Coxiella burnetii is an obligate intracellular pathogen and the causative agent for Q fever in humans. Coxiella is known to be highly resistant to extreme environmental conditions including drying, pH extremes, and high temperatures which may also allow it to persist in the environment. This organism is generally found in ruminant animals including cattle, goats, and sheep, and Q fever is thought to occur primarily through aerosol exposure to contaminated excreta and birth products. Exposure via consumption of contaminated milk and dairy products may also be a source of disease but the risk is generally considered to be low (Gale et al. Reference Gale, Kelly, Mearns, Duggan and Snary2015).
Coxiella burnetii is considered to be the most heat-resistant vegetative pathogen found in raw milk. In the late 1950s and early 1960s, researchers used raw milk containing Coxiella burnetii as the reference pathogen for the establishment of proper pasteurization conditions (Enright et al. Reference Enright, Sadler and Thomas1957; Enright, Reference Enright1961). The time and temperature recommendations given were adjusted to increase the safety of pasteurized bovine milk with additional time or temperature increases for products having increased fats, solids, or flavorings. Though the experiments were well-controlled, viability of the Coxiella in the heat-treated samples needed to be determined using guinea pigs for culture. Because direct quantitation could not be accomplished from the milk, thermal inactivation curves could not be produced at each test temperature. Instead, two regression lines were plotted delineating the point below which vials were still positive and over which no survivors were found. In this case, it had to be assumed that the survival curves were linear (Cerf & Condron, Reference Cerf and Condron2006).
Since the standard pasteurization conditions were defined using Enright's data, there has been further research into how the more common milk pathogens such as Salmonella, Listeria, E. coli O157:H7, Campylobacter and other pathogens are inactivated during various pasteurization processes (Donnelly & Briggs, Reference Donnelly and Briggs1986; Bradshaw et al. Reference Bradshaw, Peller, Corwin, Barnett and Twedt1987; Doyle et al. Reference Doyle, Glass, Beery, Garcia, Pollard and Schultz1987; Bunning et al. Reference Bunning, Donnelly, Peeler, Briggs, Bradshaw, Crawford, Beliveau and Tierney1988; D'Aoust et al. Reference D'Aoust, Park, Szabo and Todd1988; Lovett et al. Reference Lovett, Wesley, Vandermaaten, Bradshaw, Francis, Crawford, Donnelly and Messer1990; Pearce et al. Reference Pearce, Smythe, Crawford, Oakley, Hathaway and Shephard2012). However, further inactivation studies using Coxiella have not been conducted due to the difficulties in working with the organism and methods needed to determine viability (Pearce et al. Reference Pearce, Smythe, Crawford, Oakley, Hathaway and Shephard2012). Although an Integrated Cell Culture-PCR (ICC-PCR) method has been developed in our laboratory to more easily quantitate viable Coxiella in heat-treated milk samples (Stewart et al. Reference Stewart, Shieh, Tortorello, Kukreja, Shazer and Schlesser2015), the method is still quite laborious and technically difficult as it relies on tissue culture for propagation of the Coxiella.
Traditionally, Coxiella are considered to be obligate intracellular pathogens which cannot be grown outside a host cell. Recently, a specialized liquid growth medium, Acidified Citrate Cysteine Medium-2 (ACCM-2), has been developed to allow propagation of Coxiella outside a host cell under modified atmosphere conditions (Omsland et al. Reference Omsland, Beare, Hill, Cockrell, Howe, Hansen, Samuel and Heinzen2011). Using ACCM-2 medium, Coxiella have been enriched from infected mouse tissues (Omsland et al. Reference Omsland, Beare, Hill, Cockrell, Howe, Hansen, Samuel and Heinzen2011), clinical samples (Boden et al. Reference Boden, Wolf, Hermann and Frangoulidis2015), and pure cultures isolated from environmental samples (Kersh et al. Reference Kersh, Priestley, Hornstra, Self, Fitzpatrick, Biggerstaff, Keinm, Pearson and Massung2016). The potential ability to enrich Coxiella directly from contaminated milk using ACCM-2 presents a new method for conducting pasteurization research in which viability of heat-treated or injured cultures could be more easily determined through direct enrichment of Coxiella in ACCM-2. The objective of this work was to determine the ability of ACCM-2 medium to allow enrichment of Coxiella from various milk types and evaluate the potential use of this medium in an MPN-PCR scheme to quantitate heat inactivation of Coxiella.
