A Scientific Household

Born in 1966, Bertozzi grew up in Lexington, Massachusetts. Her father, William, was a physics professor at Massachusetts Institute of Technology (MIT, Cambridge, MA).

“Science was focus of the house,” she says. “We wore MIT T-shirts and went to MIT day camp.”

When teachers asked Bertozzi and her two sisters what they wanted to be, Bertozzi remembers that their answers were “nuclear physicist.”

Their father often asked where they were going to get their Ph.D.’s, Bertozzi recalls. “He had this idea that all three of us would go to MIT—a mixture of pride and the promise of free tuition,” she says.

“To my dad’s great dismay, I chose Harvard instead,” Bertozzi notes. But she wasn’t the first to stray. Her older sister, Andrea, whom Bertozzi admits she always tried to emulate, had already “broken the ice” and chosen Princeton (Princeton, NJ).

Entering Harvard University (Cambridge, MA) as an undergraduate in 1984, Bertozzi had not yet chosen a major. She selected the school because it offered strengths outside of science.

For a brief time she thought about parlaying the keyboard skills that garnered her places in several college rock bands into a music major, but “I was always centered on the sciences.”

She began with biology and eventually majored in chemistry. “After taking organic chemistry, I fell in love with it,” she recalls.

For her senior thesis, Bertozzi worked with physical organic chemist Joe Grabowski (now at the University of Pittsburgh, Pittsburgh, PA) to build a photoacoustic calorimeter.

“You can think of it as complementary to fluorescence spectroscopy,” she explains. “Instead of measuring energy from an excited state in the form of light, it measures energy given off in the form of heat.”

The heat creates a local pressure wave, which has acoustic properties that can be measured with a piezoelectric microphone.

The technique is useful because not all molecules are fluorescent, but “far more molecules can get to an excited state and give off heat.”

The project consisted of much code writing and presented a challenge. “It was the first time I had an independent project. In many ways, the project was over my head, but Joe never let on,” Bertozzi explains. “He treated me like a graduate student.”

During her senior year, Bertozzi decided to apply to doctoral programs in chemistry.

“I liked the lab culture: the freedom, flexibility, challenges, and the late-night experimental marathons,” she says.

Manifest Destiny

Bertozzi, however, wanted to go further afield and was ready to explore new territory.

While investigating graduate schools, Bertozzi visited the University of California (Berkeley, CA), and immediately felt a connection. “I just knew it was the right place,” she explains. “Back then, there weren’t many women in chemistry Ph.D. programs, either students or faculty. I was hoping I could find a place where women were more enfranchised.”

Berkeley fit the bill, and it also served as the hotspot for a then-emerging field that brought biology and chemistry together.

“The interface with biology is a well-established sector of chemistry to today’s graduate students, but the field we now call chemical biology was a new concept back then.”

Bertozzi chose to work with Mark Bednarski, who had joined the faculty only a year before.

“He was the new guy,” she explains. “He had many innovative ideas at the interface of biology and chemistry” and was developing methods to engineer sugar analogs to study cell–cell and cell–virus interactions.

Finding the Sweet Spot

The first order of business meant synthesizing sugar analogs, a tough nut to crack.

“Glycans are one of the three major biopolymers, the others being proteins and nucleic acids,” Bertozzi explains.

Techniques to synthesize proteins and nucleic acids are advanced, but even today, she says, no machine exists in common use to synthesize glycans.

The intractable nature of the problem stems from the nature of the target. Sugars are highly functionalized, and their polymers can be branched. “This complicates the process,” she explains. Sugars need analogs for study because enzymes called glycosidases chew them, breaking the glycosidic linkage between the sugar building blocks.

Bertozzi sought to synthesize a class of stable sugar analogs called C-glycosides (2). These analogs possess a carbon atom where nature would position an oxygen atom. Bertozzi developed methods to prepare C-glycoside analogs of glycopeptides and glycolipids. Their glycans resisted glycosidase action, thus the analogs could be used in biological assays.

