Assessment of gross lesions preservation methods for the direct diagnosis of bovine tuberculosis: a constant search for improvement

The present study assessed the sensitivity and biosafety of the FTA Elute Card ® (FTA) (GE Healthcare) as an alternative preservation method of gross lesions for bovine tuberculosis diagnosis, that can endure large distances and high environmental temperatures, which are found in the Brazilian territory. The FTA card was compared to the freezing and sodium borate solution methods to preserve 134 gross lesions samples from slaughterhouses located in three Brazilian states. The calculated value for the sensitivity of sodium borate solution (SBS) stored for 30 days was higher than those estimated for the methods of freezing and SBS stored for 60 days, when the PCR and isolation results were interpreted in parallel (0.55). The FTA cards presented the lowest diagnostic sensitivity and did not inactivate all mycobacteria. Given the rush of a slaughterhouse routine, the quantity of lesions with different pathological stages cannot be measured on each FTA card. Thus, as a preservation method of gross tuberculosis lesions, the FTA cards can present infection risks for people who have direct or indirect contact with them.


Introduction
Bovine tuberculosis (bTB) is an important zoonosis and generates losses in the production of meat and milk. The Brazilian cattle presents a low prevalence, which varies from 0.36-7.6% for infected herds and 0.035-0.7% for infected animals, regardless of the differences of production purpose, animal density and number of properties among states [1].
Given the bacteriological con rmation of bTB by direct diagnostic methods is crucial [2,3], the preservation of the gross lesions during transportation from the slaughterhouses to the laboratories plays an important role, avoiding the growth of contaminants and preserving mycobacteria.
Chilling, freezing and immersion in saturated sodium borate solution had been studied [2,3,4], and are highly recommended by OIE. Although these methods' main goal is to maintain viable mycobacteria for culture, direct diagnosis by molecular techniques represents a fast and sensitive option that can overcome the need of living mycobacteria and the microbial contamination issues.
The FTA Elute Card ® (FTA) was developed for puri cation, preservation, room temperature collection, and transportation of nucleic acids from biological samples, as the chemical formula on the cards promotes cell lysis and proteins denaturation [5]. Despite the cards showed partial lysis for some gram-positive species and Mycobacterium species, inactivation was observed for cell suspensions with densities below ≤ 10 4 cfu ml − 1 [6]. Furthermore, the cards presented good e cacy to store and to transport samples for the direct detection of Mycobacterium tuberculosis from sputum specimens [7], and to extract DNA of M. bovis from bovine tissues [8].
The present study evaluated the sensitivity and biosafety of the FTA cards as a tool for the preservation of gross lesions collected at the slaughterhouses for the direct diagnosis of bTB.

Samples processing
Along seven months, 134 gross lesions were detected and sampled, during the routine inspection of bovine carcasses at slaughterhouses from the states of Santa Catarina (South region), Mato Grosso (Midwestern region) and Paraiba (Northwestern region).
Each lesion was divided and preserved by freezing, immersion in sodium borate solution (134 g L -1 ) for 30 (B30) and for 60 days (B60) and imprinting on FTA Elute Card ® (GE Healthcare) (FTA). At the division, several lesions showed various degrees of calci cation, which were observed by cutting them with sterile surgical scissors. Twenty-three lesions were too small to be divided into freezing, B30 and B60, then the samples tested for B30 and B60 were fewer than the frozen ones.
For culture, all frozen, B30 and B60 samples were decontaminated with 0.75% 1-hexadecylpyridinium chloride [9,10], while a piece of 6 mm in diameter of the FTA card was removed using a skin biopsy punch and suspended in 1 ml of sterile saline, followed by inoculation into Stonebrink and Löwenstein-Jensen media and incubation at 37 °C.
Isolates with morphological aspects compatible with mycobacteria were suspended in 400 µL TE buffer (10 mM Tris, 0.1 mM EDTA pH 8.0) and boiled during 5 min to expose the DNA. For the clinical specimens, 30 milligrams of each frozen, B30 and B60 lesion sample and a piece of 6 mm in diameter of each FTA were submitted to a modi ed DNA extraction protocol [11], which included a pre-digestion with 20 µL of proteinase K (20 mg mL -1 ) and 100 µL of lysozyme (10 mg mL -1 ).