Materials and methods
Organisms and reagents
Coxiella burnetii Nine Mile Phase II RSA 439 was kindly provided by Dr Robert Heinzen at Rocky Mountain Laboratories (National Institutes of Health, Hamilton, MT). Acidified Citrate Cysteine Medium-2 (Sunrise Science Products, San Diego, CA), was prepared and pH-adjusted to 4·75 with 6 N NaOH. The 2× solution was diluted to 1 × ACCM-2 with di H2O, filter sterilized, and stored at 4 °C. The media were used in experiments within one week of preparation.
Milk products
Two bovine milk products (cream and whole milk) and three non-bovine whole milk types were selected to represent a range of source species. For the bovine products ultra-high temperature (UHT) processed whole bovine milk from Gossner Farms (Logan, UT) and HTST-processed heavy cream was purchased from a local vendor. The non-bovine whole milks included HTST-processed whole goat milk from Meyenberg, Turlock, CA; batch-pasteurized whole camel milk from Desert Farms, Santa Monica, CA; and raw water buffalo milk obtained from a Nevada farm by FDA personnel during a facility inspection. The raw water buffalo milk was treated using an isothermal heat treatment at 95 °C for 60 min before use to eliminate the background microbiota which were able to grow in the ACCM-2 medium and stop Coxiella growth even under modified atmosphere (5% CO2 and 2·5% O2) at pH 4·75.
PCR
Nucleic acid extracts for Coxiella burnetii stocks and ACCM-2 enrichment cultures were prepared by extracting 200 µl of sample using the Quickgene Mini80 system with the QuickGene DNA tissue Kit S (AutoGen, Holliston, MA) with final elution using 200 µl of the kit-supplied buffer. Extracted DNA samples were stored at −20 °C prior to PCR amplification.
The real-time PCR primers and minor groove binder (MGB) probe sequences and concentrations used for amplification of the Coxiella-specific IS1111a gene were as per Howe et al. (Reference Howe, Loveless, Norwood, Craw, Waag, England, Lowe, Courtney, Pitt and Kulesh2009). Additional details are available in the Supplemental Materials and Methods. A standard curve was produced over a 6-log range using extracted 10-fold serial dilutions of purified Coxiella burnetii with triplicate reactions giving a slope of −3·393, with an efficiency of 97·1% and R2 of 0·952.
Evaluation of ACCM-2 medium for growth of Coxiella burnetii in milk products
C. burnetii frozen stocks [~4 × 1010 genome equivalents (ge)/ml] were thawed and diluted to various levels using ACCM-2 medium prior to inoculation into milk products. For the initial evaluation of growth, milk samples were diluted 1 : 10 with 1× ACCM-2 media in vented T-25 flasks (BD, Franklin Lakes, NJ) and grown in 7 ml volumes at 37 °C under 5% CO2 and 2·5% O2 for up to 14 d. Positive controls were prepared without milk matrix and negative controls were prepared without added Coxiella. At each sampling point, 500 µl from each flask was removed and immediately frozen at −20 °C prior to DNA extraction and quantitation by PCR. Modifications tested for improved growth in the water buffalo and whole milk are detailed in the Supplemental Materials and Methods.
Demonstration of thermal inactivation of Coxiella burnetii in milk products by MPN-PCR using ACCM-2 medium
Whole bovine milk (1·5 ml) containing ~7·2 log ge/ml Coxiella was aliquoted into 2 ml crimp-top glass vials (Fisher Scientific, Pittsburgh, PA), sealed, and placed into an ice bath for 5 min to cool. Vials were secured in a vial holder and submerged in a shaking 64 °C water bath (Thermo Scientific, West Palm Beach, FL) for heat treatment. Vials were removed at different intervals (2, 4, 6, 8, 9, and 10 min) to an ice bath to stop thermal inactivation prior to the MPN-PCR assay. Unheated vials were used as the 0 min treatment samples. Thermally treated milk samples were serially diluted in ACCM-2 and 1 mL of each dilution was transferred into triplicate T-25 flasks each containing 9 ml ACCM-2 media. Negative controls were prepared by diluting 1 ml milk with 9 ml ACCM-2. All flasks were incubated for 10 d at 37 °C under 5% CO2 and 2·5% O2 and sampled as previously described. Quantitation of the remaining viable Coxiella in each heat-treated sample was determined by Most Probable Number (MPN) method using the number of flasks at each dilution level showing at least a 0·5 log increase over the 10-d enrichment period. Coxiella MPN/ml milk were calculated using an online freely available MPN Excel spreadsheet created by the U.S. Food and Drug Administration (Blodgett, 2010).
