Antibody Technique Shows Diagnostic Promise

April 2008

In the February 1 issue of the journal Biochemical and Biophysical Research Communications, a team of NIH researchers report early results with a tremendously sensitive and accurate new diagnostic technique to quantify antibodies in blood and saliva. Known by the acronym LIPS, the technique performed without error in a small validation study involving a well-known antigen that is frequently elevated in people with a rare disorder called Stiff-Person Syndrome. Additional articles will be published in the months ahead for more common autoimmune conditions, ranging from primary Sjögren’s syndrome to type-1 diabetes. The Inside Scoop spoke to two of the authors to learn more about the technique and its potential. They are NIDCR scientists Dr. Peter Burbelo, lead author on the study, and Dr. Michael Iadarola, the paper’s senior author.

Iadarola – The acronym stands for luciferase immunoprecipitation technology. But don’t let the jargon fool you. LIPS is a straightforward diagnostic system, and that’s one of its major advantages. Peter came up with the idea a few years ago as a researcher at Georgetown University, and it has great potential as a clinical tool to profile the level of essentially any antibody of interest in blood or saliva involved in an autoimmune response.

Burbelo – Right, LIPS produces such a high signal. Let me give you an example. The existing ELISA antibody assays typically have a dynamic range of between 5 and 15,000 units of signal. In plain English, it’s a measure of light intensity. How bright is the green, or how bright is the red. So, a negative sample might be 5,000 and under, and a positive outcome might register between 5,000 and 15,000 units of signal. But that’s the highest level that they see. With LIPS, we see samples that range from zero to sometimes over a million.

Burbelo – Several things. When you do an ELISA, it means you’ve taken a protein made by E. coli or another non-mammalian source and spot it on a plastic dish. You’re assuming the protein looks like it does naturally in solution. But its three-dimensional shape will be very different once it is affixed to plastic. So, the antibodies will recognize a stretch of amino acids here and maybe there. But they won’t recognize other pieces of the protein that have denatured and lost their natural conformations.

Iadarola – Exactly. Another part of the story is we think it’s a cleaner method.

Iadarola – As Peter mentioned, other assays use non-mammalian cells as cloning vectors to make a human protein of interest as bait in an ELISA test. But the proteins typically are impure in that they bear remnants of the bacterial, yeast, or insect vector that made them. It’s possible that other antibodies in blood will recognize these foreign remnants and, since they are stuck on the walls of the plastic dish, detect them. That compromises the sensitivity of the test. With LIPS, that’s not the case. We create a fusion protein using a mammalian cell. The fusion protein consists of two parts: (a) Any antigen of interest, and (b) a luminescent enzyme called Renilla luciferase that produces a flash of light.

Could you quickly define the terms?

Iadarola – Sure, an antigen is a molecule that stimulates the immune system to attack and, in our case, to produce antibodies. The luciferase enzyme comes from a marine organism called a sea pansy, or Renilla reniformis. We fuse this enzyme to the antigen. Why? It serves as our signal. In other words, after antibodies have bound to the antigen, we add a chemical substrate to the mix called coelenterazine. The luciferase snips, or converts, the substrate and, in the process, emits a flash of light like a firefly. We can read the intensity of the flash, and that tells us how much antigen is bound to the antibody.

Okay, so let’s walk through the process?

Burbelo – Let’s say hypothetically you wanted to know whether you’ve been exposed to Hepatitis C. If you gave me a pindrop of your blood, I’d walk back into the laboratory, where I have a partially automated robotic system set up to run the antigen-antibody binding reaction. Once everything is loaded, the robot adds the necessary fluids to a plastic plate with 96 individual small wells. Each well contains a bead. As the assay proceeds, the antibody and Hepatitis C antigen/luciferase fusion protein stick on the bead. The robot washes the beads to remove any excess fluids and proteins. But the antigen-antibody complex remains stuck to the bead. Then we add coelenterazine, the flash occurs, and we read the intensity of light with a luminometer. If a lot of light is produced, it means you have a lot of antibodies to the hepatitis C antigen. If we see no light, it means you’ve probably never been infected. What’s nice is the system will quantify the intensity of the flash, and I’ll have the results in matter of minutes.

