Inhibition of Morganella morganii growth and histamine production using a bacteriophage cocktail

Abstract

Histamine poisoning is a foodborne illness related to the consumption of histamine-accumulated foods. Morganella morganii is one of the major histamine-producing bacteria that contaminate fresh fish. In this study, we isolated and characterized five bacteriophages that infect M. morganii. The host ranges and adsorption patterns of the isolated phages suggest that they have different host recognition mechanisms. In this study, a M. morganii phage cocktail with a combination of seven phages was prepared. Compared to infection with individual phages, treatment with the phage cocktail delayed the growth of M. morganii for a longer duration (with two phages being the exception). The growth of M. morganii and histamine accumulation in the canned tuna flake were inhibited by the phage cocktail. These results suggest that the M.morganii phage cocktail has the potential to serve as a good biocontrol agent with respect to theprevention of M. morganii–associated histamine poisoning.

Introduction

Seafood is one of the most important protein sources; however, there are many food poisoning incidents associated with seafood due to its susceptibility to decomposition. Histamine poisoning is one of the major types of food poisoning related to seafood. This poisoning is caused by the ingestion of histamine, which is produced by the decarboxylation of histidine by histamine-producing bacteria (Yamaki and Yamazaki, 2019). Therefore, the major seafoods associated with histamine poisoning are histidine-rich fish, such as tuna, bonito, sardine, and mackerel (Becker et al., 2001; Feng et al., 2016; Kanki et al., 2004).

Histamine-producing bacteria related to fish meat are mainly gram-negative bacteria, such as Morganella morganii, M.psychrotolerans, Raoultella planticola, Photobacteriumphosphoreum, and P. damselae, which produce histamine under various conditions and in fish meat by the action of histidine decarboxylases (Bjornsdottir-Butler et al., 2020; Kanki et al., 2007; Wang et al., 2020). Among these bacteria, M. morganii is a mesophilic histamine producer that is frequently isolated from fish. Torido et al. (2014) isolated 96 histamine-producing bacteria from histamine-accumulated fish homogenates incubated at 30 °C for 24 h and reported that 24.0% of the isolated histamine-producing strains were M. morganii. Of the isolates tested, M. morganii was the second most frequently encountered bacterium, after P. damselae damselae, which was the most frequently isolated bacterium, occurring in 42.7% of the isolates. Among the isolates, 22 of 23 M. morganii isolates produced more than 1 000 mg/kg histamine, while only 24 of 41 P. damselae subsp. damselae isolates showed histamine production of more than 1 000 mg/kg. These findings indicated that M. morganii is a strong histamine producer. A previous study on the prevalence of histamine-producing bacteria in fish from the Gulf of Mexico showed that the detection rate of M. morganii was 14% (Bjornsdottir-Butler et al., 2015). In addition, M. morganii was detected on a conveyer belt and plastic tote in a fish processing plant during processing; however, the bacterium was not detected before processing (Kim et al., 2003), indicating that M.morganii in fish may also cross-contaminate other histidine-rich seafood in processing environments. These reports indicate that the control of M. morganii is important for improving the safety of seafood.

Recently, bacteriophages (phages) have been demonstrated to be well suited for use as antimicrobial agents for reducing the growth of foodborne pathogens (Sulakvelidze, 2013). Phages are bacterial viruses that specifically infect and lyse bacterial cells. To reduce the risk of histamine poisoning in seafood, we isolated and investigated phages infecting histamine-producing bacteria, especially M. morganii and P. damselae (Yamaki et al, 2014; Yamaki et al., 2015; Yamaki et al., 2020). A previous study showed that the M. morganii phage FSP1 inhibits the growth of M. morganii and reduces histamine accumulation in fresh tuna (Yamaki et al., 2018). Moreover, a phage mixture, i.e., a combination of M. morganii phages (ΦMV-1 and ΦMV-4) suppressed bacterial growth more effectively than a single phage and decreased histamine accumulation in tuna (Yamaki et al., 2020). The use of a phage cocktail is an efficient way to improve the limited spectrum of single phages and to suppress the development of phage-resistant mutants (Lu and Koeris, 2011). In this study, we isolated some M. morganii phages and evaluated the inhibitory effect of a M. morganii phage cocktail on the growth of M. morganii in liquid media and tuna flakes.

Materials and Methods

Bacterial strains    The bacterial strains used in this study are listed in Table 1. Each bacterial strain was incubated at 30 °C for 18 h in tryptic soy broth (TSB; BD, Franklin Lakes, NJ, USA) and used for further experiments.

Table 1. Bacterial strains used in this study and host ranges of phages determined by EOP test.