Results and discussion
Enrichment of Coxiella burnetii from various milk products using ACCM-2 liquid media
ACCM-2 liquid medium was used to enrich Coxiella burnetii from a variety of milk products, including bovine cream and whole milk and camel, goat and water buffalo whole milk inoculated at ~4 × 102 ge/ml. Figure 1 shows the growth of Coxiella from these samples after 14 d of enrichment. Coxiella levels in all samples showed a slight initial decrease during the first 2 d which may have been due to DNA degradation from non-viable cells or extraneous free DNA. Growth of Coxiella in the ACCM-2 medium control showed a rapid increase to 7 log ge/ml by day 6. Thereafter, growth slowed and plateaued at a final level of ~log 8·7 ge/ml. Growth from camel milk very closely mirrored that of Coxiella in the ACCM-2 media control. Growth curves for bovine cream and whole goat milk were similar. Growth from bovine whole milk showed an extended lag phase (~6 d) with a log phase growth rate of 0·77 log ge/d which is similar to the goat and bovine cream growth rates of 0·68 and 0·84 log ge/d, respectively. A maximum level of ~6·8 log ge/ml was attained by day 14. Coxiella growth from whole buffalo milk was not present after 14 d. Experiments to improve or allow growth of Coxiella in bovine and water buffalo milks using pre-enrichment dilution or pH adjustment were also undertaken and the results are shown in Supplemental Data Figs. 1 and 2. Adjustment of the initial pH of the milk/media mixture from ~4·90 to ~4·75 (the optimal pH for Coxiella growth) allowed improved growth from bovine whole milk but not water buffalo milk. Severe dilution (1 : 1000) of the water buffalo milk with ACCM-2 media was required to allow even moderate (1 log ge/ml) growth of the Coxiella in the first 7 d of enrichment. Due to the need for this extra dilution, it is believed that some inhibitory substance was created in the water buffalo milk during isothermal heat treatment. Unfortunately, no sources for commercially pasteurized water buffalo milk were available so no further investigation was possible. Experiments to determine the initial concentration of Coxiella needed to show growth in bovine milk were also completed. An initial level of 6 ge/ml bovine milk allowed a 2 log increase in Coxiella ge/ml after 10 d (Supplemental Fig. 3).
Fig. 1. Growth of Coxiella burnetii in ACCM-2 media from inoculated milk products and in media alone. Coxiella genome equivalents (ge)/ml were measured by qPCR. Data points represent the average and standard deviations of 6 replicate flasks from two trials.
Quantitation of Coxiella burnetii heat inactivation in whole bovine milk
The potential use of ACCM-2 for enrichment and quantitation of viable Coxiella present in milk after heat inactivation was evaluated using vial-heated log 7·2 Coxiella ge/ml in UHT whole bovine milk treated at 64 °C for up to 10 min. Serial dilutions of the heat-treated milk were enriched in a 3-tube MPN format in ACCM-2 media with PCR quantitation at day 0 and day 10. A ≥0·5 log ge/ml increase in Coxiella over the 10 d enrichment period was defined as positive for propagation and the samples were scored as viable. Table 1 details the MPN results of the 64 °C thermal inactivation trial. The 0- and 2-min. samples showed little inactivation, with calculated MPN/ml milk near the upper assay detection limit of 6·7 log MPN/ml. Samples heated for 4 min showed decreased viability with a calculated level of log 4·55 Coxiella MPN/ml milk. Viable levels of Coxiella in the milk were further reduced to log 2·12 MPN/ml after 6 min. and fell below detectable levels after 8 min. Inactivation was linear after the first 2 min of heating with an R 2 = 0·997 and a linear regression equation of y = −1·115x + 8·8767 for the data points from 2–6 min, indicating a D-value of 53·8 s; however a linear regression using all data points would have a greater D-value due to the lack of inactivation seen in the first 2 min of heating. In this case, the correlation coefficient was reduced to 0·896 with a linear regression equation of y = −0·7885x + 7·353, correlating to a D-value of 76 s. Using the D-value of 76 s, 5 × 104 Coxiella infective doses (as used by Enright et al. Reference Enright, Sadler and Thomas1957) would be inactivated after 5·95 min. This value is less than would be expected based on the regression lines created by Enright which show that 105 Coxiella infective doses in 2 ml milk still contained surviving Coxiella after ~8 min at 64 °C. Based on calculations by Gale et al. (Reference Gale, Kelly, Mearns, Duggan and Snary2015), it is estimated that there are between 2 and 112 C. burnetii organisms per guinea-pig infectious dose; therefore, for 5 × 104 infective doses/ml, the number of Coxiella used in Enright's experiments as estimated by PCR would be between 1·0 × 105 and 5·6 × 106 ge/ml. The inoculation level used in our trials is ~4 × 107 ge/ml milk; therefore, it is plausible that the numbers of Coxiella being inactivated are similar to what was used in Enright's experiments. It is possible the that MPN method is overestimating inactivation since enrichment in ACCM-2 medium is influenced by the presence of the milk, the injury imposed on the cells, and their ability to recover and grow to the assay cutoff with at least 0·5 log over the enrichment period. In addition, this inactivation trial was completed without adjustment of the pH to 4·75, which may have influenced detection. In our initial evaluation, 5 log ge/ml growth of Coxiella was possible from bovine milk that was not heat-treated, however it is not known if enrichment would also be improved with pH reduction of the samples after heat treatment.