Burbelo – We’re already doing it in fact. 

Iadarola – Right now, protein profiling technologies spot proteins on glass or a plastic membrane. Like the ELISAs, spotting can alter the natural three-dimensional folds of the proteins and affect the sensitivity of the antibody recognition and binding. Again, LIPS sees more of the protein, and that feature was extremely appealing for us to pursue because of its many clinical possibilities.

Burbelo: Also, gene profiles typically evaluate a person’s susceptibility to develop a disease during their lives. It’s a measure of relative, not absolute, risk. With LIPS, it tells you whether you’ve actually been exposed to an infectious agent or whether an autoimmune response is under way and its level of intensity.

Burbelo – That’s right. Another thing is because of the zero-to-million range of the signal, LIPS also has the potential to serve as a health maintenance tool.

Burbelo – With the existing antibody tests, remember your diagnostic window is limited. But, if you’re starting way up here with a zero-to–million range, your reading of the disease process is much more sensitive and informative. You may actually be able to monitor disease progression or therapeutic response once treatment begins. Also one may be able to detect the emergence of a disease early in its course and in a preventive or predictive fashion.

Burbelo – Yeah, that’s what’s really exciting about this system. I can take a pindrop of blood and say, “From two weeks ago, your antibody titer has dropped significantly since you’ve started taking this antioxidant.” It looks like the treatment is working. It’s potentially something really useful.

Iadarola - SPS is a rare but often devastating neurological disorder. People develop a debilitating rigidity in their spine and legs. This leads to changes in posture and muscle spasms, along with a spate of other possible symptoms. In my laboratory, we’re very interested in the role of antibodies in sustaining persistent pain states due to nerve injury. SPS isn’t necessarily painful but is representative of a central nervous system autoimmune disorder. The patients raise an antibody to a very important enzyme that is found in neurons in the brain and spinal cord. About 70 percent of patients have greatly elevated levels of glutamic acid decarboxylase, or GAD, particularly a subtype of the enzyme called GAD65. It plays a role in the synthesis of a major nerve signaling chemical.

Burbelo – LIPS batted 1,000. Our NIH collaborators had previously collected 40 blood samples – half belonged to people diagnosed with SPS and the others were drawn from healthy controls. In all instances, we were blinded to the identity of the samples to avoid bias. As we reported, the LIPS assay of GAD65 had 100 percent sensitivity and specificity for the SPS patients. In our statistical analysis, we found the signals, or light intensity, from the controls and the lowest detected antibody level in the SPS patients were separated by the mean, or average control GAD65 antibody titer, plus 23 standard deviations. Existing antibody tests can show differences of three or four standard deviations. In other words, LIPS sensitivity was off the charts.

Iadarola – That’s right. This enzyme can be elevated to a lesser degree in people with Type-I diabetes, drug resistant epilepsy, cerebellar degeneration syndromes, and Batten disease. Our strategy here was first to try the technique in Stiff-Person Syndrome. But we’ve already begun to test it in other conditions with anti-GAD65 antibodies, and the results will be forthcoming over the next several months. We’re also looking at other autoimmune conditions such as primary Sjögren’s syndrome.

Burbelo – As we mentioned a minute ago, that’s a real possibility. We already have reagents to profile 23 different infectious agents. In some cases, such as HIV and hepatitis C, we have the whole proteome, or the complete collection of proteins that these viruses manufacture. Over the last three years, I’ve generated hundreds of antigen-fusion proteins for antibody profiling and probing disease mechanisms, and this raises another likely application of this technology. Its high signal is custom made to discover new antigens and protein biomarkers for virtually any autoimmune, neurological, and infectious disease, as well as some other diseases not commonly associated with provoking an antibody response. Now, we just need to generate the data and hopefully translate this promise into additional tangible results.