We develop new biophysical and optical methodologies for biomedical research and clinical applications. Of late, we have been focusing on the refinement of technological approaches to characterizing early stages of disease and to developing strategies for prevention of progression and for monitoring responses to therapy in cancer and age-related macular degeneration (AMD). In further developments of Laser Capture Microdissection and related laser technologies, we have coupled the isolation of small populations of cells from sections of complex tissues with global analysis of gene expression. We have applied statistical analysis of gene expression to identify candidate lists for stage-specific molecular markers within microdissected cell populations, focusing on improvements necessary to identify critical but less abundantly expressed genes.

Laser Microdissection Technology Development
Mertts, Obiakor, Parlato, Bonner; in collabora-tion with Mage, Obiakor, Emmert-Buck
As the list of expressed human genes is completed, a major scientific and medical challenge will be to track the complex molecular events that drive normal tissue differentiation and progression of pathologic lesions. With refinements in PCR, hybridization microarrays, and mutation screening, both DNA and mRNA can be extracted from tissue biopsies or cytology specimens and analyzed with a parallel panel of hundreds or even thousands of genetic markers. Given that cells in complex tissue are biochemically and physically affected by surrounding cells as well as by remote stimuli from greater distances, the task of analyzing critical gene expression patterns in development, normal function, and disease progression depends on the extraction of specific cells from their complex tissue milieu. Our laboratory, in collaboration with NCI, has invented and developed Laser Capture Microdissection (LCM) (Arcturus Engineering, Inc., has commercialized the technology). LCM provides a rapid, reliable method to procure pure populations of specified cells from microscopic regions of tissue sections in a form useful for subsequent quantitative, multiplex molecular analysis. Routine clinical use of this approach will require the reliable preservation of the macromolecules of interest within a sample, assurance that critical clonal populations are included and can be identified within an ex vivo sample, and quantitative extraction and analysis of critical subsets of marker macromolecules.

In an effort to lead technology integration with biological and clinical research, we are studying the polymer physics and chemistry of the LCM processes and competitive ablative laser microdissection techniques. Recent improvements include “noncontact LCM” prototypes that use automated beam steering and pulse sequences. The prototypes allow reliable activation of flexible transfer films and thus accommodate the injection of picoliter volumes of polymer into individually targeted cells, permitting capture and retraction of the cells over precisely controlled air gaps of 10 to 20 microns. A new method for simultaneous complete release of targeted cells from the microscope slide is based on simultaneous ablation of 100-nm-thick polymer coatings. Using standard aqueous tissue extraction buffers, we have demonstrated quantitative recovery of DNA, mRNA, and proteins from the injected polymer matrix. The LCM process, which loads picoliter cell volumes onto equivalent volumes of polymer matrix, is being applied to integrated purification and analysis of specific macromolecular constituents. We have been able to emboss a patterned microfluidic circuit into the LCM polymer tapes and directly load selected cells into a specified well of approximately one nanoliter volume.

In parallel efforts, we have been investigating the biophysics of UV laser ablation and have demonstrated the role played by polymer breakdown and ionization in electrostatic interactions determining target collection efficiency. Polymer multilayer coatings with charge neutralization appear to improve ablative laser microdissection, which has particular advantages for collecting larger populations of cells required for automated cytometry and many proteomics discovery efforts. In collaboration with NIAID, we have used nested PCR and DNA sequencing to demonstrate reliable molecular analysis in LCM-isolated immunolabeled single cells at the single-molecule level. As we proceed to single-cell isolation and greater knowledge of molecular changes in pathology, specific molecular targeting will become increasingly important to ensure rapid, precise collection of subpopulations that are not discernible by morphology alone. We are incorporating fluorescence microscopy and real-time image analysis methods, including remote image–guided sample targeting.

Gene Expression Complexity and Statistics of Gene Expression
Kuznetsov, Stepanov, Bonner; in collaboration with Emmert-Buck, Mulshine
If microdissection and molecular analysis can be made clinically practical, the expression levels of sets of approximately 20 to 100 critical, stage-specific disease markers within a selected cell population might provide reliable diagnosis and intermediate endpoints of response to molecular therapies in individual patients. We have been developing robust statistical tools to facilitate identification of critical genes and pathways that characterize both normal and aberrant cellular function, particularly those suitable for “multiplex” analysis of a small number of specific cells. Analysis of large gene expression databases (SAGE and LCM cDNA libraries) suggests that a large fraction of all genes is expressed in any specific cell type and that the genes’ expression levels universally exhibit a highly skewed power-law distribution similar to those characterizing many other complex systems. We are developing mathematical models for the evolution of such distributions that evidence increasing biological complexity.

