News Release

You have to break a few (hundred) eggs to make a good crystal

Peer-Reviewed Publication

NASA/Marshall Space Flight Center--Space Sciences Laboratory

Bell curve shape to crystal quality may point to best candidates for flight

Sept. 20, 1999: Did you ever ask the teacher to grade a tough test "on the curve"? What you were asking was that the grades be adjusted so that a "C" fell under the part of the curve where most of your classmates had scored. A few were to the left and got a D or F; and few were to the right and got a B or an A.

That's basically how the bell curve works. In nature, objects and events quite often can be grouped along a bell curve. In a population of adult animals, most will be around the same size. A few will be larger and a few will be smaller.

"If you talk to statisticians," noted Dr. Russell Judge of the University of Alabama in Huntsville, "variations within populations in nature can be described in terms of distributions."

Judge is not a statistician but a chemical engineer who has found that protein crystals may follow much the same pattern as they form. In the current issue of Biophysical Journal, Judge, working at NASA's Marshall Space Flight Center with other researchers from UAH, NASA/Marshall, and Michigan State University at East Lansing, report that even under ideal conditions, crystals exhibit a variation in crystal quality.

The question now is whether scientists can use the microgravity of space to shift the curve to the right to grow the large, nearly perfect crystals they need for molecular lock-picking, the first step in designing drugs that can treat a broad range of diseases and disorders.

"We want to determine how the growth of crystals effect their quality," Judge said in May when NASA selected his investigation for development, "and then take that into space to see how microgravity is enhancing the growth characteristics that lead to good crystals. From this we want to develop techniques, so that by observing crystal growth on the ground, we can predict which proteins are likely to benefit the most from microgravity crystallization."

Proteins are complex chemicals that regulate virtually every function in a body, such as a virus's attack and the body's response.

These functions are a result not just of a chemical formula, but of structure which can be quite large (on the atomic scale) and fragile. If the shape isn't right, the protein cannot match up with other proteins or chemicals to do its job, just as the wrong key won't unlock a door. Sickle cell anemia, for example, results from structural differences in the hemoglobin that carries oxygen in red blood cells. Designing new treatments means designing altered proteins or other chemicals that act as a skeleton key or as a sophisticated lock pick.

Proteins can form crystals, generated by rows and columns of molecules that form up like soldiers on a parade ground. Shining X-rays through a crystal will produce a pattern of dots that can be decoded to reveal the arrangement of the atoms in the molecules making up the crystal. Like the troops in formation, uniformity and order are everything in X-ray crystallography. X-rays have much sorter wavelengths than visible light, so the best looking crystals under the microscope won't necessarily pass muster under X-rays.

This has become an invaluable tool for understanding the structure and the function of dozens of proteins. But many proteins remain shrouded in mystery because on Earth crystal imperfections are introduced by fluid flows and the settling of the crystals to the bottom of the container. This leaves internal defects that distort or blur the view of the structure.

"In order to have crystals to use for X-ray diffraction studies," Judge said, "you need them to be fairly large and well ordered." Scientists also need lots of crystals since exposure to air, the process of X-raying them, and other factors destroy the crystals. Getting just one perfect specimen isn't enough. Dozens may be needed, and the quality might not be known until well into the analysis.

In 1998, NASA accepted Judge's proposal to determine the best conditions under which the best crystals would grow.

Growing protein crystals in the microgravity of space has yielded striking results, such as determining to a fine resolution how certain molecules of insulin join so scientists can improve injectable insulin needed by diabetics. There have also been disappointments when crystals in other experiments did not grow as expected.

Since the 1970s, scientists have used a variety of approaches in trying to determine what leads to the growth of a large, perfect crystal. Judge tried a different approach that built on results noted by researchers dating as far back as 1946.

He and his team looked at the effects of concentration, temperature, and pH (acid vs. base) on the growth of lysozyme, a common protein in chicken egg white. Lysozyme's structure is well known and it has become a standard in many crystallization studies on Earth and in space. Although lysozyme has an atomic mass of 14,300 Daltons - almost 92 times that of the ordinary sugar that many of us crystallized in elementary school science - it's a relative lightweight in the protein world.

To exclude impurities often found in commercial lysozyme preparations, Judge and his team purified lysozyme extracted from eggs obtained from a local egg farm. While one experiment run required only five dozen eggs, the full series of experiment consumed about 200 eggs.

