Mass Spectrometry-based Identification of Lyme Disease-associated Proteins as a Model for the Universal Detection of Bacterial Infections

Lyme disease is an infectious disease that can affect multiple organ systems, including the nervous system, cardiovascular system, joints, and muscles.  The prototypical causative agent in North America, the bacterium Borrelia burgdorferi [1, 2], is transmitted to humans through the bite of infected ticks, and nearly 300,000 people in the United States are afflicted each year [3].  If left untreated, Lyme disease may lead to chronic joint inflammation, neurological disorders, cognitive defects, and heart rhythm irregularities, resulting in decreased quality of life and significant health care costs.  Ultimately, analytical methods for B. burgdorferi detection are critical for early diagnosis, and evaluation of treatment efficiency and post-treatment symptoms.

At present, Lyme disease is diagnosed based on a number of symptoms, including fever, headache, fatigue, and a characteristic bull’s-eye skin rash called erythema migrans, and the possibility of exposure to infected ticks.  Blood tests are available for Lyme disease, but the accuracy of the tests depends upon the stage of the disease, and the tests are indirect in that they measure the patient’s antibody response to the infection, and not the infection itself.  During the first 4 to 6 weeks of infection, these tests are unreliable because most patients have not yet developed the antibody response the tests measure.  If a patient exhibits erythema migrans, diagnosis and treatment can begin earlier; however, the rash does not occur in 20 to 30 percent of Lyme disease patients.  Direct detection of B. burgdorferi is technically feasible using polymerase chain reaction (PCR) assays [4-7], but a number of drawbacks exist.  For one, the concentration of B. burgdorferi in the blood is low at all stages of infection, and so there may not be enough DNA for the assay, resulting in a false negative.  In cases where DNA can be detected, there is no way to distinguish between live and dead bacteria, creating the possibility of a false positive [8].  Clearly, more reliable and direct detection methods are needed to improve Lyme disease diagnosis and treatment.

Researchers at IBBR, NIST, and Johns Hopkins University School of Medicine have developed an analytical method that allows for early and direct detection of B. burgdorferi.  The assay involves mass spectrometry-based identification of a target B. burgdorferi protein, ospA, which is normally associated with the bacterial outer membrane, in human serum.  The challenge was to detect the bacterial membrane protein among the more abundant host serum proteins.  The researchers hypothesized that bacterial protein-containing membrane vesicles released from B. burgdorferi cells as a result of the host innate immune response [9-11] could be isolated from serum by high-speed centrifugation and the proteins subsequently detected using mass spectrometry.

The detection and quantification of a target protein by mass spectrometry requires the addition of a stable isotope-labeled internal standard to the sample to be measured.  The researchers needed an unambiguous indicator of the bacterium and ultimately chose five B. burgdorferi proteins as targets because their amino acid sequences were distinguishable from human serum proteins.  These five target proteins were then used to design a QconCAT internal standard.  Ultimately, one particular target protein, ospA, the major B. burgdorferi surface lipoprotein, could be detected at an average level of 4.0 fmol of ospA per mg of serum protein in the serum obtained from three patients diagnosed with Lyme disease, which means the method can detect a target protein that is 107 lower in abundance than the major serum proteins.

Although a long way off from being implemented in the clinic, this method could prove useful for Lyme disease detection and treatment, and could be expanded to identify other bacterial infections in humans, particularly where existing diagnostics are unreliable.

This work was published as: C. S. F. Cheung, K. W. Anderson, K. Y. Villatoro Benitez, M. J. Soloski, J. N. Aucott, K. W. Phinney, and I. V. Turko, “Quantification of Borrelia burgdorferi Membrane Proteins in Human Serum: A New Concept for Detection of Bacterial Infection.” Analytical Chemistry (2015) 87, 11383-11388. doi: 10.1021/acs.analchem.5b02803

Read more here:


  1. Lyme disease graphs: Reported Cases of Lyme Disease by Year, United States, 1995-2014 (2015, November 5) Retrieved from
  2. Hinckley, A. F., et al. (2014) Clin. Infect. Dis. 59, 676-681.
  3. How many people get Lyme disease? (2015, September 30) Retrieved from
  4. Borchers, A. T., et al. (2015) J. Autoimmun. 57, 82-115.
  5. Nocton, J. J., et al. (1994) N. Engl. J. Med. 330, 229– 234.
  6. Cerar, T., et al. (2008) J. Clin. Microbiol. 2008, 46, 3375– 3379.
  7. Eshoo, M. W., et al. (2012) PLoS One 7, e36825.
  8. Iyer, R., et al. (2013) J. Clin. Microbiol. 51, 857-862.
  9. Rus, H., et al. (2005) Immunol. Res. 33, 103-112.
  10. Kochi, S. K., et al. (1991) J. Immunol. 146, 3964-3970.
  11. Breitner-Ruddock, S., et al. (1997) Med. Microbiol. Immunol. 185, 253-260.