angiogenic activity and the cloning and expression of platelet-derived endothelial cell growth factor. Nature 1989;338:557–562.[CrossRef][Medline]
  51. Moghaddam A, Zhang HT, Fan TP, Hu DE, Lees VC, Turley H, Fox SB, Gatter KC, Harris AL, Bicknell R. Thymidine phosphorylase is angiogenic and promotes tumor growth. Proc Natl Acad Sci USA 1995;92:998–1002.[Abstract/Free Full Text]
  52. Reynolds K, Farzaneh F, Collins WP, Campbell S, Bourne TH, Lawton F, Moghaddam A, Harris AL, Bicknell R. Association of ovarian malignancy with expression of platelet-derived endothelial cell growth factor. J Natl Cancer Inst 1994;86:1234–1238.[Abstract/Free Full Text]
  53. Wright L, Tuder RM, Wang J, Cool CD, Lepley RA, Voelkel NF. 5-Lipoxygenase and 5-lipoxygenase activating protein (FLAP) immunoreactivity in lungs from patients with primary pulmonary hypertension. Am J Respir Crit Care Med 1998;157:219–229.
  54. Simon R, Radmacher MD, Dobbin K, McShane LM. Pitfalls in the use of DNA microarray data for diagnostic and prognostic classification. J Natl Cancer Inst 2003;95:14–18.[Free Full Text]



This article has been cited by other articles:


Home page
 Am. J. Respir. Cell Mol. Bio.Home page
I. P. Laumanns, L. Fink, J. Wilhelm, J.-C. Wolff, R. Mitnacht-Kraus, S. Graef-Hoechst, M. M. Stein, R. M. Bohle, W. Klepetko, M. A. R. Hoda, et al.
The Noncanonical WNT Pathway Is Operative in Idiopathic Pulmonary Arterial Hypertension
Am. J. Respir. Cell Mol. Biol., June 1, 2009; 40(6): 683 - 691.
[Abstract] [Full Text] [PDF]

Home page
 BioinformaticsHome page
K.-S. Lynn, L.-L. Li, Y.-J. Lin, C.-H. Wang, S.-H. Sheng, J.-H. Lin, W. Liao, W.-L. Hsu, and W.-H. Pan
A neural network model for constructing endophenotypes of common complex diseases: an application to male young-onset hypertension microarray data
Bioinformatics, April 15, 2009; 25(8): 981 - 988.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Cell Mol. Bio.Home page
T. M. Bull, C. A. Meadows, C. D. Coldren, M. Moore, S. M. Sotto-Santiago, S. P. Nana-Sinkam, T. B. Campbell, and M. W. Geraci
Human Herpesvirus-8 Infection of Primary Pulmonary Microvascular Endothelial Cells
Am. J. Respir. Cell Mol. Biol., December 1, 2008; 39(6): 706 - 716.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
M. T. Saavedra, G. J. Hughes, L. A. Sanders, M. Carr, D. M. Rodman, C. D. Coldren, M. W. Geraci, S. D. Sagel, F. J. Accurso, J. West, et al.
Circulating RNA Transcripts Identify Therapeutic Response in Cystic Fibrosis Lung Disease
Am. J. Respir. Crit. Care Med., November 1, 2008; 178(9): 929 - 938.
[Abstract] [Full Text] [PDF]

Home page
 GutHome page
M Persico, M Capasso, R Russo, E Persico, L Croce, C Tiribelli, and A Iolascon
Elevated expression and polymorphisms of SOCS3 influence patient response to antiviral therapy in chronic hepatitis C
Gut, April 1, 2008; 57(4): 507 - 515.
[Abstract] [Full Text] [PDF]

Home page
 Proc Am Thorac SocHome page
M. M. Wurfel
Microarray-based Analysis of Ventilator-induced Lung Injury
Proceedings of the ATS, January 1, 2007; 4(1): 77 - 84.
[Abstract] [Full Text] [PDF]