Bacterial species	Strain	Phage
		FSP1	ΦMV-4	Momo2	Momo8	Momo10	Momo13	Momo22
Morganella morganii subsp. morganii	NBRC 3848T	+++*	+++	−	−	−	−	−
Morganella morganii	MFS 1801	+	+/−	+	−	+/−	++	+
Morganella morganii	MFS 1802	+	+	+	−	+	++	++
Morganella morganii	MFS 1803	+	−	+	−	+	++	+
Morganella morganii	MFS 1804	+	+	−	−	++	+++	+++
Morganella morganii	MFS 1805	+++	+	−	−	++	−	+++
Morganella morganii	MFS 1806	++	+	−	−	+++	+++	+++
Morganella morganii	MFS 1807	++	+	−	−	++	−	+++
Morganella morganii	MFS 1808	−	−	−	+++	−	++	+++
Morganella morganii	MFS 1809	−	−	−	+++	+/−	++	+++
Morganella morganii	MFS 1810	++	+	−	−	++	−	++
Morganella morganii	MFS 1811	+	−	+++	−	−	+	+/−
Morganella morganii	MFS 1812	+/−	−	+++	−	+++	+	+
Morganella morganii	MFS 1813	+	−	+++	−	++	+	+
Morganella morganii	MFS 1814	−	+/−	+++	−	+++	+	+
Morganella morganii	MFS 1815	+	−	+++	−	++	+	+/−
Morganella morganii	MFS 1816	−	−	−	+++	+/−	+++	+++
Morganella morganii	MFS 1817	−	−	−	+++	+/−	++	+++
Morganella morganii	MFS 1818	−	−	−	−	−	−	−
Morganella morganii	MFS 1819	−	+	−	−	−	−	−
Morganella morganii	MFS 1820	−	++	−	−	−	−	−
Morganella morganii	MFS 1821	−	++	−	−	−	−	−
Morganella morganii	MFS 1822	−	++	−	−	−	−	−
Morganella morganii	MFS 1823	−	+	−	−	−	−	−
Morganella morganii	MFS 1824	−	++	−	−	−	−	−

*  EOPs ≥ 0.5 marked as +++, 0.1 ≤ EOPs < 0.5 was marked as ++, 0.001 ≤ EOPs < 0.1 was marked as +, EOPs < 0.001 was marked as +/-, and an absence of plaques was marked as -.

M. morganii phages    The following M. morganii phages were isolated in this study: Momo2, Momo8, Momo10, Momo13, and Momo22. These phages were isolated from river water in Hokkaido, Japan, using the method described below. In addition, the M. morganii phages FSP1 (Yamaki et al., 2014) and ΦMV-4 (Yamaki et al., 2020) were also used in the M. morganii phage cocktail.

Phage counts were determined by the double agar overlay method using reference strains for each phage. The reference strains for each phage were as follows: MFS 1804 (Momo13 and Momo22), MFS 1808 (Momo8), MFS 1814 (Momo2 and Momo10), and NBRC3848^T (FSP1 and ΦMV-4). One hundred microliters of M. morganii culture and 100 µL of phage solution were added to 4 mL of 0.5% molten soft agar. The mixture was overlaid onto a tryptic soy agar (TSA; BD) plate and incubated at 30 °C for 24 h. After incubation, the plaque forming units (PFUs) were counted.

Isolation, propagation, and purification of phages    M. morganii phages were isolated from river water in Hokkaido, Japan. Equal or three volumes of the sample (river water) were added to 2-fold or 4-fold concentration TSB to achieve a 1-fold concentration. Ten strains of M. morganii (NBRC 3848^T, MFS 1801, MFS 1804, MFS 1805, MFS 1808, MFS 1811, MFS 1816, MFS 1818, MFS 1820, and MFS 1821) were used to isolate the M. morganii phages. The cultures of these strains were mixed, and the mixture was inoculated into the river water sample. The inoculum was incubated at 30 °C for 24 h. After centrifugation (10 000 × g, 10 min, 4 °C) of the culture, the supernatant was passed through a membrane filter (pore size, 0.45 µm) and the filtrate was used for spot testing to detect the phage. Each bacterial culture was inoculated into 4 mL of 0.5% molten soft agar and overlaid onto TSA plates. The filtrates were spotted onto each bacterial lawn and the plates were incubated at 30 °C for 24 h. Plaques were picked and suspended in SM buffer (100 mM sodium chloride (NaCl), 8 mM magnesium sulfate (MgSO₄), 50 mM Tris-HCl, and 0.01% gelatin, pH 7.5). Single plaques from the suspension were produced by the double agar overlay method, and a single plaque was picked again and resuspended in SM buffer. This procedure was repeated at least three times to purify the isolated phages.

