"At the center of every cell, there beats a heart of centrin."
From a plaque outside Dr. Salisbury's office

As a Ph.D. botany student, Jeffrey Salisbury, Ph.D., made an observation that green algae have an organelle that attaches the nucleus to the centrioles and can contract and extend. Thus he discovered a small calcium binding regulatory protein that he subsequently named centrin. The discovery was published in Science (Salisbury, J. and G. Floyd (1978). "Calcium-induced contraction of the rhizoplast of a quadriflagellate green alga." Science 202: 975-977)

A man with a zest for life and a self-deprecating humor, Dr Salisbury is a self-described "midlife poster child." His office is decorated with artifacts from Einstein to the Grateful Dead. He rides a Harley-Davidson Fatboy and is an avid inline skater. The day he turned 50 ("a scary day") he celebrated by skating 50 miles. That's the kind of enthusiasm he brings to his lab. He extols the virtues of a green algae model system called Chlamydomonas reinhardtii, which he uses to study his beloved centrin.

"The genetics in algae make for a wonderful model system," says Dr. Salisbury. "We are diploid--we carry one set of chromosomes from each parent. But algae are haploid, they only carry one copy of their DNA, so when you mutate it--boom!--you see the effect immediately."

To generate the mutations, Dr. Salisbury's lab scrambles chromosomes through abnormal mitosis. He calls it "shuffling the deck." The effect is massive changes in chromosome number as well as gains and losses of chromosomes. This is called aneuploidy, which is a characteristic of all high grade tumors.

Studying Something Relevant

The leap from algae to breast cancer cells came years later at the suggestion of a Mayo colleague to study something relevant--like breast cancer.

"When we examined human breast tissue we were blown away by how dramatic and obvious the centrosome abnormalities were in the tumors," says Dr. Salisbury. "And that was literally on day one."

The first author on the paper (Proc. Natl. Acad. Sci. Vol. 95, pp 2950-2955, March 1998) was Wilma Lingle, Ph.D., who was a tumor biology post-doctoral fellow in Dr. Salsibury's lab at the time, but now directs her own lab as well as two of Mayo's core labs: the Biospecimens Lab and the Tissue and Cell Molecular Analysis Lab. Her passion for microscopy began with a Christmas gift of a microscope at the age of seven after which she stabbed her finger with a glass shard to see what blood looked like. The "use-by date" for her fascination is not yet up.

"The centrosomes are beautiful under fluorescence microscopy," says Dr. Lingle. "You can see the nucleus and tell what part of the cycle the cell is in. Then you can put together the many different pieces of information. It's a puzzle."

Dr. Lingle is particularly proud of "the Nikon Cool Scope." It's a fluorescence microscope and the first of its kind sold in the U.S. when Mayo purchased it in 2005. It doesn't look like a microscope. It looks like a computer and it can analyze tissue samples. In a breast tumor sample, for example, nuclei show up as little blue spots, which the software analyzes to measure the amount of DNA in each nucleus and gauge whether the cells have gained or lost chromosomes. The researchers then correlate the information with the mitotic behavior that they observe in the centrosomes, and also with the presence or absence of centrosome abnormality.

The 1998 study solved one piece of the puzzle. It demonstrated centrosome abnormalities in human tumors and characterized the abnormalities. Subsequent studies showed a correlation between centrosome abnormalities and chromosomal instability and a probability that centrosome abnormalities actually drive chromosomal instability in human tumors causing the aneuploidy seen in cancer.

A search of the early literature revealed that this was not a new idea.

"Amazingly, Theodor Bovari (who discovered the centrosome in 1888) first proposed that centrosome abnormalities underlie the origin of malignant tumors back in 1914," says Dr. Salisbury. "We have continued to study this in cancer and we've made a lot of headway but, frankly, we've just been rediscovering what we already knew in the algae and applying it to cancer biology so it's more relevant."

Relevant indeed. Their discoveries opened up a new field in cancer research -- centrosome amplification. Many investigators have since repeated the studies in other solid tumors and consolidated the concept that centrosome amplification is characteristic of aggressive tumors.