Bertozzi went beyond synthesis, collaborating with Francisco Gonzalez-Scarano and coworkers at the University of Pennsylvania (Philadelphia, PA) to test a particular C-glycoside analog of a glycolipid for its ability to bind to the HIV receptor gp120 (3).

Respect for Glycans

As Bertozzi continued with graduate school, glycans, “the elusive biopolymers” as she calls them, were gaining notability.

In the late 1980s, the field of glycobiology expanded considerably with the discovery that selectins, a family of adhesion molecules central to inflammation, bind to glycan ligands.

Suddenly, there was a “frenzy of activity directed toward figuring out what glycans structures bind to the selectins,” she recalls. “History will look back on that period as the time that glycobiology transformed from a niche field into something bigger in the minds immunologists and pharmaceutical researchers.”

One of the primary researchers searching for the structures of the glycans binding these selectins was Steve Rosen at the University of California (San Francisco, CA) who had a collaboration with Larry Lasky at Genentech (South San Francisco, CA). Rosen and Lasky’s groups had cloned L-selectin, the selectin expressed on leukocytes (4).

“Cloning a gene was a big deal,” Bertozzi says. “It was a different undertaking prior to the availability of the human genome sequence.”

Rosen’s group had performed a crude analysis of L-selectin’s glycan ligands, but the race was on to determine the detailed structure, and Bertozzi wanted to join Rosen’s team. Emphasizing the potential usefulness of her background in synthetic chemistry, she soon found herself as the lone chemist among immunologists.

“We were all working around the clock to figure out the structure,” she recalls. “If you can get enough material, mass spectrometry can reveal everything.” But pulling only small amounts of L-selectin ligands from mouse lymph nodes did not provide enough material.

Instead they performed radiolabeling studies, which involved digesting the labeled glycans with enzymes to see what sugars fell off. After approximately four years they figured out the structure (5, 6).

Determining the location of sulfate groups proved the hardest part.

“The actual glycan structure was not that unusual, but what was special was the fact that it was sulfated at specific positions,” says Bertozzi.

“I liked the lab culture: the freedom, flexibility, challenges, and the late-night experimental marathons.”

The enzymes that installed the sulfate groups changed the sugar from a low-affinity to a high-affinity ligand (7). The sulfate’s role in regulating the underlying glycan’s function reminded Bertozzi of a phosphate’s stimulatory role for the many proteins activated by kinase-mediated phosphorylation.

She wanted to know more about how sulfates regulate the activities of sugars in nature and sought to develop inhibitors of sulfating enzymes when she joined the Berkeley faculty in 1996.

TB for Two

In 1998, researchers published the Mycobacterium tuberculosis genome sequence (8). Bertozzi recalls that out of curiosity, graduate student Joseph Mougous, now on the faculty at University of Washington (Seattle, WA), scanned the genome for sulfotransferases by looking for genes similar to human ones.

Surprisingly, he found several, although the sulfotransferases were thought to be mainly eukaryotic. Mougous wanted to work on understanding the roles of the enzymes in tuberculosis, and Bertozzi agreed to the experimentation despite her lab’s lack of experience with the organism.

“We had little understanding of what the challenges would be,” she says. “Fortunately, we had some wonderful colleagues in our midst,” such as Lee Riley, also at Berkeley, and Jeffrey Cox (UCSF), among others.

The group showed that sulfated molecules help mediate host–pathogen interactions (9) and also characterized the molecular machinery underlying the biosynthesis of these sulfated molecules (10, 11).

Seeing Sugars

Since the start of her research career, Bertozzi has sought ways to image sugars in vivo.

In the early 1990s, her postdoctoral work feeding cells with radiolabeled sugars to probe the structures of L-selectin’s ligands seeded an idea for imaging sugars in vivo: feeding cells with unnatural sugars could act as a tagging mechanism.

Her ideas were supported by an encounter in 1994 with German biochemist Werner Reutter at a meeting in Southampton, England, where Bertozzi represented her then-mentor Rosen.

Reutter had discovered that a biosynthetic precursor of the simple sugar sialic acid could be structurally modified without significant detriment to its cellular metabolism (12).