Table 1. Inactivation of Coxiella burnetii in whole bovine milk at 64 °C at various treatment times
ND, not determined
Initial inoculation level ~ log 7·2 ge/ml
† Number positive flasks/total number test flasks
Previous work using the ICC-PCR method for quantitation of viable Coxiella in milk after heat-treatment at 60 °C also showed non-linear inactivation with no quantifiable reduction in viable cells prior to 20 min of heating followed by a 3-log reduction during the next 20 min period (Stewart et al. Reference Stewart, Shieh, Tortorello, Kukreja, Shazer and Schlesser2015). After an additional 20 min of heating the level of Coxiella fell below the detection limit. Per Enright's regression lines (Enright et al. Reference Enright, Sadler and Thomas1957), Coxiella survival might still be expected at ~62 min whereas no survival would be expected after ~80 min when treating 2 ml of milk containing 5 × 104 Coxiella infective doses/ml. In the ICC-PCR heat-treatment, cells were not found viable above the lower detection limit of 1·15 log ge/ after treatment of ~7·2 Coxiella ge/ml for 60 min at 60 °C, but over 4 log ge/ml remained viable after 40 min (Stewart et al. Reference Stewart, Shieh, Tortorello, Kukreja, Shazer and Schlesser2015).
Both the ICC-PCR and MPN-PCR assays offer several advantages over the animal model used by Enright et al. (Reference Enright, Sadler and Thomas1957). The ICC-PCR and MPN-PCR methods both allow enrichment of the Coxiella directly from the heat-treated milk without the need for primary or secondary infection of animals. In addition, PCR is used as a sensitive way to evaluate viability in both methods, though the ICC-PCR method is more difficult to accomplish due to the need to extract the nucleic acids from the tissue culture. The MPN-PCR method is more streamlined in that only preparation of the media is needed prior to analysis of the inactivation. Another advantage of the MPN-PCR method over the ICC-PCR method is that, if needed, growth of the Coxiella can be monitored over time simply by quantitating the Coxiella in a small aliquot of ACCM-2 enrichment over a number of days. In this way, viability could be confirmed before the 10-d enrichment is complete.
To our knowledge, this is the first instance of enrichment of Coxiella from a food using ACCM-2 medium and the first use of the medium to quantitate viable Coxiella via an MPN-PCR scheme. The medium has been shown to allow growth of Coxiella from the milks of several animal species without the need for adaptation or adjustment of the method. The number of viable cells needed for enrichment of Coxiella from bovine whole milk using the standard 1 : 10 dilution without pH adjustment was ~6 Coxiella ge/ml milk. Further, we used enrichment in ACCM-2 to prove viability of Coxiella in heat-treated milk in an MPN-PCR format to quantitate an increase in ge/ml over a 10 d enrichment period and ultimately characterize heat inactivation of Coxiella in milk.
Acknowledgements
This study was supported in part by grant number 5U01FD003801 from the U. S. Food and Drug Administration to the Institute for Food Safety and Health of the Illinois Institute of Technology. The authors thank Dr Mary Lou Tortorello for critical review of this manuscript and Peien Wang, Xiaoxia Li, and Chen Chen for their technical assistance.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022029918000699