Phages were propagated by incubation with reference strains in TSB. Reference strains were inoculated into 400 mL of TSB at a concentration of 10⁶ CFU/mL and incubated at 30 °C for 2 h. Phages were inoculated into the culture at a concentration of 107 PFU/mL and incubated at 30 °C for 6.5 h. The culture was centrifuged (8 000 × g, 4 °C, 30 min) and the supernatant was passed through a membrane filter (pore size, 0.45 µm). The liquid lysate was treated with 1 µg/mL DNase I and RNase A (Sigma-Aldrich, St. Louis, MO, USA) at 37 °C for 30 min before adding 1 M NaCl and incubating the mixture for 1 h on ice. Polyethylene glycol (PEG) 6000 (FujiFilm Wako Pure Chemical, Osaka, Japan) was added to the lysate at a concentration of 10–15% (w/v) and the mixture was incubated at 4 °C for more than 12 h. The lysate was then centrifuged (10 000 × g, 4 °C, 20 min) and the precipitated phages were resuspended in SM buffer and mixed with an equal volume of chloroform to remove the PEG. After centrifugation (10 000 × g, 4 °C, 10 min), the supernatant (water) was then collected and used as a phage concentrate. The phage concentrate was then purified by cesium chloride (CsCl) density gradient centrifugation (100 000 × g, 20 °C, 3 h) and dialysis with SM buffer using dialysis tubing (SpectraPor7, MWCO 8 000; Repligen Corporation, WM, USA).

Host range analysis    The host range of the tested phages was examined by the efficiency of plating (EOP) method described by Mirzaei and Nilsson (2015) with some modifications. Fresh cultures of bacterial strains were diluted and 100 µL of each suspension (10⁶ CFU/mL) was inoculated onto 4 mL of molten soft agar. Each phage was diluted to a concentration of 10^2–5 PFU/mL, and 100 µL of the phage diluent was added to the mixture. The mixture was overlaid onto TSA plates and incubated at 30 °C for 24 h. Plaque forming units were counted, and the EOP was calculated by dividing the PFU of the tested strains by the PFU of the reference strain.

Transmission electron microscopy (TEM)    The morphology of the isolated phages was observed using TEM (JEM-1011; JEOL, Tokyo, Japan). Purified phage suspensions were placed onto a grid (Excel support film; Nisshin EM, Tokyo, Japan) and incubated for 10 min to allow for adsorption of the phages onto the grid. Excess suspension was removed and the phages were negatively stained with 2.5% samarium triacetate (Nakakoshi et al., 2011). Excess staining solution was removed quickly, and phages were stained again with 2.5% samarium triacetate. Any remaining solution was removed in the same way and the grid was washed with 10 µL sterile distilled water. After drying the grid, the stained phages were observed using TEM.

Genome size estimation    The genome sizes of the isolated phages were estimated by pulse-field gel electrophoresis (PFGE), using a slight modification of the method of Lingohr et al. (2009). PFGE was performed using 1.1% agarose (SeaKem Gold Agarose; Lonza, Rockland, ME, USA) at 6 v/cm at 14 °C for 18 h with incremental pulses of 0.1–10 s on a CHEF-DR II system (Bio-Rad Laboratories, Hercules, CA, USA).

Analysis of adsorption kinetics    The adsorption test of the isolated phages was performed using the method described by Kropinski (2009) with minor modifications. The reference bacterial strains for each phage were inoculated at a concentration of 10⁶ CFU/mL and incubated at 30 °C until the OD₆₀₀ reached 0.5. After centrifugation (10 000 × g, 4 °C, 2 min), the bacterial cells were suspended in 10 mL of fresh TSB. Each phage (Momo2, Momo8, Momo10, Momo13, and Momo22) was inoculated into the culture at a concentration of 10⁶ PFU/mL, and the inoculum was incubated at 30 °C. One milliliter of the sample was collected, centrifuged (10 000 × g, 1 min), and filtered using a membrane filter (pore size, 0.45 µm) to remove the phages that adsorbed onto the bacterial cells. The number of phages in the filtrate was determined, and the rate (%) of free phages was calculated.

Analysis of one-step growth curve    The reference bacterial strains for each isolated phage were inoculated at a concentration of 10⁶ CFU/mL and incubated at 30 °C until the OD₆₀₀ reached 0.1. One hundred microliters of each phage suspension (100 µL) was inoculated at a concentration of 10⁶ PFU/mL into 10 mL of the culture, and the inoculum was incubated at 30 °C for 5 min for phage adsorption to the host cells. After centrifugation (10 000 × g, 4 °C, 2 min), the supernatant was removed and the bacterial pellet was resuspended in 10 mL of fresh TSB. The culture was incubated at 30 °C and samples were collected every 10 min. The number of phages was measured as described above, and the latent period and burst size of the isolated phages were determined.

