Have advanced research technologies made real impact on respiratory medicine?

INTRODUCTION

The past 20 years have seen remarkable advances in technology applied to biologic samples to help understand the physiologic and pathologic processes involved in human health and disease. Successes such as the sequencing of the human genome in 2001¹ brought great excitement to the scientific community, as did many of the technologic advances, such as proteomics and broad spectrum RNA analysis (microarray). Significant advances in both genome and protein sequencing have followed and along with genetic manipulation of rodent species, we now have many new tools at our disposal to attack the dilemma of lung disease. Most recent grant submissions require us to incorporate the newest approaches to investigation of both human and experimental models of lung disease. Some findings have been highly instructive, others add to a large mountain of information that still needs to be processed. Below, we summarize a number of the advanced technologies and what these have done, or as yet failed to do, for human lung diseases. Microarray analysis and genetic manipulation have dramatically influenced experimental research in the past decades, while systematic use of proteomics has just begun. Small interfering RNA are currently evaluated for their therapeutic potential and may become the first highly specific molecular therapies available for human disease.

MICROARRAY GENE EXPRESSION ANALYSIS

Microarray Gene Array, also known as ‘Gene Chip’, is a multiplex technology which allows detection of the expression of thousands of genes that are actively transcribed in a living cell or tissue. Since the introduction of high throughput profiling methods in the last decade of the 20th century,^2,3 they have rapidly become a mainstay of basic, translational and clinical research but have not yet entered clinical practice. More than 35 000 papers using gene expression microarrays have been published, a number that only underestimates the usage as the majority of such studies remain unpublished. Microarray studies that analyse gene expression patterns in IPF, COPD, asthma, acute lung injury, cystic fibrosis and other lung disease have been published.⁴ Among others, these studies led to identification of new molecular targets, definition of new molecular classes and identification of biomarkers.

Two of the major barriers to implementation of these technologies in human pulmonary research have been addressed in the first decade of the current century. In human lung disease the major barrier was historically the availability of the target organ as the majority of lung diseases are managed without a biopsy. Thus, centres that had very active lung transplantation programs have been at the forefront of this research because they could collect end-stage explanted tissues from patients with IPF, COPD and CF. More recently, an NHLBI initiative ameliorated this issue by creating the Lung Tissue Resource Consortium (LTRC) that collected well-phenotyped samples from lung biopsies for advanced lung disease, lung biopsy excess material from tumour and nodule resection and explanted tissues.⁵ The public availability of LTRC tissues has made it feasible for many investigators to engage in profiling studies without requiring a large clinical operation. Other approaches for sample access have emerged, such as airway epithelial cell brushings in asthma and COPD and the use of peripheral blood gene expression patterns. The other barrier to implementation of high throughput technologies for gene expression is funding—these technologies are cost intensive and investigators have been frustrated with the difficulty to get grant support for profiling studies. The increase in publications and the maturation of analytical approaches provided sufficient preliminary data to overcome the traditional aversion of grant reviewers to support ‘fishing expeditions’. Moreover, funding agencies, especially NHLBI, have actively engaged in creating requests for applications (RFA) that sought implementation of such technologies—these RFAs led to the funding of large projects that characterize tissue and peripheral blood gene expression profiles in most of the advanced lung diseases, even as recently as the last American Recovery and Reinvestment Act funds (http://www.nhlbi.nih.gov/recovery/).

Taking into consideration the published papers as well as the recently funded research projects one expects that the impact of gene expression profiling technologies will be felt in a significant way in pulmonary research and medicine within the next years. One has also to posit that this impact will be felt beyond the impact on investigators careers, not only measured in amounts of funding or high impact papers but also in implementation in clinical management of patients with advanced lung disease. The most probable impacts of these technologies in lung disease include:

GENE THERAPY AND GENE EXPRESSION MODULATION

Over the past 20 odd years there have been a number of significant advances in our knowledge of the pathogenesis of lung diseases made through the use of genetic analysis, and application of animal models in which a gene of interest has been modulated through genetic manipulation. These advances in molecular understanding are remarkable, but unfortunately, few findings have yet made their way to clinical benefit. Gene therapy, the buzz word of the 1990s, was expected to be the therapeutic advance that provided the solution for Cystic Fibrosis, but despite exciting in vitro and in vivo experimental data, including mouse knock out data, translation to functional long term correction of the human genetic defect still eludes us.

