Bibliographic search and methodological considerations
The experiments were designed based on the data collected through a review of the scientific literature on strategies for C. fetus isolation from biological samples. The literature selection was based on the follow criteria: i) the abstract of the article or chapter’s book included the use of transport, enrichment and/or culture media for the isolation of Cfv from field samples (prioritizing publications that included a comparison of different media and culture techniques) and ii) protocols were used to recover Cfv from the reproductive tract of cattle. We excluded works published before 1973 (Veron, 1973) due to the possible misidentification of C. fetus subsp. venerealis.
According to the literature: i) the transport media is used to preserve the viability of Cfv from sampling to processing at the laboratory (Chaban et al., 2013; Lander, 1990a; Monke et al., 2002), ii) enrichment media is employed to improve the growth of Cfv in presence of other potential microorganisms on the samples (Lander, 1990a), and iii) culture plate agar media is used to recover and isolate Cfv colony-forming units (CFU) (Monke et al., 2002). Besides, additional methodologies to decrease the number of non-Campylobacter microorganisms isolated have been also reported in the literature, and were considered in this study: i) the use of passive filtration prior to the inoculation of plate agar media (Chaban et al., 2013), and ii) the incubation of plates at 42ºC instead of 37ºC (WHOA, 2021). Also, according to previous reports and in order to avoid the rupture of Cfv cells, the use of Digralsky loop technique during culture in plates was limited to 5 seconds (Hedderich et al., 2011).
From sampling to laboratory (experiment 1): evaluation of transport media, temperature and time of conservation for C. fetus
A total of four transport media were compared (Table 1): i) Weybridge (De Lisled, 1982), where the 5-fluorouracil and cycloheximide were replaced by amphotericin B as previously suggested (Koya, 2016; Martin et al., 2002; Murinda et al., 2006), ii) Lander (Lander, 1990b), where the 5- fluoruracil was replaced by amphotericin B as previously suggested (Koya, 2016; Martin et al., 2002; Murinda et al., 2006), iii) Thomann (Harwood et al., 2009), and iv) commercial Stuart transport medium modified (Thermo Fisher Scientific, UK). Phosphate buffered saline (PBS) was used as a control media to assess the viability of Cfv without other compounds as antibiotics and nutrients.
Table 1
Composition of each transport media in a final volume of 250 mL. All compounds were dissolved in distilled water. Is indicated from left to right: Weybridge (We), Lander (La), Thomann (Th), Stuart (St) and PBS (P) which final pH must be 7.4.
Compound | We | La | Th | St | P | Reference (commercial brand) |
Activated carbon powder (g) | 1.25 | 1.25 | | | | 05105A (Sigma) |
Veal infusion broth (g) | 6.25 | | | | | 11748812 (Fisher) |
Laked blood horse (mL) | 17,5 | 17.5 | | | | SR 0048 (Oxoid) |
Campylobacter selective supplement | 1 vial | 1 vial | ⅜ vial* | | | SR 0069 (Oxoid) |
Amphotericin B** (mL) | 10 | 10 | 10 | | | A-2942 (Sigma) |
Mueller-Hinton broth (g) | | 5.25 | | | | 0757-17-6 (DIFCO) |
Campylobacter growth suplement | | ½ vial | ½ vial | | | SR 00232 (Oxoid) |
NaCl (g) | | | | | 2 | 131659 (Panreac) |
KCl (g) | | | | | 0.05 | 131494 (Panreac) |
Na2HPO4 (g) | | | | | 0.36 | 131679 (Panreac) |
K2HPO4 (g) | | | | | 0.061 | 131509 (Panreac) |
Stuart (modified) (g) | | | | 4 | | CM0111 (Oxoid) |
Nutrient broth Nº2 (g) | | | 6.25 | | | CM 0067 (Oxoid) |
* Using 3/8 of a vial is impractical, therefore we recommend to produce at least 333.3 mL of media in order to use ½ vial. |
** In Weybridge medium, the 5- fluorouracil and cycloheximide were replaced by amphotericin B, while in Lander medium the 5- fluoruracil was replaced by amphotericin B. Those modifications were based on Martin et al. (2002), Murinda et al. (2006) and Koya (2016) (Koya, 2016; Martin et al., 2002; Murinda et al., 2006) works, due to 5- fluorouracil and cycloheximide seems to be toxic for C. fetus. |
Finally, four combinations of temperature and times for a period of 24h were evaluated for each transport media as follows (Fig. 1A): i) 24h at room temperature (RT, 21ºC ± 2ºC), ii) 24h at refrigeration (RF, 4ºC), iii) 12h at RT followed by 12h at RF, and iv) 12h at RF followed by 12h at RT. In addition, the same combination of temperatures was tested during 48h (see Fig. 1A) instead 24h.
