Monochloramine Inactivation of Bacterial Select Agents

Abstract

Seven species of bacterial select agents were tested for susceptibility to monochloramine. Under test conditions, the monochloramine routinely maintained in potable water would reduce six of the species by 2 orders of magnitude within 4.2 h. Bacillus anthracis spores would require up to 3.5 days for the same inactivation with monochloramine.

Monochloramine is the second most common disinfectant used for drinking water, after free chlorine. A 1998 survey found that 29.4% of 200 large and medium-sized water treatment systems rely on chloramines for disinfection (2). Little information is available regarding the monochloramine susceptibility of bacteria on the Center for Disease Control's list of bioterrorism agents. Concern for the security of drinking water supplies has led to questions about the efficacy of current disinfection practices against these bacterial biothreat agents. Surface Water Treatment Rule (SWTR) guidance manuals (15, 16) provide tables of Ct values to indicate conditions necessary for a 2-log₁₀ inactivation (99%) or a 3-log₁₀ inactivation (99.9%) of a given pathogen. Ct values are defined as the product of the disinfectant concentration (in milligrams/liter) and the time (in minutes) that potential pathogens in water are in contact with the disinfectant. Ct values are assigned for various microorganisms and are specific for a given temperature and pH of the water.

In the present study, seven bacterial biothreat agents (10 isolates) were exposed to preformed monochloramine, and Ct values were calculated for 2-log₁₀ and 3-log₁₀ inactivation. These studies were conducted at three temperatures representative of a range found within water distribution systems, 5°C, 15°C, and 25°C (pH 8 for all temperatures).

Organisms included in this study were Bacillus anthracis Sterne 34F2, B. anthracis Ames, Burkholderia pseudomallei KC872, Burkholderia mallei M-9, Brucella suis MO562, Brucella melitensis ATCC 23456, Francisella tularensis NY98, F. tularensis LVS, Yersinia pestis A1122, and Y. pestis Harbin. B. anthracis was cultured on soil extract-peptone-beef extract agar (3) for 7 days to encourage spore formation. The cells and spores were then transferred to centrifuge tubes and treated with 50% ethanol for 1 h and then washed five times with sterile ultrapure water before they were stored in 50% ethanol at 5°C. F. tularensis isolates were cultured on Cysteine heart agar (3), and all other isolates were cultured on Trypticase soy agar with 5% sheep blood (Becton Dickinson Microbiology Systems, Sparks, MD) for 24 h before testing. B. anthracis spores were adjusted to 10⁷ CFU/ml, and other bacterial suspensions were adjusted to 10⁸ CFU/ml in 0.05 M KH₂PO₄, pH 8.0. All glassware and stir bars were treated to become chlorine demand-free before sterilization. Monochloramine solutions were prepared by the addition of NH₄Cl and reagent-grade NaOCl in a 2.2:1 (N:Cl) molar ratio (equivalent to approximately 2:1 Cl₂:N weight ratio) to 500 ml 0.05 M KH₂PO₄ adjusted to pH 8. One milliliter of bacterial suspension was added to a test flask containing 99 ml of the monochloramine solutions that had been equilibrated to the appropriate temperature and monochloramine concentrations (1 or 2 mg/liter). All tests were performed in triplicate. Aliquots of 3 ml were removed from the test flasks at given times and placed immediately into tubes containing 10 μl of 5% sodium thiosulfate to neutralize the disinfectant. Serial dilutions and spread plating were performed, plates were incubated (25°C for Y. pestis, 35°C for all other organisms), and CFU were counted and checked for up to 7 days after treatment. Testing methods are further described elsewhere (5). Free available chlorine (FAC) and monochloramine, as measured by total chlorine, were monitored using the N,N-diethyl-p-phenylenediamine (DPD) colorimetric analysis method (1). FAC and monochloramine were checked before the inoculation of each solution and after the final test period for all test runs, except for the B. anthracis tests, which were checked at every sampling time. Decay curves were generated using the mean log₁₀ of the CFU counts at each sample time. The time each organism was inactivated by 99.0% or 99.9% was determined by linear regression of the appropriate segment of the decay curve. Disinfectant concentrations at the times of interest were estimated by linear regression. The Ct values for each inactivation level and test temperature were then determined by multiplying the inactivation time by the estimated monochloramine concentration.