Preparation of the phage cocktail and comparison with individual phages    Five isolated phages (Momo2, Momo8, Momo10, Momo13, and Momo22), FSP1 (Yamaki et al., 2014), and ΦMV-4 (Yamaki et al., 2020) were used to prepare a phage cocktail. An equivalent mixture of each phage was used to prepare the M. morganii phage cocktail. The inhibitory effects of the phage cocktail and individual phages on M. morganii were analyzed based on changes in turbidity. Each of the M. morganii cultures (i.e., NBRC 3848^T, MFS 1804, MFS 1808, and MFS 1814) was centrifuged (10 000 × g, 2 min) and washed with phosphate-buffered saline (PBS). The bacterial suspension was diluted with TSB to a concentration of 2 × 10³ CFU/mL, and 100 µL of the culture was inoculated into the wells of a 96-well microtiter plate. The phage cocktail and each single phage were also diluted with TSB at a concentration of 2 × 10⁷ PFU/mL, and 100 µL of each phage diluent was inoculated to the wells and mixed with the bacterial culture. The plate was incubated at 30 °C and the OD₅₉₀ was measured at 0, 3, 6, 9, 12, 24, 36, and 48 h.

Challenge testing using broth medium    The M. morganii cultures NBRC 3848^T, MFS 1804, MFS 1808, and MFS 1814 were incubated with TSB supplemented with 1% histidine (TSBH) at 30 °C for 24 h, and 100 µL of the culture was transferred into 4 mL of TSBH. The inoculum was incubated at 30 °C until the OD₆₀₀ reached 0.1, and the culture was diluted with PBS. The diluents of each strain were mixed equally, and 100 µL of the mixture was inoculated into 24.8 mL of TSBH (pH 6.5) at a concentration of 1 × 10³ CFU/mL. Then, 100 µL of the M. morganii phage cocktail was inoculated into the medium at a concentration of 1 × 10⁷ PFU/mL. The medium was incubated at 12 °C for 8 d, and the viable cell counts and histamine concentrations were determined at 2-d intervals. To measure the histamine concentration, 1 mL of the sample solution was heated at 100 °C for 20 min to denature histidine decarboxylase. The sample solution was centrifuged (10 000 × g, 5 min), and the histamine concentration in the supernatant was determined using a histamine test kit (Kikkoman Biochemifa Company, Tokyo, Japan). To prevent underestimation of the viable counts due to carryover of phages, free phages were inactivated before measurement. We used a virucidal agent (TeaF) to inactivate the phages. TeaF (a mixture of black tea extract and 4.3 mM FeSO₄) was prepared using the method of Yamaki et al. (2022), which is a modification of the method proposed by de Siqueira et al. (2006). A total of 100 µL of the sample solution was mixed with 900 µL of 66.6% TeaF and incubated at room temperature for 15 min. Viable cell counts of the phage-inactivated samples were determined by spreading the sample solutions onto TSA plates. The plates were incubated at 30 °C for 24 h, and the number of colonies was counted.

Challenge testing using canned tuna    The M. morganii isolates NBRC 3848^T, MFS 1804, MFS 1808, and MFS 1814 were incubated until the OD₆₀₀ reached 0.1, diluted with PBS, and mixed equally as described above. One milliliter of the bacterial mixture was inoculated into 120 g of salt-free canned (sterilized) tuna flakes at a concentration of 1 × 10³ CFU/g, and 1 mL of the phage cocktail was inoculated into the canned tuna flakes at a concentration of 1 × 10⁷ PFU/mL. The same volume of PBS (pH 7.4) was used as the control without the phage cocktail. The tuna flakes were stored at 12 °C for 8 d. Tuna samples were collected every two days to measure the bacterial counts and histamine concentrations. For the measurement of histamine, 5 g of the collected sample was added to 45 mL of PBS. After mixing with a stomacher, 1 mL of the sample was heated at 100 °C for 20 min to denature histidine decarboxylase, and the histamine concentration was determined as described above. To measure the bacterial counts, 5 g of the collected sample was added to 25 mL of 96% TeaF. After mixing with the stomacher, the mixture was incubated at room temperature for 15 min to inactivate free phages in the tuna sample. After phage inactivation, 20 mL of PBS was added and mixed well with the sample to prepare a 10-fold food homogenate. The sample solution was serially diluted with PBS, and 100 µL of the sample was spread onto TSA plates. The plates were incubated at 30 °C for 24 h and the bacterial counts were recorded.

