Michael McClelland, PhD, Professor

Dr. McClelland's Publications
mmcclelland@skcc.org

Director, Cancer Genetics Program

Laboratory Staff: Steffen Porwollik, PhD, Yipeng Wang, MD, Gaelle Rondeau, PhD, Rocio Canals PhD, Sang Ho Choi MD, Nabil Arrach, Ph.D., Pui Cheng, Huazhen Yao, BS.

Supervisor of Fred Long BS, in Bioinformatics Core.

Supervisor of Xiao Qin Xia. PhD, in Genomics Core.

Michael McClelland works in the area of genomics, a field that aims to interpret the language of DNA. DNA methylation: Michael McClelland was the first to describe that the regulatory regions of mammalian genes, called promoters, contain sequences called “CpG islands”. Methylation of these features (addition of a methyl group to the DNA) is important in marking genes for regulation. Dr McClelland studies the alteration of methylation during the dys-regulation of genes that occurs in cancer. As part of this work the McClelland lab constructed the first “promoter microarrays” of DNA fragments from 10,000 promoters and collaborated on the first Chromatin immunoprecipitation-array (ChIP-chip) experiments, which identify proteins bound to promoters in living cells. Prostate cancer prognostics: The McClelland lab is involved in a large multicenter grant to identify gene expression changes in prostate cancer that are associated with increased risk of recurrence of the disease after prostatectomy. The McClelland lab performs the RNA work and much of the bioinformatics for this project. Genomics: Michael McClelland developed the first DNA cleavage methods for making physical maps of DNA molecules millions of bases long. He organizes the sequencing of major pathogens, including Salmonella typhimurium, resulting in papers in Nature and Nature Genetics. The lab provides microarray resources and techniques to over 50 labs worldwide for study of Salmonella genomes, transcription, and mutant library screens. One recent paper in Science coauthored by the McClelland lab describes how Salmonella recruits virulence genes from other bacteria and then uses a global trans-regulator to prevent undesired transcription of these in the absence of environmental cues. Cancer therapy: Harmless variants of Salmonella prefer to reside in tumors over any other location in the human body, by a factor of 1000 or more. These bacteria can cause cancer cures. The lab has engineered improvements in Salmonella as a therapeutic and novel delivery agent in cancer. Newly developed high-throughput tools are used to identify promoters that are turned on only in tumors. Using such specific promoters, cloned foreign therapeutic proteins could be produced only when bacteria reach tumors. In addition, tools have been developed to track thousands of bacterial mutants simultaneously, which allows mutants with favorable characteristics, such as those that accumulate even better in tumors, to be isolated and further engineered.

This laboratory is applying high-throughput methods in a number of areas:

Program 1: Salmonella Genomics.
Background. Salmonellae are among the most important orally acquired pathogens in the world. Approximately one million deaths and 100 million human infections are caused annually by these organisms, and they are major pathogens of domestic livestock. On a genomic level, strains within this species can differ by hundreds of their ~4,500 genes. This variability is the result of a mosaic of lateral transfer events within a constant genome scaffold. These differences result in an extraordinary diversity of host-ranges and pathogenic presentations between strains. Known virulence functions are encoded primarily within these laterally transferred regions. However, despite 30+ years of intensive study, the functions of most genes within these important regions are still unknown. We are in the process of rectifying this gap in knowledge.

Objective. Understand the evolution of pathogenesis in Salmonella with the objective of generating principles applicable to other diseases, allowing new methods of treatment, and also exploitation of avirulent strains for human needs.

Approaches.

A. We have sequenced strains used in laboratory research and are now sequencing 30 more strains that capture the diversity of the species. Understanding sequence diversity:

> Allows steps in evolution to be better understood.
> Allows development of DNA typing methods that correlate better with infectious manifestations.
> Allows the construction of specific in-frame knockouts of every gene (1150 so far) as a critical resource for our functional genomic studies.

B. We have developed oligonucleotides microarrays for high-throughput analysis of:

> RNA expression.
> Screening promoter-GFP libraries for all promoters.
> Studying the fitness of all genomic regions, simultaneously, in animal gut, systemic infection, and the environment.

Program 2: Salmonella as a therapeutic delivery agent in cancer.

Background. Harmless live vaccine strains of Salmonella naturally accumulate 1000X in tumors. Tumors have lower than normal levels of oxygen, where Salmonella will continue to thrive protected from the immune system. Salmonella may also further deplete oxygen to levels thereby killing and growing on the remains of tumor cells. This leads to cures.

Objective. Improve upon the natural ability of Salmonella to kill tumors. In addition, Salmonella can be used to deliver therapeutic agents, such as enzymes for drug metabolism and cytokines.

Approaches. In order to understand and improve the ability of Salmonella to kill tumors, and develop this ability for therapeutic use we are taking the following steps:

> Identifying promoters active in tumors to express therapeutics.
> Identifying mutants that grown better than wild type bacteria in tumors.
> Identifying avirulent strains that grow well in cancer.
> Determining the usefulness of these mutations in a variety of animals so they can be used to treat tumors in companion animals.

Program 3: Biomarkers prostate cancer recurrence after prostatectomy.
Background. Prostate cancer is the most common cancer in men (70% of 70 year olds!). Tens of thousands die each year because the tumor has metastasized before treatment. However, the vast majority of prostate cancers do not lead to death even if untreated but most cases are treated, nevertheless, because there is no test for accurate identification of indolent versus aggressive cases. More than 100,000 men are treated unnecessarily each year in the US, at a cost of billions of dollars, and with some side-effects in most cases.

Objective. Develop prognostic tests to accurately define those patients with progressive disease, so they can be targeted for early aggressive treatment.

Approaches. We accumulated clinical data over decades for hundreds of consented patients.
> Developed method that assigns predictive markers to tumor or adjacent reactive tissues. Note: most biopsies do not contain tumor but changes occur in these samples, nevertheless.
> RNA strategy licensed to company.
> Building large tissue microarrays for protein prognosticators.
> Developing DNA methylation assays (see below).

Program 4: DNA methylation biomarkers for cancer progression.

Background. For prostate cancer, the sources of materials for prognostics include needle biopsies, blood, and urine, as well as prostatectomies. Prognostic tests can include pathology, DNA, RNA, or protein, or combinations thereof. DNA is more stable than RNA and is found in all of these types of samples. The methylation profile is unaltered in partially degraded DNA after cell death.

Objective. Develop a cancer prognosticator based on DNA rather than RNA or protein.

Approaches. We developed method that samples hundreds of thousands of DNA methylation loci throughout genome.

> The prostate tumor samples in Program 3 are being screened to find differences in DNA methylation that are prognostic.
> Markers are being converted to an assay platform that can be used in a clinical service laboratory.

Protocols and Arrays Information

The NCI “Director’s Challenge” program for the Molecular Characterization of Early Stage Prostate Cancer