We have seen extensive use of sophisticated experimental mouse gene manipulation studies to examine the participation of gene products in the pathogenesis of various lung diseases including asthma, COPD acute lung injury and fibrosis. These studies have used transgenic mice, in which the gene of interest is randomly inserted into the genome and mouse lines are derived that over-express various levels of the transgene under control of ubiquitous or cell specific promoters or specific genes are silenced or knocked-out (KO). However, many (if not all) of the factors and signalling systems that are thought to be involved in the pathogenesis of lung diseases are apparently equally involved in developmental programs of lung biology. Therefore, the findings are prone to misinterpretation of developmental modifications rather than direct effects in the adult, and more informative methods of inducible expression or even inducible KO systems in adult mice, coupled with lung cell specific expression, are required to make the findings mean more than mere implications.

Despite the limitations of the interpretations, the gene modulation approach has lead to significant findings related to human lung disease. In asthma, there has been some disappointment in translating the findings of extensive gene manipulation studies to human disease. For example, genetic manipulation studies in mice have long praised IL-5 as a key regulator in asthma,⁶ but several clinical trials to block IL-5 in broad spectrum asthma have failed. However, recent trials clearly suggested a role for anti-IL-5 in severe steroid resistant asthma.⁷ Another example is the role of transforming growth factor (TGF)-β1 in pulmonary fibrosis. Inducible or transient over-expression of TGF-β1 in the lung initiates the fibrogenic response.^8,9 These observations, coupled with data using the integrin αVβ6 KO¹⁰ or the Smad 3 KO mouse, showed that interfering with TGF-β activation and associated signalling cascade are viable targets for therapy of human fibrosis^10–14 and some clinical trials for IPF are currently developed.¹⁵ Moreover, the same experimental approach using Smad3 KO mice showed that a direct link exists between inflammation, TGF-β generation and fibrogenesis, and that it may be more beneficial to directly target the fibrogenesis rather than inflammation in IPF.¹⁶ Certainly, however, the chemokine data from extensive genetic manipulations has not translated to any great extent into effective therapeutic interventions in human lung disease, nor has there been any significant impact to date on clinical treatment of COPD derived from findings with gene manipulated systems in mice.

The attempts to induce gene manipulation targeted at removal of specific cell types have resulted in significant findings, most of which not yet translated directly to human therapies. However, success with implicating CD4+ cells in asthma¹⁷ or CD8+ cells¹⁸ and neutrophils¹⁹ in COPD has helped target efforts towards some new therapies. Recent exciting findings in liver fibrosis, using gene modified systems to remove macrophages,²⁰ suggest we need to target this cell type to effect resolution of fibrotic disease in other tissues. These examples of success, though limited, argue that better directed and controlled systems for gene manipulation and more complex systems involving not only one gene product at a time, should yield more interpretable information for human lung diseases, and hopefully to the development of effective therapeutic interventions.

PROTEOMICS AND POST-TRANSLATION PROTEIN MODIFICATIONS

Proteomics, the wide-scale investigation of proteins and their multiple isoforms, is one of the newer technologies to be applied to investigations of lung tissue and diseases. It is a particularly challenging endeavour in lung disease as lung tissue is composed of about 50 different cell types. The possible number of proteins in a single cell can reach an enormous complexity as a single gene can generate multiple different proteins through alternative splicing and post-translational modifications. Some have estimated this number to reach several hundreds of thousands. This ‘protein pool’ is variably expressed, depending on all internal and external factors a cell might encounter. In addition, the study of one single protein is cumbersome as proteins might be localized in various subcellular compartments and bind to different biochemical complexes. As a provocative example, a single protein expressed at similar levels in two different cells could be involved in distinct molecular signalling pathways depending on its level of activation (i.e. phosphorylation) or ubiquitination (which means that a protein is tagged to be recycled through the proteosome). Separation and identification involves a combination of approaches, including automated sophisticated multi-dimension electrophoresis, specific protein labelling, gas chromatography, mass spectroscopy and high throughput rapid sequencing, as well as specific antibody arrays.

Studies investigating respiratory diseases have so far largely focused on differential identification of proteins present in whole blood (or fractions of blood), broncheoalveolar lavage fluid, pulmonary oedema fluid, saliva, sputum or exhaled breath condensate (see references in Rottoli et al.²¹). What have emerged are large lists of differentially expressed proteins in various respiratory diseases. While the overall goal of most of the above mentioned studies was to identify biomarkers in disease, newer approaches aim to investigate the temporal expression of proteins using more quantitative approaches.

Protein synthesis and degradation rates need to be estimated in order to generate a dynamic picture of the fluctuating protein pool. Several strategies have been used, for example the ‘stable isotope’ labelling of cells in culture with incorporation of radio-labelled amino acids combined with mass spectroscopy. By replacing ‘heavy’ with ‘light’ media at various time-points, investigators are able to reflect the dynamic protein pool and its turnover using the ratio between ‘heavy’ and ‘light’ proteins detected by mass spectroscopy.²² A multitude of new quantitative proteomic analyses have emerged over the past few years and the interested reader can refer to excellent reviews for further details.^21,23,24 However, no direct clinical impact of such studies has yet been manifest in lung disease.