Estrogen and Centrosome Amplification in Breast Cancer

In collaboration with Jonathon and Sarah Li at the University of Kansas, Drs. Salisbury and Lingle are investigating the role of estrogen in driving centrosome amplification in the development of breast cancer. Since most breast cancers are estrogen dependent, the group's focus is on an estrogen responsive gene, Aurora kinase, which is localized at the centrosome.

A graduate student in Dr. Lingle's lab, Kara Lukasiewicz, recently proved that Aurora kinase phosphorylates (introduces a phosphate (PO4) group into a protein molecule) centrin. That's significant because the centrosomes cannot duplicate until centrin has been phosphorylated.

"Aurora kinase is overexpressed in 80 percent of breast tumors," says Dr. Lingle. "Interestingly, that's about the same percentage of breast tumors that have centrosome amplification. We think that increased expression of Aurora kinase is phosphorylating centrin leading to aberrant duplication of centrosomes."

There was 100 percent incidence of epithelial mammary tumors in a rat model, which they gave physiological but high levels of estrogen. In a hamster model, the same experiment produces an estrogen responsive kidney tumor. The same studies in breast cancer cell lines produce similar results and the group has also observed consistencies in humans.

"There are no other groups studying centrosomes that have access to the huge bank of human tissues and their accompanying clinical annotations as we do," says Dr. Lingle. "We can retrieve older archives and study expression in relevant genes and see how that relates to a patient's response to treatment, or to overall survival of patients with breast cancer. It will help us identify patients who might respond to one kind of treatment better than another."

Lynn Hartmann, M.D., is a breast cancer specialist who co-directs Mayo Clinic's Women Cancer Program. She is collaborating with Dr. Lingle to study a cohort of patients that have benign breast disease, called atypical hyperplasia.

"In the rat model, we see hyperplasia developing just prior to when we see the cancer, so we're pretty sure that this benign breast disease is actually a precursor to cancer," says Dr. Lingle. "This is important knowledge because women with atypical hyperplasia may choose to have more frequent physical examinations, experimental prevention drugs or prophylactic mastectomy."

Studying the Molecular Mechanisms

Dr. Salisbury has spent the last decade studying breast tumors to determine the origin of aneuploidy.

"I have been fortunate, over the years, to have had a number of talented trainees join my research group," says Dr. Salisbury. "Most recently, two Italian researchers have made significant contributions."

Antonio D'Assoro, M.D., Ph.D., from the University of Catania, Italy, studied the molecular basis for the origin of different types of breast tumors.

"The progression of human breast cancer from an estrogen-dependent to an estrogen-independent and more aggressive tumor represents a major clinical problem that limits the long-term usefulness of endocrine therapeutic strategies," says Dr. Salisbury. "Dr. D'Assoro's work has shown that the tumor suppressor gene p53 plays a major role in keeping centrosomes in check during the cell cycle. He recently developed a human tumor xenograft mouse model that mimics human breast tumor progression."

Cosima Quatraro, M.D., who is doing the lab phase of her Italian Ph.D. degree with Dr. Salisbury, is a valuable addition to the lab. They are now using the mouse model to test novel targets that may block centrosome abnormalities, thereby delaying tumor progression.

In order to better understand how molecules interact to cause centrosome abnormalities, Dr. Salisbury has also worked with nephrology researcher, Rajiv Kumar, M.D. and James Thompson, Ph.D., from the Mayo Proteomics Research Center, to crystallize the centrin molecule. The group was the first to accomplish this important task. (The structure of the human centrin 2-xeroderma pigmentosum group C protein complex. J Biol Chem. 2006 Jul 7;281(27):18746-52. Epub 2006 Apr 20)

"It turns out to be a beautiful molecule," says Dr. Salisbury. "And now we understand how centrin interacts with some of its target proteins."

The group is now taking what they've learned from human tumor cultures and trying to reproduce them in the mouse to further study the mechanisms by which centrosome amplification can give rise to aneuploidy. Dr. Salisbury credits much of the group's success to the Mayo environment.

"Nowhere else could we do the kinds of things we're doing here," says Dr. Salisbury. "Mayo provides an academic and research environment such that, if you want to pursue an area of biological relevance, nothing comes between you and those studies."

- Yvonne Hubmayr