Results and Discussion

Host range and morphology of the five isolated phages    In this study, we isolated and characterized five M. morganii phages. The host ranges of the isolated phages and previously isolated phages (FSP1 and ΦMV-4) are shown in Table 1. The EOP test showed that phages Momo2, Momo8, Momo10, Momo13, and Momo22 formed plaques on 8, 4, 15, 14, and 17 strains of M. morganii, respectively. Momo10, Momo13, and Momo22 showed wide host ranges, and Momo22 had the widest host range among the five isolated phages. However, M. morganii subsp. morganii NBRC 3848^T, M. morganii MFS 1818, MFS 1819, MFS 1820, MFS 1821, MFS 1822, MFS 1823, and MFS 1824 were not sensitive to these five phages. M. morganii NBRC 3848^T was sensitive to phage FSP1 and ΦMV-4, and ΦMV-4 formed plaques on MFS 1819, MFS 1820, MFS 1821, MFS 1822, MFS 1823, and MFS 1824. Only M. morganii MFS 1818 was not sensitive to any phage. The host range of phages is typically narrow. The combination of several phages is thus one strategy to overcome the limited host range of phages (Lu and Koeris, 2011). The combination of the seven phages allowed at least one phage to infect 24 of the 25 strains of M. morganii (Table 1).

The morphologies of the isolated phages are shown in Figure 1. TEM findings showed that Momo2, Momo10, Momo13, and Momo22 had a long, non-contractile, and flexible tail, indicating that these phages are siphoviruses. The tail length of Momo2 was shorter than that of Momo10, Momo13, and Momo22, suggesting that Momo2 and other phages belong to different taxa. The tail of Momo8 was long and contractile, indicating that Momo8 was a myovirus. Genome size estimation by PFGE indicated that the genomes of Momo2 and Momo8 were approximately 37 kbp, and the genomes of Momo10, Momo13, and Momo22 were approximately 50 kbp in size (data not shown). The similar morphology and genome size suggest that Momo10, Momo13, and Momo22 are closely related to each other. In contrast, the host ranges of the five isolated phages were different (Table 1), suggesting that the host recognition mechanisms of these phages are slightly different.

Fig. 1.

Transmission electron microscopy (TEM) images of the isolated phages.

The tailed phages (order Caudovirales) were traditionally classified based on the differences in tail structure. Based on the length and contractility of the phage tail, the tailed phages are classified into three families, i.e., Siphoviridae, Myoviridae, and Podoviridae (Ackermann, 2009). Previously reported M. morganii phages were classified as Myoviridae (FSP1, ΦMV-1, ΦMV-4, and MP1) and Podoviridae (MmP1 and MP2) and their morphological characteristics (Oliveira et al., 2017; Yamaki et al., 2014; Yamaki et al., 2020; Zhu et al., 2010) suggested that Momo2, Momo10, Momo13, and Momo22 are novel M. morganii phages that have not been described to date. However, the International Committee on Taxonomy of Viruses proposed the creation of new families in the taxa of tailed phages, such as Autographiviridae, Chaseviridae, and Demerecviridae (Adriaenssens et al., 2020). As it is difficult to identify phages solely based on the TEM observations, in the future, genomic analyses will be required to obtain a more accurate identification of the isolated phages. Three Morganella phage genomes (MmP1, MP1, and MP2) have been sequenced and reported (Oliveira et al., 2017; Zhu et al., 2010). In the current taxonomy, these phages were classified into the genus Minipunavirus of the subfamily Studiervirinae of the family Autographiviridae (MmP1 and MP2) and the subfamily Tevenvirinae of the family Myoviridae (MP1). We are currently attempting to sequence the genome of our M. morganii phage for further genomic characterization.

Adsorption and one-step growth of the five isolated phages    The changes in the adsorption rates of the isolated phages are shown in Figure 2. More than 90% of each phage strain was adsorbed onto the M. morganii cells in 10 min. The adsorption rates of Momo2, Momo8, Momo10, Momo13, and Momo22 at 10 min were 98.2%, 99.1%, 92.5%, 99.9%, and 99.2%, respectively. The adsorption of Momo13 was the fastest, while that of Momo10 was the slowest. Although Momo13 and Momo22 had similar morphology and genome sizes, the different adsorption patterns on the same M. morganii strain (MFS 1804) suggest that the receptors used by these phages are different. Similarly, the different adsorption patterns of Momo2 and Momo10 on the same M. morganii strain (MFS 1814) also suggests that these phages employ different receptors. The adsorption rates of M. morganii phages FSP1 and ΦMV-4 were approximately 40% at 5 min and 50% at 10 min, respectively (Yamaki et al., 2014; Yamaki et al., 2020). The adsorption of the five isolated phages was faster than those of FSP1 and ΦMV-4, and these fast adsorption rates are considered to be suitable for the biocontrol of M. morganii.