Judge and his team grew the crystals in trays with small plastic wells filled with solutions containing a trace of salt to help stimulate crystal growth. Temperatures ranged from 4 to 18 deg C (39-64 deg F) and pH from 4.0 to 5.2 (slightly acidic; pure water theoretically has a pH of 7). Judge also varied the driving force behind the crystal growth process, called supersaturation, by varying the initial concentration of protein. Protein concentration must be set above a critical limit, the solubility, in order to form crystals (below this concentrations the protein stays dissolved and never forms crystals)

The tough part was examining each of the over 2000 wells and counting the crystals. It turned out that the solution pH had the largest effect on the growth of the crystals, possibly due to changes in charges on the surface of the molecules.

When solution conditions had been optimized to give a small number of large crystals, a sample of 50 crystals was withdrawn for X-ray diffraction analysis. Judge hoped that when the ideal conditions were found and then applied to subsequent batches, he would be able to grow consistently large, high quality crystals of lysozyme. The expectation was that with ideal conditions, quality crystals could be cranked out as if in a factory.

Instead, nature put him on the curve.

"Some variation is occurring there," Judge said, "but we haven't quite pinpointed the cause."

Judge got a bell curve when he measured the X-ray clarity, properly called the signal-to-noise ratio (a radio with static has a low signal-to-noise ratio). A graph of the number of crystals versus the signal-to-noise ratio forms a bell curve, albeit slightly skewed to one side.

"The distribution is saying a very few crystals form perfectly in solution," he continued, "and a small number are really poor. The majority of crystals are in-between."

It's doubly puzzling because the crystals were grown from the same batch of lysozyme that was poured into 120 wells in the experiment tray and crystallized under the same conditions.

"We have some ideas," Judge said, "but we haven't tested them yet, so we're hesitant to say it might be this or that." The research will continue with insulin, the crucial protein that conveys sugar from the blood stream into a body's cells, and with glucose isomerase, a larger (46,000 Daltons) protein used in industrial processes to convert glucose to a sweeter sugar called fructose.

"In all of the proteins we're using the structure is pretty well known," Judge added.

In addition to ground-based experiments, Judge hopes to conduct flight experiments in the next year or so. He would use the Vapor Diffusion Apparatus, a device developed by the University of Alabama in Birmingham and well-proven in a number of Space Shuttle flights.

"Most researchers say that crystals grown in microgravity will be better than those on the ground," Judge said. And a number of experiments bear out that belief. "Somehow, microgravity pushes up the end of the distribution curve."

With expected flight experiments on lysozyme, insulin and glucose isomerase, Judge will have crystals grown in conditions as close as possible to the ideal conditions he had determined so far. At the same time, he will grow crystals on Earth from the same mix as the flight batch and using identical hardware and conditions so that microgravity is the only variable.

Eventually, he hopes that his studies will lead to a tool for screening candidate proteins for flight.

###

The Effect of Temperature and Solution pH on the Nucleation of Tetragonal Lysozyme Crystals. Biophysics Journal, September 1999, p. 1585-1593, Vol. 77, No. 3

Russell A. Judge, 1 Randolph S. Jacobs, 2 Tyralynn Frazier, 3 Edward H. Snell, and Marc L. Pusey4

1Alliance for Microgravity Material Science and Applications, NASA/Marshall Space Flight Center, Huntsville, Alabama 35812;
2Department of Chemical Engineering, University of Alabama in Huntsville, Huntsville, Alabama 35899;
3Biochemistry Department, Michigan State University, East Lansing, Michigan 48825;
and 4Biophysics SD48, NASA/Marshall Space Flight Center, Huntsville, Alabama 35812 USA

Part of the challenge of macromolecular crystal growth for structure determination is obtaining crystals with a volume suitable for x-ray analysis.

In this respect an understanding of the effect of solution conditions on macromolecule nucleation rates is advantageous. This study investigated the effects of supersaturation, temperature, and pH on the nucleation rate of tetragonal lysozyme crystals. Batch crystallization plates were prepared at given solution concentrations and incubated at set temperatures over 1 week. The number of crystals per well with their size and axial ratios were recorded and correlated with solution conditions. Crystal numbers were found to increase with increasing supersaturation and temperature. The most significant variable, however, was pH; crystal numbers changed by two orders of magnitude over the pH range 4.0-5.2. Crystal size also varied with solution conditions, with the largest crystals obtained at pH 5.2. Having optimized the crystallization conditions, we prepared a batch of crystals under the same initial conditions, and 50 of these crystals were analyzed by x-ray diffraction techniques. The results indicate that even under the same crystallization conditions, a marked variation in crystal properties exists.


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.