Bacterial Spore Detection and Determination by Use of Terbium Dipicolinate Photoluminescence David L. Rosen* U.S. Army Research Laboratory, AMSRL-IS-EE, 2800 Powder Mill Road, Adelphi, Maryland 20783-1197 Charles Sharpless and Linda B. McGown Department of Chemistry, Duke University, Box 90346, Durham, North Carolina 27708 A new method to detect bacterial endospores and deter- mine their concentration was demonstrated by the addi- tion of a solution of terbium chloride to a suspension of bacterial endospores. The terbium chloride reacted with the calcium dipicolinate in the spore case to form terbium- (III) dipicolinate anion. Solid particles, including residual bacterial particles, were removed by filtering. The pho- toluminescence from the solution was measured as a function of excitation wavelength, emission wavelength, and bacterial endospore concentration. The photolumi- nescence from terbium(III) dipicolinate anion in the solution was easily identified. A new method for detecting bacterial endospores and deter- mining their concentration is presented that could be very useful in many applications. For example, manufacturers could rapidly monitor bioreactors that make endospore suspensions (such as bacterial insecticides). Public health workers could more rapidly monitor indoor environments, water quality, or food quality. 1,2 Paleontologists, who now can detect only viable endospores in rock and soil, 3,4 could detect nonviable endospores. Bacterial endospores are much more durable than vegetative (i.e., biologically active) bacterial cells. Endospores resist anti- septics, antibiotics, desiccation, and ordinary boiling more than vegetative cells. Bacterial strains are sometimes identified by these types of resistance. Therefore, the new method could be useful in evaluating some antibacterial and diagnostic techniques by measuring the number of endospores independently of the vegetative cells. Bacterial endospore concentrations are not easy to determine. The main methods of quantification, microscopy and plate culture counting, are slow and tedious. 5 Although previous investigators have studied the intrinsic photoluminescence of bacteria for identification, 6-8 the intrinsic photoluminescence is not specific to bacterial endospores. Our new method is faster than any other method that is specific to bacterial endospores. We here explain and show experimental validation of the new method. The method of detecting and determining bacterial spores consists of the following three steps. First, terbium chloride (i.e., TbCl 3 ) is added to an aqueous suspension that may contain bacterial endospores. The terbium cation (Tb 3+ ) reacts with calcium dipicolinate (i.e., Ca(dpa)) in any spore case present to form terbium(III) dipicolinate (i.e., [Tb(dpa) 3 ] 3- ) anion, a chelate. Second, particles (including the residual bacterial particles) are removed from the terbium-treated suspension by filtration. Be- cause [Tb(dpa) 3 ] 3- is soluble, it is easily separated from insoluble concomitants. Third, the photoluminescence of [Tb(dpa) 3 ] 3- is measured. To explain why this method works, we compare Ca- (dpa) to [Tb(dpa) 3 ] 3- . Calcium dipicolinate (i.e., Ca(dpa)) is the major component in spore cases of bacterial endospores 2 but is otherwise uncom- mon. Previous investigators have used the absorbance spectrum of Ca(dpa) to measure bacterial endospore concentrations. 9 For many other compounds, measuring photoluminescence is a more sensitive method for determining concentration than measuring absorbance. However, Ca(dpa) does not generate photolumines- cence. Terbium(III) dipicolinate (i.e., [Tb(dpa) 3 ] 3- ), in contrast to Ca- (dpa), has a very strong and distinctive photoluminescence spectrum. Terbium cation (i.e., Tb 3+ ) reacts with dipicolinate anion (i.e., dpa 2- ) to form [Tb(dpa) 3 ] 3- . The positions of the peaks in the emission spectrum of [Tb(dpa) 3 ] 3- are unchanged from those of Tb 3+ , and the emission bands of [Tb(dpa) 3 ] 3- are very narrow because of shielding by outer s and p orbital electrons in the terbium atom. 10 At a given concentration and excitation intensity, the photoluminescence intensity of [Tb(dpa) 3 ] 3- is far greater than that of terbium cation, Tb 3+ , alone. 11 Previous investigations that used dipicolinic acid to detect Tb 3+ led to the idea of the new endospore determination method. 12,13 Dipicolinic acid (i.e., H 2 dpa) was added to a sample, releasing dipicolinate anions (i.e., dpa 2- ), which reacted with Tb 3+ to form [Tb(dpa) 3 ] 3- , from which photoluminescence spectra were mea- sured. Because both Ca(dpa) and H 2 dpa release dpa 2- in solution, we predicted that the Ca(dpa) from bacterial endospores would (1) Alcamo, I. E. Fundamentals of Microbiology; Addison-Wesley: Reading, MA, 1984. (2) Salle, A. J. Fundamental Principles of Bacteriology, 7th ed.; McGraw-Hill: New York, 1973. (3) Fischman, J. Science 1995 , 268, 977. (4) Bello, N. L. Sci. Dig. 1967 , 61, 70-73. (5) Brock, T. D.; Brock, K. M. Basic Microbiology with Applications; Prentice- Hall: Englewood Cliffs, NJ, 1973; pp 69-78. (6) Bronk, B. V.; Reinisch, L. Appl. Spectrosc. 1993 , 47, 436-440. (7) Dalterio, R. A.; Nelson, W. H.; Britt, D.; Sperry, J. F.; Tanguay, J. F.; Suib, S. L. Appl. Spectrosc. 1987 , 41, 234-241. (8) Nachman, P.; Chen, G.; Pinnick, R. G.; Hill, S. C.; Chang, R. K.; Mayo, M. W.; Fernandez, G. L. Appl. Opt. 1996 , 35, 1069-1076. (9) Lewis, J. C. Anal. Biochem. 1967 , 19, 327-337. (10) Richardson, F. S. Chem. Rev. 1982 , 82, 541-552. (11) Metcalf, D. H.; Bolender, J. P.; Driver, M. S.; Richardson, F. S. J. Phys. Chem. 1993 , 97, 553-564. (12) Ma, W.; Hwang, K. J.; Lee, V. H. L. Pharm. Res. 1993 , 10, 204-207. (13) Barela, D. B.; Sherry, A. D. Anal. Biochem. 1976 , 71, 351-357. Anal. Chem. 1997, 69, 1082-1085 1082 Analytical Chemistry, Vol. 69, No. 6, March 15, 1997 S0003-2700(96)00939-0 CCC: $14.00 © 1997 American Chemical Society