The ability to analyse large datasets will ultimately yield a greater knowledge about cell and tissue biology/pathology. Pathways to meaningful discovery will depend on the ability to analyse and understand large datasets obtained by different investigators. A number of simplifications can reduce the difficulty of analysis. Harmonized gold standard sample preparation and reduced sample complexity are critical factors. Techniques such as laser capture micro-dissection (LCM) allows capture of single cell populations from a culture or tissue slice (rather than total tissue homogenate) and can be used in conjunction with microarray analysis and also proteomic assessment.²⁵ Subcellular fractionation or organelle purification from single cells has been used to reduce the total pool of proteins and to investigate protein signalling.^26,27

Proteomic approaches have started to deliver knowledge in respiratory diseases, but the future might see datasets that need to be understood in the larger context of systems biology as investigators will be able to assess and correlate protein pools from various fluids, organelles, cells and tissue from experimental models and from patients, in order to yield new hypotheses and mechanistic insight of respiratory cell and tissue pathobiology.

LOCALIZED SIRNA AS THERAPEUTIC OPTIONS FOR LUNG DISEASE

The discovery of RNA interference (RNAi), a natural mechanism for downregulation of gene expression, opened up new perspectives for the development of RNA-based therapies.^28,29 The method of small interfering RNA (siRNA)-mediated knockdown of specific genes was thus heralded as a method with infinite therapeutic options due to its simplicity and target specificity.³⁰ RNAi is mediated by siRNA, which are 19–23 base pair-spanning double-stranded RNA that specifically lead to the degradation of target mRNA and thereby indirectly reduce target protein expression. The RNAi pathway operates in most eukaryotics and is initiated by the enzyme Dicer, which cleaves long double-stranded RNA molecules into short fragments of ∼20 nucleotides. In therapeutic applications, siRNA is targeted to a specific gene, introduced into the cells of interest, where the siRNA unwinds and is incorporated into a stable protein-RNA complex. If the siRNA is targeted to an mRNA that is involved in certain diseases, this mRNA undergoes degradation, thereby interrupting protein synthesis of this gene. Thus, the outcome of siRNA application is post-transcriptional gene-specific silencing.²⁸

In medicine, effective delivery, as well as sufficient potency, specificity and stability of the siRNA are prerequisites required to achieve selective mRNA cleavage for efficient gene silencing in vivo.^31–33 In recent years, intranasal or orotracheal administration of formulated siRNA resulted in effective gene silencing in several diseases, including experimental pulmonary fibrosis,³⁴ viral respiratory tract infections,³⁵ or ischemia-reperfusion injury.³⁶ In these studies, siRNA were applied in different ways: first, local delivery to the lungs was performed as well as systemic delivery. Second, ‘naked’ siRNA were applied as well as siRNA packaged into liposomes or similar lipophilic solutions. And third, non-stabilized siRNA were used as well as stabilized and cholesterin-coupled siRNA. The most promising siRNA-mediated therapeutic approach is most likely the reduction of viral infection with respiratory syncytial virus,³⁵ as this was successful both before and after infection. While other diseases have been reported to benefit from siRNA application as well, confirmatory studies are still lacking in most cases.

In principle, the lungs represent an optimal organ for local siRNA delivery, as small substances can be delivered to the lung either via the vasculature or locally via the airways. Despite being a promising novel therapeutic tool, the in vivo applicability of siRNA is currently limited by apparent siRNA instability in blood, inadequate cellular uptake and low bioavailability. The most challenging innovation that is required for effective siRNA-mediated therapy in lung diseases is the targeting of siRNA to specific lung cell populations. Although no direct clinical benefit has yet emerged from the siRNA technology, further refinement of delivery systems that will allow delivery of siRNA into the relevant intracellular compartment will surely make a substantial contribution to specific therapies for chronic lung disease.

SUMMARY

Tremendous sophistication in our ability to detect, measure and manipulate the genes and proteins involved in lung disease has provided unique insights to the pathobiology of a number of lung disorders affecting humans today. To date, there has been little direct impact on therapeutic interventions available to clinicians. However, the rapidity with which these early data have accumulated and the expected advances in bioinformatics and systems biology should bring forward innovative approaches and druggable targets for many acute and chronic lung diseases within the next 10 years and, after all, three decades from start to clinic is not a long time in traditional drug discovery, to provide meaningful and important advances of clinical benefit in human lung disease.