Fig. 2.

Adsorption kinetics of the isolated M. morganii phages. Symbols represent Momo2 (●), Momo8 (□), Momo10 (○), Momo13 (△), and Momo22 (▲). The results are shown as the mean ± standard deviation from three independent experiments.

One-step growth curves for each of the isolated phages are shown in Figure 3. The latent periods of Momo2, Momo8, Momo10, Momo13, and Momo22 were 40, 20, 50, 30, and 30 min, and the burst sizes of these phages were 428, 279, 22, 138, and 254 PFU per infected cell, respectively. The burst sizes of Momo2, Momo8, Momo13, and Momo22 were larger than 42 PFU per infected cell of FSP1 and 62 PFU per infected cell of ΦMV-4 (Yamaki et al., 2014; Yamaki et al., 2020). Moreover, the burst sizes of M. morganii phages MP1 and MP2 have been reported as 41 and 16 PFU per infected cell, respectively (Oliveira et al., 2017). These results indicate that Momo2, Momo8, Momo13, and Momo22 have large burst sizes.

Fig. 3.

One-step growth curves of Momo2 (A), Momo8 (B), Momo10 (C), Momo13 (D), and Momo22 (E). The results are shown as the mean ± standard deviation from three independent experiments.

Antimicrobial potential of individual phages and the M. morganii phage cocktail    The use of a phage cocktail can suppress the emergence of phage-resistant bacteria (O'Flynn et al., 2004; Tanji et al., 2004; Yuan et al., 2019). The antimicrobial effects of the individual phages and the phage cocktail containing seven phages (Momo2, Momo8, Momo10, Momo13, Momo22, FSP1, and ΦMV-4) were evaluated (Fig. 4). Increases in OD₅₉₀ of the control samples were observed after 9 h. In the single-phage treatments (Momo2, Momo10, Momo13, Momo22, or ΦMV-4), the growth of M. morganii was delayed, but M. morganii titers began to increase after 12 h. An increase in turbidity was not detected in samples treated with Momo8 or FSP1, suggesting that the antimicrobial efficacy of these two phages was high. The phage cocktail also completely inhibited the growth of M. morganii during the experimental period; the exception being M. morganii MFS 1804 (Fig. 4A). However, the growth of M. morganii MFS 1804 was retarded by the phage cocktail (compared to single phage treatment), suggesting that the phage cocktail could effectively inhibit the growth of M. morganii for a longer period.

Fig. 4.

Growth of M. morganii MFS 1804 (A), MFS 1808 (B), MFS 1814 (C), and NBRC 3848^T (D), with or without phage treatment. Symbols represent the control without phages (◇), single phage treatment (△, □), and phage cocktail treatment (○). For single phage treatments in panels (A) to (D), phages were used as follows: Momo13 (□) and Momo22 (△) for M. morganii MFS 1804 (A), Momo8 (□) for M. morganii MFS 1808 (B), Momo2 (□) and Momo10 (△) for M. morganii MFS 1814 (C), and FSP1 (□) and ΦMV-4 (△) for M. morganii NBRC 3848^T (D).

Inhibition of adsorption is a strategy by which phage-resistant bacteria prevent phage infection (Labrie et al., 2010). To increase the efficacy of the phage cocktail, it is important to select appropriate phages for the phage cocktail. For example, Tanji et al. (2004) reported combinations of phages that could and could not inhibit the emergence of phage-resistant bacteria. This study also suggests that a phage cocktail consisting of more than two phages that use different receptors may inhibit the emergence of phage-resistant bacteria (Tanji et al., 2004). In this study, we prepared an M. morganii phage cocktail using seven different phages. The different host ranges (Table 1) and different adsorption patterns (Fig. 2) suggest that the host recognition mechanisms of the phages in this study were not the same. Moreover, Paez-Espino et al. (2015) showed that increased phage persistence against CRISPR immunity occurs through recombination in the targeted region by CRISPR immunity between multiple phage genomes. This suggests that combining multiple M. morganii phages may improve resistance of the M. morganii phage cocktail to CRISPR immunity. Further, these results indicate that the use of the M. morganii phage cocktail is more advantageous for the biocontrol of M. morganii than that of individual phages, as it suppresses the growth of phage-resistant bacteria and expands the host range. However, some M. morganii strains (MFS 1819–1824) were sensitive only to ΦMV-4 (Table 1), suggesting that it is difficult to suppress the emergence of phage-resistant strains of M. morganii MFS 1819–1824 with the phage cocktail prepared in this study. Therefore, improvements in the selection of phage components in the cocktail will be necessary in the future in order to obtain superior effects for controlling the M. morganii in seafood.