To follow macromolecular changes during cancer progression within prostate, breast, lung, and ovary tissues, we have developed, in collaboration with NCI, standard procedures for the isolation of normal and pathological cells from clinical specimens. Using statistical multivariate analysis, we have developed mathematical algorithms for the selection of “candidate genes” from cDNA libraries. The candidate genes’ expression frequencies have a strong correlation with disease progression. We use our models of the statistics of expression levels in cell populations to identify genes differentially expressed in cancer progression. To date, our analysis points to a critical role for many less abundantly expressed genes at a critical stage of ovarian cancer progression and suggests that, for most cancers, critical diagnostic marker sets should include such low-abundance transcripts. This notion is guiding our research in statistics of low-level gene expression and suitable detection methods.

Gene Expression during Normal Development and Pathology Progression
Parlato, Malekafzali, Bonner; in collaboration with Gruber, DiLauro, Mushinski
We have applied single-cell LCM capture technology to studies of gene expression during normal development (spermatogenesis and thyroid bud development) and in stage-specific pathological cells (e.g., cancer precursors). In collaboration with Roberto DiLauro’s laboratory, we examined gene expression patterns associated with the primordial thyroid and the adjacent cells from which it derives in both wild-type and knock-out mice. In collaboration with Fred Mushinski, we are applying the same LCM methodology to study early changes in the well-established plasmacytoma model in BALB/c mice. Jointly, we have developed robust LCM techniques for isolation of specific cells from mouse cryosections and recovered good yields of high-quality mRNA. In collaboration with Life Technologies, Inc., we have made cDNA libraries of high diversity and message length from 100 ng of microdissected mRNA (~5,000 cells). Using Affymetrix microarray hybridization, we can determine global gene expression profiles in these libraries while preserving the libraries for isolation of the full-length message of identified stage-specific genes. We are planning to apply these methods, which have undergone validation in reproducible animal models, to clinical specimens of early stages of disease progression in which microdissection of critical rare cell populations is required. We foresee an evolution of molecular diagnosis from one based on the qualitative or quantitative analysis of a few key macromolecules to one in which special clustering algorithms analyze complex multivariate databases. Such analyses should permit a more complete identification of highly correlated clinical cases and allow us to characterize their response to molecular therapies specifically designed to prevent progression. To this end, we have refined the LCM of rare cell populations, particularly those that might be accessible by serial, minimally invasive cell harvesting from patients. We have continued to organize and hold an annual conference at NIH on Laser Capture Micro-dissection and Macromolecular Analysis of Normal Development and Pathology, bringing together about 500 researchers to discuss research and methodological advances made possible by LCM.

Prevention of Progression of Age-Related Macular Degeneration through Photoprotection
Meyers, Ostrovsky, McConnell, Bonner
Age-related macular degeneration (AMD) is the most common cause of severe visual loss in the elderly. Early stages of the disease are ophthalmoscopically detected with increasing frequency with age (particularly over 60 years). Lipofuscin accumulates with age in the retinal pigment epithelium (RPE). It is the dominant fluorphore in the retina, and, in Stargardt’s disease, its increased rate of accumulation is associated with early macular degeneration. This finding led us to hypothesize in the early 1980s that lipofuscin is photochemically active and responsible for acute retinal damage at high blue light levels, causing chronic photochemical (free radical production) stress that increases with age (lipofuscin accumulation). If lipofuscin photo-chemistry plays a critical role in initiation and progression of early AMD, then simple optical filters (yellow-colored lens) might be expected to reduce dramatically the rate of photochemical free radical production. We designed such filters (Bonner: 500-nm-long pass sunglasses; Ostrovsky: 430-nm-long pass intraocular lens), which have been used clinically. The recent discovery of a photochemical active product (A2E) created by rod photoreception at high light levels and our experimental determination of A2E photo-chemical action spectrum in RPE cell culture led us to reinvestigate optimal designs for optical filters to reduce or prevent progression of early AMD. We have generated models of spectral irradiance in the human macular RPE following cataract surgery and intraocular lens (IOL) implantation and as the lens ages. The models suggest the importance of the reduction of both the maximal bleaching of rod rhodopsin and the rate of singlet oxygen generation from photo-toxic photoproducts (trans-retinal, A2E, and its photoproducts), along with the significance of the role of photodegradation of A2E and its toxic photoproducts by visible light. We have proposed new IOLs and long wave-length transmitting “sunglasses” for reducing rod activation in bright ambient light as well as the accumulation and activation of toxic byproducts of photo-transduction. In addition, we are seeking to characterize spectrally different RPE photoproducts to evaluate the potential for noninvasive monitoring of the molecular effects of such filters in patients.