The Effect of Long-Term Storage on Mycobacterium bovis

For many years, tuberculosis has remained relevant in veterinary and human medicine worldwide and in Ukraine. According to WHO, tuberculosis is one of the top ten causes of death in the world. In 2017, 10,000,000 people had a tuberculosis infection, and 1,600,000 died from the disease. In 2018, according to WHO, the overall incidence of tuberculosis in Ukraine was 36,000 (per 37,000,000 population).

Complexities of overcoming tuberculosis and mycobacteriosis can be interrelated to the variability of the bacterial cells that enable pathogens to survive both in hostile macrophages (Kumar and Sanyal 2012) and outside the organism. Mycobacteria have an extraordinary capacity to adapt to the changing environment (Velayati and Farni 2012). It has been shown that nutrient deficiency, hypoxia, temperature, pH, NaCl, and various exogenous stress can significantly influence the metabolism of mycobacteria and, at the same time, the microorganism’s morphology (Vera and Rettger 1940; Smeulders et al. 1999; Bentrup and Russell 2001; Shleeva et al. 2002; Young et al. 2005; Anuchin et al. 2009; Velayati et al. 2009; 2011; Velayati and Farnia 2012).

It has been established that mycobacteria have a high degree of variability, and changes in the biological properties of Mycobacterium bovis (cultural, tinctorial, and morphology) may be accompanied by changes in their metabolism (Tkachenko et al. 2020b). The biological properties of mycobacteria and M. bovis, in particular, are far from being fully characterized. It leads to the tense situation of tuberculosis in individual countries, both among animals and humans.

Understanding the processes in a bacterial cell in response to the external factors could push the knowledge on the complex mechanism of mycobacterial adaptation and survival in macroorganisms and the environment.

There is no sufficient research on the impact of long-term storage of vaccine strains on the effectiveness of preventive immunization for infectious diseases. The BCG strain is a good example. Long-term storage of a parent strain in various countries led to its variability and the formation of BCG sub-strains: Russia (Moscow), Japan (Tokyo), Moreau (Brazil, Rio), Sweden (Gothenburg), Birkhaug, Danish (Denmark, Copenhagen), Prague (Czech Republic), Glaxo (London), China, Phipps (New York, Park, Philadelphia), Tice (Chicago), Frappier (Montreal), Pasteur (Paris), Bulgaria (Sophia), Connaught (Toronto), etc. BCG sub-strains slightly differ in genetic traits and immunogenicity. Therefore, considering potential changes in bacteria’s biological features after the long-term storage is essential for a comprehensive assessment of vaccines’ safety and efficiency (Ritz et al. 2008; Zhang et al. 2016).

Therefore, there was necessary to investigate the morphology, growth rate, tinctorial properties, biochemical activity, pathogenic, and sensitizing properties of mycobacteria in long-term conditions of hypoxia, nutrient deficiency, and low temperature (3°C).

Initial cultures on a Lövenstein-Jensen nutrient medium were characterized by a predominantly moderate growth rate in the inoculation line (Fig. 1), the color of ivory. Some strains also grew as individually formed colonies. The colony surface was rough (R) in three subculture variants (passages No. 135, 171, and 180) or smooth (S) – also in three subculture variants (passages No. 54, 115, and 143).

Fig. 1.

The mycobacteria from passage 54 appeared as the acid-fast forms (thick rod with rounded ends and single grains) (Fig. 2a). All other subculture variants (passages No. 115, 135, 145, 171, and 180) displayed mixed morphological forms (Fig. 2b), i.e., acid-fast rods, non-acid-fast filamentous and rod-shaped forms, and grains.

Fig. 2.

After inoculation of the Lövenstein-Jensen nutrient medium, the mycobacteria were capable of growing at two temperatures (3°C and 37°C) (passages No. 115, 171), at one selected temperature 37°C (passages No. 54, 143, 180), or 3°C – one strain (passage No. 135). The subcultures obtained (Fig. 3) differed from the initial cultures, as shown in Fig. 1.

Fig. 3.

At 37°C, the bacterial growth became visible on the fourth day – passages No. 115, 143, on the fifth day – passages No. 171 and 180, and on the twentieth day – the passage No. 54. The growth rate in all cultures was considerable. In four subculture variants (passages No. 54, 115, 171, and 180), the growth was characterized by an oily coating (fur) of yellow or orange colors (Fig. 3), in one subculture variant (passage No. 143) – small translucent ivory-colored S-colonies were located in clusters.