Inhibitory effect of the M. morganii phage cocktail on histamine accumulation    The antimicrobial effect of the M. morganii phage cocktail was more effective than that of a single phage (Fig. 4). The inhibitory effect of the phage cocktail on histamine accumulation by M. morganii was evaluated in broth medium and canned tuna flakes incubated at 12 °C. Changes in the viable cell count and histamine concentration in the broth medium are shown in Figure 5. The phage cocktail delayed the growth of M. morganii and accumulation of histamine in the broth medium. Viable counts in the control sample exceeded 8 log CFU/mL at 4 d, but the M. morganii concentration in the phage-treated samples did not exceed 8 log CFU/mL for 8 d. The histamine concentration of the control samples was 4 285 mg/L at 8 d; however, that of the phage-treated samples was 804 mg/L.

Fig. 5.

Viable cell counts of M. morganii (A) and histamine concentrations (B) in the broth medium, with or without phage cocktail treatment. In (A), symbols represent the viable cell count of the control without phages (◆) and phage cocktail treatment (●). In (B), black and white columns represent the histamine concentration of the control without phages and phage cocktail treatment. The results are shown as the mean ± standard deviation from three independent experiments.

Changes in the viable cell counts and histamine concentrations in the canned tuna flakes are shown in Figure 6. In the control sample, M. morganii rapidly increased and viable counts reached 7.8 log CFU/g at 4 d, but an increase in viable counts in the phage-treated samples was slower than that of the control (Fig. 6A). In the control sample, 208 mg/kg of histamine was detected at 4 d, and the histamine concentration exceeded 3 000 mg/kg four days later. In the phage-treated samples, histamine accumulation was effectively inhibited by the phage cocktail. Histamine was detected on day 6, and the histamine concentration in the final sampling period was 345 mg/kg (Fig. 6B).

Fig. 6.

Viable cell counts of M. morganii (A) and histamine concentrations (B) in the canned tuna flakes, with or without phage cocktail treatment. In (A), symbols represent the viable cell counts of the control without phages (◆) and phage cocktail treatment (●). In (B), black and white columns represent the histamine concentrations in the control without phages and phage cocktail treatment. The results are shown as the mean ± standard deviation from three independent experiments.

Regulation of the histamine concentration in fresh or canned fish products—established by Codex Alimentarius—is the average of samples less than 100 ppm, and none of the samples are over 200 ppm (DeBeer et al., 2021). When considering this regulation, the storage period in which the standard was exceeded was 4 d in the control sample (208 mg/kg) and 8 d in the phage-treated samples (345 mg/kg), indicating that the phage cocktail prolonged the shelf-life for 4 d. Moreover, the no-observed-adverse-effect level (NOAEL) of histamine is 50 mg (FAO/WHO, 2012), and the experiment showed that the phage cocktail could reduce the risk of being affected by the health hazards caused by histamine poisoning (Fig. 6).

Our previous study indicated that the M. morganii phages (FSP1 and ΦMV-4) inhibit the growth of M. morganii and histamine accumulation (Yamaki et al., 2018; Yamaki et al., 2020). However, the host ranges of FSP1 and ΦMV-4 were not broad enough to inhibit M. morganii strains used in this study (Table 1), indicating that more M. morganii phages were needed for effective prevention of histamine accumulation by M. morganii. In this study, five phages were newly isolated, and their characteristics suggested that they are novel M. morganii phages. The M. morganii phage cocktail combining seven phages was able to suppress the growth and histamine accumulation in a four-strain mixture of M. morganii in canned tuna flakes (Fig. 6). The results of the EOP test and challenge testing show that the M. morganii phage cocktail effectively improves the inhibitory effect of histamine accumulation compared to using FSP1 or ΦMV-4 alone. These results also suggest that the M. morganii phage cocktail has the potential for use as a biocontrol agent for the inhibition of M. morganii growth and the accumulation of histamine in fishery products.

Acknowledgements    This work was supported by a grant from the Mayekawa Houonkai Foundation. We would like to thank Editage (www.editage.com) for English language editing.

Conflict of interest    There are no conflicts of interest to declare.