The Ct values for inactivation of the organism tested at 2 and 3 orders of magnitude are given in Table 1. The B. anthracis spores were less susceptible to monochloramine disinfection than all other bacteria tested, with Ct values 10- to 1,000-fold greater.

TABLE 1.

Monochloramine Ct values of bacterial bioterrorism agents^a

Isolate	Temp (oC)	Ct (mg·min/liter) value for:
		2-log10 inactivation	3-log10 inactivation
Bacillus anthracis Sterne spores	5	10,532	15,164
	15	2,793	3,925
	25	1,442	1,847
Bacillus anthracis Ames spores	5	3,499	6,813
	15	1,072	1,691
	25	785	1,204
Brucella suis MO562	5	134.3	156.8
	15	99.8	120.4
	25	47.8	56.1
Brucella melitensis ATCC 23456	5	501.8	579.5
	15	204.0	223.9
	25	104.4	116.6
Burkholderia pseudomallei KC872	5	116.7	156.1
	15	87.6	103.9
	25	38.8	45.9
Burkholderia mallei M-9	5	158.6	194.1
	15	89.4	102.4
	25	52.5	64.6
Francisella tularensis LVS	5	76.0	97.9
	15	61.2	71.1
	25	26.3	30.4
Francisella tularensis NY98	5	84.0	116.0
	15	48.7	64.8
	25	31.3	37.1
Yersinia pestis A1122	5	92.0	115.6
	15	71.4	86.4
	25	27.6	33.1
Yersinia pestis Harbin	5	80.7	91.4
	15	33.5	40.8
	25	21.9	25.0

^aAll tests were performed at pH 8.

The spores of the two B. anthracis strains tested demonstrated differences in susceptibility to monochloramine. Ct values for the Sterne strain were 1.5 to 3 times greater than those of the Ames strain, depending on the temperature (Table 1).

B. pseudomallei, B. mallei, B. suis, F. tularensis, and Y. pestis demonstrated a 2-log₁₀ inactivation at Ct values of 21.9 to 52.5 at 25°C. Using similar conditions of temperature and pH, Ward et al. reported a mean Ct value of 54 mg·min/liter for Escherichia coli (17). The B. melitensis isolate was more resistant than the other gram-negative organisms, with a Ct value of 104.4 at 25°C (Table 1). If we consider that distribution systems target a residual monochloramine concentration of 2 mg/liter at the average residence time location (14) and if the Ct of Y. pestis Harbin is 25, then it would require 12.5 min to achieve a reduction of 3 log₁₀ of this organism in a distribution system if water temperature and pH were similar to these test parameters (25°C and pH 8). Brucella melitensis, however, with a Ct of 116.6, would require 58.3 min to achieve the same 3-log₁₀ level of inactivation.

The Ct values for all organisms increase with decreasing water temperature (Table 1) so that B. melitensis would require 112 min (almost 2 h) at 15°C and 290 min (almost 5 h) at 5°C to be reduced by 3 log₁₀ organisms. The Brucella strains were the only alphaproteobacterial isolates of the bacteria tested. Preliminary evidence suggests that differences in membrane lipopolysaccharide core structures may exist between Brucella spp. and other gram-negative organisms, with variations within the genus reported (6). This variation within the genus is evidenced in these data in that B. suis was not as resistant to the disinfectant as B. melitensis. B. melitensis did not, however, show a marked tolerance to FAC (12), but this can likely be explained by the mechanistic differences of inactivation for the two disinfectants.