For this study, a suspension of Cfv in 4 mL of PBS was prepared with an approximated concentration of 105 UFC/mL, a Cfv concentration commonly found in preputial bull samples (Indjein, 2013). Then, 300 µL of the suspension were inoculated in 9.7 mL of each transport media to achieve a final concentration of approximately 3 x 103 UFC/mL.
The Cfv strain used in the study (strain ICM18/00036), provided by Zootecnia S.L. laboratories, was recovered from a preputial wash bull sample and identified by MALDI-TOF mass spectrometry as C. fetus and by PCR as Cfv according to the PCR protocols published in a previous work (Polo et al., 2021). After each incubation protocol (Fig. 1A), 100µL of each transport media were inoculated on Skirrow plates (Monke et al., 2002; Skirrow, 1977) (Table 2) in duplicates. The Skirrow plates were incubated in boxes of 2.5L (Thermofisher, UK) at 37ºC during 4 days in microaerophilic atmosphere with the use of CampyGen 2.5L sachets (Thermo Fisher, UK).
Table 2
Composition of each enrichment and culture media in a final volume of 250mL. All compounds were dissolved in distilled water. The enrichment media were dispensed in tubes (12mL) while the culture mediums were plating. Is indicated from left to right: Preston enrichment medium (Pr), Brucella enrichment medium (Br), Bolton enrichment medium (Bo), Skirrow plate culture medium (Sk), Preston plate culture medium (Pr-ag) and Blood Sheep agar culture medium (Bl).
Compound | Pr | Br* | Bo | Sk | Pr-ag | Bl | Reference (commercial brand) |
Nutrient broth NO.2 (g) | 6.25 | | | | | ** | CM0067 (Oxoid) |
Blood agar base (g) | | | | 9.75 | | ** | CM0331B (Oxoid) |
Campylobacter agar base (g) | | | | | 9.25 | ** | CM0689 (Oxoid) |
Dehidrated Brucella broth (g) | | 7 | | | | ** | R452662 (Remel) |
Bacteriologic agar (g) | | 0.4 | | | | ** | A01718 (Condalab) |
Bolton broth (g) | | | 6.9 | | | ** | CM0983 (Oxoid) |
Laked blood horse (mL) | 12.5 | | 12.5 | 12.5 | 12.5 | ** | SR 0048 (Oxoid) |
Campylobacter selective supplement | | | | ½ vial | | ** | SR 0069 (Oxoid) |
Preston campylobacter selective supplement | ½ vial | | | | ½ vial | ** | SR 0204 (Oxoid) |
Campylobacter growth supplement | ½ vial | | | | ½ vial | ** | SR 00232 (Oxoid) |
Bolton broth selective supplement | | | ½ vial | | | ** | SR0183 (Oxoid) |
Bacitracin (g) | | 0.05 | | | | ** | 11702-5g (Sigma) |
Novobiocin (g) | | 0.0013 | | | | ** | N1628 (Sigma) |
Anphotericin B (mL) | | 10 | | | | ** | A-2942 (Sigma) |
* In the Brucella enrichment medium (Br), cycloheximide was replaced by anphotericin B based on Martin et al. (2002), Murinda et al. (2006) and Koya (2016) (Koya, 2016; Martin et al., 2002; Murinda et al., 2006) works, due to cycloheximide seems to be toxic for C. fetus. Polymyxin B was also removed due to seems to inhibit the growth of C. fetus subsp. fetus (Public Health Laboratory Network, 2000) and some strains of C. fetus subsp. venerealis (Ballabene, 1992). **Commercial medium (Biomérieux, 2022). |
In the case of Stuart medium, which is semisolid, both the transport media and Skirrow were inoculated using a cotton swab. This was done since it was observed in a trial that the cotton swab could contain approximately the same volume used to inoculate liquid transport media (100 µL). The trial consisted in placing one cotton swab in PBS and then measuring the volume of PBS recovered after centrifuging the cotton end of the swab at 1,200g. Therefore, a cotton swab was immersed in the bacterial suspension for 2 minutes and then the swab was introduced into Stuart tubes. After 24h, the swab was dragged on a Skirrow plate (this is the only one case in which the culture in plates was not duplicated).