Bertozzi thought that modified precursors would be a good way to introduce a sort of “chemical handle” that could serve as a reactive site for the attachment of imaging probes. But the reaction between modified sugar and probe would have to occur in living organisms for in vivo imaging applications and without unwanted side reactions with the myriad of biological functional groups.

Bertozzi’s group coined a term to describe reactions that can achieve this level of selectivity in biological systems—bioorthogonal—and they now refer to the two-step process of metabolic labeling followed by chemical reaction as the “bioorthogonal chemical reporter” method.

A few classic reactions from organic chemistry literature appeared to have the requisite bioorthogonality at first, such as the condensation of ketones with aminooxy or hydrazide reagents (13).

The process, however, proved too slow under physiological conditions.

“We had to invent new reactions,” adapting what was already in the literature for better performance in a biological environment, Bertozzi says. The work was time-consuming, with seemingly straightforward research papers failing to reflect the volume of trial and error involved. “One sentence took five years,” she explains.

They eventually found a path to success using the azide as a chemical reporter, which they incorporated into cellular glycans by feeding cells or injecting organisms with azidosugar precursors. The advantages of the azide include its small size and relative chemical inertness. Azides are not normally found in biological systems and have unique chemical properties.

Bertozzi and her colleagues mined the classic synthesis literature for reactive partners for the azide and eventually identified two reactions that were promising starting points for development of a bioorthogonal transformation: the Staudinger reaction of azides and phosphines and the Huisgen cycloaddition reaction of azides and alkynes.

Her group modified these reactions, both first reported in the early- to mid-1900s, to create what is now termed the Staudinger ligation (14) and strain-promoted reaction of azides and cyclooctynes, also called “copper-free click chemistry” (15).

The research team recently employed azidosugar metabolism followed by copper-free click chemistry in the first imaging study of glycans in a live organism (16).

Growing zebrafish embryos in media with azidosugars, Bertozzi and colleagues labeled temporally distinct populations of glycans in the fish, producing multicolor images to illuminate trafficking patterns of the glycans during development.

She chronicles the development of the imaging strategies and offers a look into work still needed in her Inaugural Article (1). In the article, the authors suggest that future research will need to focus on developing more unnatural sugars and additional chemical reporters to cover full the extent of the glycome—the totality of a cell’s glycans—as it changes over an organism’s development.

Vision in Action

In 2006, Bertozzi took on directorship of the Molecular Foundry, a Department of Energy-funded facility for nanoscience research located at the Lawrence Berkeley National Laboratory (Berkeley, CA). She had been involved with the project since its inception, first as the director of the institute’s Biological Nanostructures Facility.

The Molecular Foundry operates on the same model as a synchrotron facility, Bertozzi explains. Any researcher can apply for cost-free time to use the center’s equipment or to call upon its expert staff for help and training.

The work feeds her personal interest in materials science, which stems from her days as an intern at Bell Labs (Holmdel, NJ) in 1988 and work with her graduate mentor Bednarski. Materials science projects in her lab focus on merging synthetic materials with biological ones (17), making synthetic polymers to coat living cells to direct them to specific targets in the body (18) and developing new technologies to probe cells with nanomaterials (19).

“A lot of people are interested in molecular characterization of biological systems at the nanometer scale,” Bertozzi notes. “Optical probes have been useful, but they have limitations. Fluorescent reporters such as GFP can light up proteins, but many molecules are not amenable to its use. And the resolution one can achieve using optical probes is limited by the wavelength of visible light; it is difficult to resolve features smaller than 50 nm, even with modern superresolution techniques.”

Bertozzi wants to explore how mass spectrometry can be used to image molecules, as doing so may be able to reveal the location of a molecule in a natural biological system without the need for attached probes.

Bertozzi explains that using nanoscale tools to manipulate molecules for mass spectrometry could, for instance, help understand the molecules expressed by the tuberculosis bacterium while it is actually in the lung, a feat that is difficult, “especially if they are not proteins.”

For Bertozzi, the tiny, seemingly imperceptible scale of nano and the hard-to-study glycans are coming together and her work is helping to illuminate important fundamentals of biology every day.