As a result of recent research conducted by scientists from the University of California at San Diego, the palette of colors available to scientists conducting image-based live-cell analysis has grown.
As powerful a technique as high-content screening and other image-base cellular analysis techniques have recently become, they are still somewhat difficult to multiplex because of the limited availability of bright, stable fluorescent proteins.
Furthermore, assays such as those based on FRET (fluorescence resonance energy transfer) have become quite popular in the last several years as a way to track and quantify protein-protein interactions in living cells, and have become a key part of high-content screening.
The problem is, if scientists want to genetically engineer cells to express a fluorescent protein, it is usually GFP from Aequoria victoria, or one of its variants — typically yellow fluorescent protein or cyan fluorescent protein. These have been and continue to be useful fluorescent markers, but they are somewhat limiting for multiplexing and FRET-based experiments.
Instrument makers have even had to design their platforms with these most popular fluorescent proteins in mind: Almost every high-content imaging system has laser lines or excitation filters designed for GFP and family.
A few years ago, UCSD research Roger Tsien and colleagues developed a red fluorescent protein, mRFP1 — an improved version of the Discosoma DsRed fluorescent protein. The latter had a tendency to clump together inside cells, which is why Tsien’s group engineered the monomeric mRFP1. But even mRFP1 demonstrated decreased brightness and stability, making it a relatively poor choice for most live-cell imaging applications.
Other academic scientists and companies like Quantum Dot and Evident Technologies have attempted to bridge the color gap with semiconductor nanocrystal technology. And although these “quantum dots” have far superior optical qualities to any fluorescent protein, and reach well into the red and even infrared regions of the spectrum, their relative size has limited them somewhat for live cell applications. Another obvious limitation of quantum dots is that they are externally applied dyes, and cannot be genetically linked and expressed in cell alongside normal proteins.
In the December 2004 issue of Nature Biotechnology, however, a team led by Tsien introduced several new fluorescent proteins with some rather fruity names — mCherry, mBanana, mOrange, mStrawberry, mTangerine, and mHoneydew — that may just fill the void between traditional GFP and mRFP1.
The fluorescent proteins, which range in color from yellow to red, descended from the mRFP1 molecule, but have superior properties to their predecessor, the paper states.
“We subjected mRFP1 to many rounds of directed evolution using both manual and fluorescence-activated cell sorting-based screening,” the paper noted. “The properties of the resulting variants include several new colors, increased tolerance of N- and C-terminal fusions, and improvement in extinction coefficients, quantum yields, and photostability.”
According to the paper, to engineer the fluorescent proteins, Tsien and colleagues used PCR, followed by FACS analysis on a BD Biosciences cell-sorting instrument to screen the resulting large libraries of mutants.
Although each of the fluorescent proteins had specific strengths and weaknesses, the researchers wrote that the mOrange protein was the brightest monomer. They also demonstrated that mOrange would make an ideal FRET partner for GFP, providing an alternative to its typical shorter-wavelength CFP and YFP partners.
The researchers also may have found a better alternative to the mRFP1 molecule that they developed only a few years back in mCherry.
“Among the true monomers, mCherry offers the longest wavelengths, the highest photostability, the fastest maturation, and excellent pH resistance,” the paper stated. “Its excitation and emission maxima are just 3 nm longer than those of mRFP1, for which it is the closest upgrade.
“Although mCherry’s quantum efficiency is slightly lower, its higher extinction coeffecient, tolerance of N-terminal fusions, and photostability make mRFP1 obsolete,” the paper continued.
Tsien and colleagues acknowledged that none of the new fluorescent proteins is perfect yet. For instance, they report that although tdTomato, a dimer protein, exhibited the highest brightness, it was at the expense of a doubling in molecular weight. MOrange, meanwhile, had maturation time, pH sensitivity, and photostability that were “far from optimal,” the researchers wrote.
Despite any shortcomings, the expanded palette of colors may provide a starting point for developing genetically engineered fluorescent proteins that span the spectrum — and may thus provide even more content to high-content screening.
But the researchers also believe that the technology they used to design the proteins — namely, better ways to manipulate their genetic constructs and analyze the results.
“Evolution of such proteins, or recombination of the best features of the existing proteins will probably require even higher-throughput means to generate genetic diversity, coupled with new screens run in parallel for all requisite performance criteria,” the paper stated.