Home page
 Proc Am Thorac SocHome page
T. M. Bull, C. D. Coldren, M. W. Geraci, and N. F. Voelkel
Gene Expression Profiling in Pulmonary Hypertension
Proceedings of the ATS, January 1, 2007; 4(1): 117 - 120.
[Abstract] [Full Text] [PDF]

Home page
 Physiol. GenomicsHome page
T. Dolinay, N. Kaminski, M. Felgendreher, H. P. Kim, P. Reynolds, S. C. Watkins, D. Karp, S. Uhlig, and A. M. K. Choi
Gene expression profiling of target genes in ventilator-induced lung injury
Physiol Genomics, September 14, 2006; 26(1): 68 - 75.
[Abstract] [Full Text] [PDF]

Home page
 Clin. Cancer Res.Home page
I. Osman, D. F. Bajorin, T.-T. Sun, H. Zhong, D. Douglas, J. Scattergood, R. Zheng, M. Han, K. W. Marshall, and C.-C. Liew
Novel blood biomarkers of human urinary bladder cancer.
Clin. Cancer Res., June 1, 2006; 12(11): 3374 - 3380.
[Abstract] [Full Text] [PDF]

Home page
 HypertensionHome page
H. Chon, M. C. Verhaar, H. A. Koomans, J. A. Joles, and B. Braam
Role of Circulating Karyocytes in the Initiation and Progression of Atherosclerosis
Hypertension, May 1, 2006; 47(5): 803 - 810.
[Abstract] [Full Text] [PDF]

Home page
 Circ. Res.Home page
E. D. Michelakis
Spatio-Temporal Diversity of Apoptosis Within the Vascular Wall in Pulmonary Arterial Hypertension: Heterogeneous BMP Signaling May Have Therapeutic Implications
Circ. Res., February 3, 2006; 98(2): 172 - 175.
[Full Text] [PDF]

Home page
 ChestHome page
M. Rabinovitch
Cellular and Molecular Pathobiology of Pulmonary Hypertension Conference Summary
Chest, December 1, 2005; 128(6_suppl): 642S - 646S.
[Full Text] [PDF]

Home page
 ChestHome page
M. Rabinovitch
Cellular and Molecular Pathobiology of Pulmonary Hypertension Conference Summary
Chest, December 1, 2005; 128(6_suppl): 642S - 646S.
[Full Text] [PDF]

Home page
 Am. J. Respir. Cell Mol. Bio.Home page
A. Zaiman, I. Fijalkowska, P. M. Hassoun, and R. M. Tuder
One Hundred Years of Research in the Pathogenesis of Pulmonary Hypertension
Am. J. Respir. Cell Mol. Biol., November 1, 2005; 33(5): 425 - 431.
[Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
T. D. Bradley, Y. E. Miller, F. J. Martinez, D. C. Angus, W. MacNee, and E. Abraham
Interstitial Lung Disease, Lung Cancer, Lung Transplantation, Pulmonary Vascular Disorders, and Sleep-disordered Breathing in AJRCCM in 2004
Am. J. Respir. Crit. Care Med., April 1, 2005; 171(7): 675 - 685.
[Full Text] [PDF]

Home page
 Am. J. Respir. Crit. Care Med.Home page
D. Sheppard
Fishing in the Bloodstream: Insights into the Mechanisms of Pulmonary Hypertension?
Am. J. Respir. Crit. Care Med., October 15, 2004; 170(8): 827 - 828.
[Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Online Supplement
Right arrow All Versions of this Article:
200312-1686OCv1
170/8/911    most recent
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Bull, T. M.
Right arrow Articles by Geraci, M. W.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Bull, T. M.
Right arrow Articles by Geraci, M. W.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Proc. Am. Thorac. Soc. Am. J. Respir. Cell Mol. Biol.
Copyright © 2004 American Thoracic Society