This experiment was performed eight times over three days so that 16 readings were available for each TEM and temperature-time combination (except for those involving the Stuart medium). The CFU counts obtained in the Skirrow plates for each transport media and temperature-time combinations were compared through a Poisson model including the transport media (Weybridge, Lander, Thomann, Stuart and PBS as the reference) and the temperature-time combination (24h RT, 24h RF, 12h RT + 12h RF and 12h RF + 12h RT; 48 hour periods were not considered since very limited growth was obtained, see results) as the explanatory variables.
Processing at laboratory (experiment 2): evaluation of enrichment and culture media for C. fetus.
Three enrichment media were tested (Table 2, Fig. 1B): i) Preston (Bolton and Robertson, 1982), ii) Bolton (Hunt, 2021), and iii) Brucella (Marcellino et al., 2015), where cycloheximide was replaced by amphotericin B as previously suggested (Koya, 2016; Martin et al., 2002; Murinda et al., 2006) and Polymyxin B was removed due to its potential inhibition of the growth of some strains of Cfv (Ballabene, 1992). PBS was used as negative control to assess the growth of Cfv in a saline medium.
The transport medium selected based on the results of the previous analysis (see results) was inoculated with the same Cfv suspension used in experiment 1 in order to achieve a final concentration in transport medium of approximately 3 x 103 CFU/mL in a total volume of 10 mL. Then, 120 µL from this spiked transport medium were inoculated in each enrichment media (final volume = 12 mL, in tubes of 12.5 mL to ensure microaerophilic conditions) and incubated at 37ºC during 48h (Fig. 1B). After incubation, 100 µL of each enrichment media were cultured in duplicate in three different culture media (Table 2, Fig. 1B): i) Skirrow (Chaban et al., 2013; Monke et al., 2002; Skirrow, 1977), ii) Preston agar (Bolton and Robertson, 1982) and iii) commercial sheep blood agar (SBA) (Biomérieux, FR). The incubation was carried out at 37 ºC during 4 days in microaerophilic atmosphere as above mentioned. The experiment was repeated in three different days yielding six observations for each enrichment-culture media combination. Results per each combination were compared using a Poisson model as explained before with the enrichment-culture media as covariates in order to identify the combination yielding the higher C. fetus counts for its consideration in the following experiment (experiment 3) to assess the diagnostic performance of the selected protocol (transport + enrichment media) using spiked bull preputial samples as indicated below.
Diagnostic performance of selected protocol (experiment 3): spiked bull preputial samples assessing the impact of temperature of incubation and passive filtration.
A total of 32 samples consisting of preputial washings in PBS from 32 breeding bulls raised under extensive management conditions were collected from different Spanish farms. Samples were confirmed as negative to C. fetus using a PCR designed by Genetic PCR Solutions (Orihuela, Spain). A total of 4.5 mL of each preputial washings were spiked with 500 uL of a solution containing approximately 105 CFU/mL of the same Cfv strain used in previous experiments. A first group of samples (Group 1, n = 12) was used to assess the recovery rate of Cfv in the presence of possible non-Cfv microorganisms (contaminants) in the sample considering different processing conditions (Fig. 1C): i) culture temperature of 37 ºC vs 42 ºC (most Cfv strains are able to grow at 42 ºC (WHOA, 2021)), ii) apply or not a passive filtration step previous to the inoculation of the culture media [no filters vs. filters with pore size of 0.65 µm or 0.45µm according to Chaban et al. (2013) (Chaban et al., 2013)] and iii) use of solid culture media [non selective (SBA) or selective (Skirrow, 1977)]. The best protocol based on the combination of temperature, filter usage and culture media was selected based on a logistic regression model that included these three variables as covariates and the isolation of Cfv (yes/no) as the outcome. In case growth of non-specific bacteria (other than Cfv) was observed in the whole surface of the culture plates, the plate was classified as negative, while in case Cfv was observed in the presence of other non-specific colonies the plate was considered positive.
Based on the results from the first group of samples, a second group (Group 2, n = 20) was analysed to further discriminate between the performance of the following processing conditions for recovering Cfv from field samples after the storage of inoculated TEM for 24 h at 21 ± 2 ºC: 37 ºC vs. 42 ºC, use or not of passive filtration (using filters with a pore size of 0.65 µm) and two selective culture media (Preston vs. Skirrow) (Fig. 1D).