Monochloramine is thought to react with sulfur-containing amino acids and tryptophan in the bacterial cell wall (7, 8). The mechanism of FAC as a disinfectant differs in that FAC targets many aspects of the cell structure and metabolism, such as oxidation of the cell membrane, DNA damage, and respiration inhibition (11).

Bacillus spores are known to be more resistant to disinfectants than vegetative bacteria because of their complex spore coat (13), and resistance to monochloramine is no exception (Table 1). Water with the target residual of 2 mg/liter monochloramine would reduce the viability of B. anthracis Ames spores by 3 log₁₀ in 10.0 h, 14.1 h, and 56.8 h at 25°C, 15°C, and 5°C, respectively. B. anthracis Sterne spores proved to be more resistant to inactivation, requiring 15.4 h, 32.7 h, and 126.4 h for a 3-log₁₀ reduction of viable spores at 25°C, 15°C, and 5°C, respectively (Table 2).

TABLE 2.

Hours required for 2- or 3-log₁₀ inactivation of Bacillus anthracis spores^a

Isolate	Temp (°C)	Time (h) to:
		2-log10 inactivation	3-log10 inactivation
Bacillus anthracis Sterne	5	87.8 (3.6 days)	126.4 (5.2 days)
	15	23.3	32.7
	25	12.0	15.4
Bacillus anthracis Ames	5	29.2	56.8 (2.4 days)
	15	8.9	14.1
	25	6.5	10.0

^aInactivation by 2 mg/liter monochloramine at pH 8.

The American Water Works Association found the median time to the first point of use to be 45 min for the 283 distribution systems responding to a survey (18). A second survey indicated that the median (and target) concentration was 2 mg/liter monochloramine at the average residence time in the responding distribution systems (14). We can estimate that an organism with a 3-log₁₀ Ct of 90 would be inactivated by 3 log₁₀ before the median first point of use (45 min) if introduced early in the distribution system when the monochloramine concentration is at least 2 mg/liter. F. tularensis and Y. pestis can be reduced by 3 log₁₀ within 45 min if the water temperature is 15°C or higher and the pH is maintained at 8. The Burkholderia strains tested would be reduced by 2 log₁₀ under these conditions, but the Brucella strains tested would require a longer contact time or a higher disinfectant concentration for a 2-log₁₀ reduction.

Monochloramine is a less effective disinfectant for all organisms tested when they are exposed at lower temperatures. At 5°C and 2 mg/liter monochloramine, none of the gram-negative bacteria tested can be inactivated, as defined by a 3-log₁₀ reduction, within the 45 min mean time to the first point of use. At 5°C, 46 min to 4.8 h would be required for a 3-log₁₀ inactivation, depending upon the organism.

B. anthracis spores cannot be reduced by 2 or 3 log₁₀ under these median treatment conditions (45 min and 2 mg/liter monochloramine) regardless of temperature and would require hours or days of disinfectant exposure (Table 2).

Monochloramine, though a less effective disinfectant than free chlorine, is being used increasingly as a secondary disinfectant because of the tendency to form lower levels of the disinfection by-products (DBPs) closely regulated by the SWTR. Fewer taste and odor complaints from consumers also make monochloramine use attractive. Disadvantages include problems with controlling excess ammonia to avoid nitrification and the need to control pH for better efficacy. Many treatment facilities have opted to alternate between FAC and monochloramine to control nitrification problems and biofilm formation, to boost disinfection efficacy, and to reduce DBPs (18).

Water conditions such as pH (19), turbidity (9), nutrient availability (4), and the presence of biofilms (10) can alter the efficacy of chemical disinfectants; hence, the Ct values stated in this work are valid for these laboratory test conditions only. Additionally, given the variations in susceptibility observed among strains of the same species in these studies, continued testing of a variety of strains is essential to better prepare for protecting public health if potable water is ever contaminated with any of these bacterial agents.