Statistical analysis
The results of MTC detected by isolation and PCR from frozen samples were considered as gold standard. The sensitivity and speci city for each preservation method were calculated based on the results of isolation and PCR in parallel. Associations between preservation methods and the gold standard were performed by Chi square test. Analogously, Cohen's Kappa index was used to calculate the agreement [13]. All the analysis were performed in R software [14] through the epiR and fmsb packages [15,16].

Results
The number of MTC isolates was much higher (50/134) in the frozen samples than in the other preservation methods (Table 1). Given B60 and FTA card presented one isolate each (Table 1)  The results for "PCR" and "isolation and PCR in parallel" showed no statistical differences among the sensitivities of the preservation methods (Table 1). However, the calculated value for the B30 sensitivity was higher than those estimated for the other preservation methods when the PCR and isolation results were interpreted in parallel (Table 1). For this study we found kappa values below 0.60, which indicates inadequate agreement (fair to moderate agreement) among the diagnostic protocols. The speci cities calculated were always high and statistically equal, but they are not able to represent the test itself because the samples suffered mycobacterial loss, instead of being negative for culture.

Discussion
According to OIE, con rmation of Mycobacterium bovis infection requires traditional mycobacterial culture. On the other hand, molecular techniques have gained their place among the routine diagnostic tests applied for bovine tuberculosis (bTB).
The FTA cards are expected to provide a safe method for storage and transportation, including mailing, by killing all mycobacteria [6,7]. Moreover, they would maintain a high-quality DNA for the direct detection of MTC species [7,8].
The present study observed one isolate of MTC from the FTA card, and showed that this preservation method may offer risks, for it does not guarantee the innocuousness of the sample. Regarding the PCR, the detection of MTC is not dependent on the mycobacteria viability, and whereas the results were similar for the frozen and sodium borate preserved samples, the FTA cards displayed a lower MTC detection.
According to Moura et al. [8], the FTA Elute Micro Card was shown as an optimistic DNA extraction method for its convenience and simplicity. However, the material impregnated on the FTA cards was tissue previously submitted to enzymatic digestion [8], which could have increased the available mycobacteria from the tissue compared to an imprinting from the lesion.
The MTC survival on the FTA card and the low PCR detection might be due to the excessive cell tissue and/or mycobacteria in the lesion sample [6]. Nevertheless, measuring the amount of content, from lesions with different pathological stages, on the FTA cards would be challenging in the rush of a slaughterhouse routine.
Regarding the numbers of MTC isolates and of MTC detected by PCR, the freezing method showed better results for culture than molecular techniques, though it is expected to be otherwise or at least the same, considering the viable and unviable mycobacteria available in each sample. However, calcium ions are known to present inhibitory effects on the PCR, especially at high concentrations [17], and several gross tuberculosis lesions were in advanced pathological stages showing various degrees of calci cation.
The proportions of MTC isolates for B30 and for B60 compared to the gold standard are 32% and 2%, and they are quite lower than 82.6% and 26.1%, respectively, found by Morato et al. [4]. Minerals containing calcium carbonate can remove boron from water supplies by adsorption and coprecipitation processes, accordingly to pH and boron concentration [18]. Although the concentration of sodium borate solution was the same in both studies, the caseous lesions observed by Morato et al. [4] showed no calci cation, differently from the present study.
It is di cult to determine if there was a chemical reaction between the sodium borate and the lesions calci cation, and, if any, how it could have affected the lesion samples preservation and the PCR. Even so, a decrease of boron concentration may have diminished the preservation quality of the samples for culture, while a decrease of calcium ions would have improved the PCR detection.

Limitations
Although the FTA cards offer many advantages as a preservation method, they did not match up to the expectations to preserve gross lesions from slaughterhouses for bovine tuberculosis. They showed an exceptionally low diagnostic sensitivity and can present risks for people who may have contact with them. Furthermore, our ndings show that different pathological stages of gross lesions may affect the performances of sodium borate solution preservation properties and the e ciency of molecular diagnosis. Additional studies are necessary to con rm these observations.

Declarations
Ethics approval and consent to participate The present study was approved by the Ethic Committee on Animal Use (CEUA/FMVZ) under the protocol number 2580031016.