At 3°С, the growth was noted on the sixth day – passages No. 115 and 171, and on the fiftieth day – passage No. 135. In two cultures (passages No. 115, 171), there was considerable growth in the form of blue-green mucous ropy fur, and in one culture (passage No. 135) – yellow fur.

Microscopy of smears prepared from the isolated cultures revealed the difference between the parental strain of M. bovis (as shown in Fig. 2) and the organisms that were stored for many years (Fig. 4). Thus, in passage No. 54 only blue thick thread-like forms with a barely noticeable red granularity inside could be found. The morphology of mycobacteria was quite diverse: short and long, straight and curved, thin and thick rod-shaped bacteria with rounded ends, coccal and rod-shaped forms, and grains. In passages No. 171 and 180, the L-forms and other rod-shaped mycobacteria, thread-like forms, and grains were noticed. In passage No. 143 only L-forms with a different optical density were recorded. As proof of the affiliation of L-forms to mycobacteria is the destruction of vesicular L-forms and the release of acid fast-negative grains and acid fast-positive shaped bacteria (rods).

Fig. 4.

At the end of the animal experiments and euthanasia, an autopsy was performed to reveal any pathological-anatomical changes characteristic for tuberculosis in the experimental animals. The infection of guinea pigs by all studied mycobacterial subcultures (passages) was not accompanied by sensitization of the macroorganism to PPD for mammals, the formation of ulcers at the site of mycobacteria suspension, or bodyweight decrease.

Guinea pigs infected with the parental pathogenic strain of M. bovis responded to tuberculin after 30 and 60 days of the experiment, and their body weight decreased. At the point of inoculation of the mycobacterial suspension, the ulcers were formed after 27 days. The death of one guinea pig was reported on day 69. The other guinea pigs were euthanized after 90 days. The pathological changes characteristic of tuberculosis were noted in infected all guinea pigs (Fig. 5).

Fig. 5.

Through bacteriological investigations of specimens from each guinea pig, eight cultures of mycobacteria were isolated: after four days of culture, there were three cultures at 3°C (passages No. 115, 135, and 171) and five cultures at 37°C (passages No. 54, 115, 143, 171, and 180). Non-acid-fast rod-shaped mycobacteria and grains were therefore visible under the immersion in the microscope.

In animals infected with a second passage, no allergic reactions to PPD for mammals, formation of ulcers in the injection site, or tuberculosis characteristic features were observed. However, the bacteria growth from tissue specimens at 37°C was observed on the 6^th day (passages No. 54 and 180) and on the 10^th day (passages No. 171 and 143). Under microscopy, the smears from cultures revealed acid fast-negative rod-shaped bacteria and grains.

The animals infected with a third passage did not suffer allergic reactions to the PPD for mammals, the formation of ulcers in the site of inoculation of material, decrease in body weight, or and symptoms of tuberculosis. No bacteria were grown from the animal tissues.

These mycobacteria could grow on nutrient media. In the nutrient broth, colony growth was observed for all passages except for passage No. 54. In the Lövenstein-Jensen medium with sodium salicylate at 3°C, colony growth was observed for all passages except for 54, 135, and 171 (Fig. 6).

Fig. 6.

Concerning the biochemical activity of the cultures studied, variations in the activity of mycobacterial enzymes were ascertained depending on the number of passages and the cultivation temperature. The tendency to change the biochemical activity of the microorganisms studied was established: catalase activity increased along with the decrease of the cultivation temperature in passages 115 and 171 (Fig. 7).

Fig. 7.

Peroxidase activity was observed only in the control samples of the highly virulent strain. Dehydrogenase activity in the first 15–30 minutes was not detected in any culture, but after 24 hours was observed in almost all cultures, except those on the 54 passage (Fig. 8).

Fig. 8.

Nitrate reduction was recorded in all cultures except for passages No. 143 and 171 grown at 37°C (Fig. 9).

Fig. 9.

Hydrolysis of Tween-80 in the first 4 hours was not found in all subcultures (passages) of M. bovis. Passages No. 115 and 171, which were grown at a low temperature, had a higher ability to hydrolyze Tween-80 on the fifth day.

When the biochemical activity of mycobacteria at subsequent passages in guinea pig was compared with the initial strain, the following phenomena become apparent: an increase in the activity of the enzyme dehydrogenase – passages No. 54, 135 (grown at 37°C) and 171 (grown at 3°C); loss of ability to reduce nitrates, especially in cultures grown at low temperature (3°C) – passages No. 115, 171, and at 37°C – for the passage No 195; increase of catalase activity – passages No. 115 and 171 at both cultivation temperatures (3°C and 37°C); the ability to hydrolyze Tween-80 in the vast majority of cultures was decreased for all passages except passage No. 54 (Fig. 10).

Fig. 10.

It should also be noted that the biochemical activity of one subculture strain No. 143 after passage through the body of guinea pigs remains unchanged.

Using PCR, the amplicons were detected in the initial pathogenic M. bovis fast-growing strain and both Vallee and BCG strains (Table I). At the same time, the amplicons were not observed after amplification of DNA derived from the stored for long-time M. bovis fast-growing strain.

According to the experimental findings of this study, mycobacteria stored in the nutrient medium for 9–12 years manifested viability on the Lövenstein-Jensen medium. After long-term storage, it was possible to trace the non-acid-fast elements and L-forms of vesicular type in the subsequent cultures.

Other authors have also reported that the causative agent of tuberculosis of the horned presented highly adaptive properties (Djachenko et al. 2008). The transformation of mycobacteria may occur when environmental hostility leads to the change in their lifestyle (Gomez and McKinney 2004; Lewis 2007). To minimize the need for nutrients, bacteria may undergo hibernation and increase their resistance.

The phenotypic change that occurred in mycobacteria in our laboratory may stem from the decrease in the oxygen level in tubes due to its gradual use for bacterial metabolism during storage under strictly closed stoppers.

According to Kumar and Sanyal (2012), such a process of increasing resistance by reducing metabolism may be caused by the decrease in oxygen concentration in the habitat, which leads to anaerobiosis. It has appeared that the slight decrease in oxygen content in the tubes allows mycobacteria to survive. In contrast, the rapid decrease of oxygen concentration leads to the sharp decrease in the number of colony-forming units (Wayne and Lin 1982; Dick et al. 1998; Usha et al. 2002; Lewis and Falkinhan 2015).

The oxygen limitation causes some specific changes in mycobacteria. The microorganisms become thermotolerant, reduce overall protein synthesis, lose acid resistance, and their cell wall display a thickening process (Gillespie et al. 1986, Cunningham and Spreadbury 1998, Hu et al. 1998, Primm et al. 2000).

Adaptation to nutrient deficiency is also crucial for the long-term survival of mycobacterial cells. Mycobacterium tuberculosis can survive on starvation for up to two years in vitro while maintaining the ability to resuscitation on the nutrient media (Nyka 1974; Primm et al. 2000). Mycobacterium smegmatis retain the viability for more than 650 days and long-term maintenance of stable numbers of colony-forming units (Smeulders et al. 1999). After the long-term storage (10–13 years), M. bovis retained its viability in this study. Moreover, the mycobacteria could grow on a nutrient medium and Levenstein-Jensen medium with sodium salicylic acid. They could also reduce nitrates, hydrolyze Tween-80, and display an increase in catalase, peroxidase, and dehydrogenase activity.

The results of guinea pig infection in this study demonstrated that mycobacteria underwent a strong transformation during the years of storage because they became avirulent and lost their sensitizing properties. Even multiple direct passages through the laboratory animals failed to restore their pathogenicity. On the contrary, after the second passage through guinea pigs, only half of the mycobacterial subcultures were isolated from animals, and no cultures were grown after the third passage. No allergic reactions to tuberculin in any of the infected guinea pigs were observed for three passages. In our opinion, the multiple passages of mycobacterial subcultures through the nutrient medium and subsequent long-term preservation at 3°C affected their vital properties. Due to the influence of the above-named factors, mycobacteria adapted so much to the conditions of existence in vitro that the immune system overcame them when they enter the macroorganism.

Already described by many authors (Tkachenko 2004; Djachenko et al. 2009; Yavorska and Sybirna 2009; Glebenjuk and Telizhenko 2015; Lysenko et al. 2019), the high variability of mycobacteria is confirmed by our investigations of the biochemical activity. M. bovis (unlike M. tuberculosis) is traditionally considered to have no pronounced nitroreductase activity (Bönicke et al. 1970; Fritz et al. 2002). Some researchers have suggested that M. bovis BCG (vaccine strain) uses nitrates as a critical source of nutrition, supporting bacterial metabolism in the lungs, liver, and kidneys by restoring nitrate to nitrite. They argue that nitrate could provide energy for bacterial metabolism even in anaerobic environments (Philippot and Højberg 1999; Fritz et al. 2002). In addition, nitrate reductase enzyme analysis is widely used in laboratory practice as an alternative method for detecting the resistance of nitrate reductase-positive strains of mycobacteria to anti-tuberculosis medicines such as isoniazid, rifampicin, ethambutol, ofloxacin, and streptomycin. A mycobacterial strain is considered to be resistant if it has nitrate reductase activity (Angeby et al. 2002; Martin et al. 2005; Montoro et al. 2005; Lemus et al. 2006; Fonseca et al. 2012).

On this basis, we can assume that the changes that have occurred in M. bovis enzymatic activity reflect the rearrangement of metabolic reactions within the bacterial cell. Such processes took place throughout long-term storage in the same medium and aimed at providing nutrition to the microorganisms at the expense of the energy obtained by reducing nitrates.

After the passage of M. bovis through the living organism, the bacterial activity of dehydrogenase and catalase increased. These are multifunctional heme-dependent enzymes (Bertrand et al. 2004) actively involved in the antioxidant protection of the microbial cell (Kondratjuk and Sybirna 2008).

It has been reported (Li et al. 1998) that catalase promotes M. tuberculosis ability to survive in host tissues. In the work of Manca et al. (1999), the catalase and peroxidase activity of the laboratory, clinical, and recombinant strains of M. tuberculosis were investigated. These strains had minimal catalase activity both intracellularly (in human monocytes) and in the nutrient medium. Significantly more bacteria (85 %) survived when they were exposed to the influence of exogenous H₂O₂.

According to some assertions, catalase and peroxidase activity acts as a virulence factor of mycobacteria (Manca et al. 1999; Kondratjuk and Sybirna 2008).

However, our results indicate that mycobacteria’s final typing and species identification based only on biochemical tests are impossible. These characteristics may differ between members of the same species depending on the environmental conditions of the specific microorganism. (Torkko et al. 1998; Yavorska and Sybirna 2009; Lysenko et al. 2011). Thus, mycobacteria can rapidly alter metabolic processes and adapt to environmental conditions. To identify mycobacteria, researchers (Wayne and Sramek 1992; Torkko et al. 1998; Tkachenko 2004) recommend other tests, for instance, fatty acid analysis, detection of bacterial growth rate, sequence analysis, or PCR.

In the microbiological diagnosis of tuberculosis, PCR is recognized as a method of high specificity and sensitivity (Rodriguez et al. 1995; Portillo-Gómez et al. 2000). However, sequencing of the complete genome of M. tuberculosis revealed a similar frequency of mutations both during latency and during active disease or in a logarithmically growing culture. Such a mutational load on mycobacteria in vivo is assumed to be due to oxidative DNA damage (Ford et al. 2011). After long-term storage of the BCG strain, numerous genetic differences were detected in the form of large-sequence polymorphisms, including deletions and duplications, as well as single nucleotide polymorphisms among different groups of the BCG strain (Zhang et al. 2016).

Repeatedly passaged M. bovis (acid fast-positive rod-shaped bacteria, acid fast-negative thread-like, and rod-shaped forms) under long-term persistence in the unfavorable conditions, underwent permanent changes leading to growth as yellow, orange, or blue-green pigment colonies. Also, the generation of L-forms of vesicular type, acid fast-negative thread-like, rod-shaped variants, and elementary bodies pushed out of the latter were observed. Secondly, the mycobacterial metabolism changed because they were able to grow on nutrient medium and Lövenstein-Jensen medium with sodium salicylate. Their capability to reduce nitrate and hydrolyze Tween-80 was diminished. The catalase, peroxidase, and dehydrogenase activities of sub-cultured mycobacteria increased and were not dependent on the number of passages. After passages through animal tissues, the activity of the dehydrogenase and catalase enzymes increased while virulence and sensitizing activity were lost. We conclude that M. bovis was considerably changed during long-term storage in tubes under hypoxia conditions.