References

•  Ackermann,  H.W. (2009). Phage classification and characterization. In “Bacteriophages.” ed. by  M.R.  Clokie and  A.M.  Kropinski. Methods in Molecular Biology, vol 501, Humana Press, New York, pp. 127-140.
•  Adriaenssens,  E.M.,  Sullivan,  M.B.,  Knezevic,  P.,  Zyl,  L.J. van,  Sarkar,  B.L.,  Dutilh,  B.E.,  Alfenas-Zerbini,  P.,  Łobocka,  M.,  Tong,  Y.,  Brister,  J.R.,  Switt,  A.I.M.,  Klumpp,  J.,  Aziz,  R.K.,  Barylski,  J.,  Uchiyama,  J.,  Edwards,  R.A.,  Kropinski,  A.M.,  Petty,  N.K.,  Clokie,  M.R.J.,  Kushkina,  A.I.,  Morozova,  V.V.,  Duffy,  S.,  Gillis,  A.,  Rumnieks,  J.,  Kurtböke,  İ.,  Chanishvili,  N.,  Goodridge,  L.,  Wittmann,  J.,  Lavigne,  R.,  Jang,  H.B.,  Prangishvili,  D.,  Enault,  F.,  Turner,  D.,  Poranen,  M.M.,  Oksanen,  H.M., and  Krupovic,  M. (2020). Taxonomy of prokaryotic viruses: 2018-2019 update from the ICTV Bacterial and Archaeal Viruses Subcommittee. Arch. Virol. 165, 1253-1260.
•  Becker,  K.,  Southwick,  K.,  Reardon,  J.,  Berg,  R., and  MacCormack,  J.N. (2001). Histamine poisoning associated with eating tuna burgers. JAMA, 285, 1327-1330.
•  Bjornsdottir-Butler,  K.,  Bowers,  J.C., and  Benner,  R.A. (2015). Prevalence and characterization of high histamine-producing bacteria in Gulf of Mexico fish species. J. Food Prot., 78, 1335-1342.
•  Bjornsdottir-Butler,  K.,  May,  S.,  Hayes,  M.,  Abraham, and  Benner,  R.A. (2020). Characterization of a novel enzyme from Photobacterium phosphoreum with histidine decarboxylase activity. Int. J. Food Microbiol., 334, 108815.
•  DeBeer,  J.,  Bell,  J.W.,  Nolte,  F.,  Arcieri,  J., and  Correa,  G. (2021). Histamine limits by country: a survey and review. J. Food Prot., 84, 1610-1628.
•  de Siqueira,  R.S.,  Dodd,  C.E.R., and  Rees,  C.E.D. (2006). Evaluation of the natural virucidal activity of teas for use in the phage amplification assay. Int. J. Food Microbiol., 111, 259-262.
•  Feng,  C.,  Teuber,  S., and  Gershwin,  M.E. (2016). Histamine (scombroid) fish poisoning: a comprehensive review. Clinic. Rev. Allerg. Immunol., 50, 64-69.
• Food and Agriculture Organization of the United Nations/World Health Organization (FAO/WHO) (2012). “Joint FAO/WHO expert meeting on the public health risks of histamine and other biogenic amines from fish and fishery products, meeting report.” FAO Headquarters, Rome, Italy.
•  Kanki,  M.,  Yoda,  T.,  Ishibashi,  M., and  Tsukamoto,  T. (2004). Photobacterium phosphoreu caused a histamine fish poisoning incident. Int. J. Food Microbiol., 92, 79-87.
•  Kanki,  M.,  Yoda,  T.,  Tsukamoto,  T., and  Baba,  E. (2007). Histidine decarboxylase and their role in accumulation of histamine in tuna and dried saury. Appl. Environ. Microbiol., 73, 1467-1473.
•  Kim,  S.-H.,  An,  H.,  Wei,  C.-I,  Visessanguan,  W.,  Benjakul,  S.,  Morrissey,  M.T.,  Su,  Y.-C., and  Pitta,  T.P. (2003). Molecular detection of a histamine former, Morganella morganii, in albacore, mackerel, sardine, and a processing plant. J. Food Sci., 68, 453-457.
•  Kropinski,  A.M. (2009). Measurement of the rate of attachment of bacteriophage to cells. In “Bacteriophages.” ed. by  M.R.  Clokie and  A.M.  Kropinski. Methods in Molecular Biology, vol 501, Humana Press, New York, pp. 151-155.
•  Labrie,  S.J.,  Samson,  J.E., and  Moineau,  S. (2010). Bacteriophage resistance mechanisms. Nat. Rev. Microbiol., 8, 317-327.
•  Lingohr,  E.,  Frost,  S., and  Johnson,  R.P. (2009). Determination of bacteriophage genome size by pulsed-field gel electrophoresis. In “Bacteriophages.” ed. by  M.R.  Clokie and  A.M.  Kropinski. Methods in Molecular Biology, vol 502, Humana Press, New York, pp. 19-25.
•  Lu,  T.K. and  Koeris,  M.S. (2011). The next generation of bacteriophage therapy. Curr. Opin. Microbiol., 14, 524-531.
•  Mirzaei,  M.K., and  Nilsson,  A.S. (2015). Isolation of phages for phage therapy: a comparison of spot tests and efficiency of plating analyses for determination of host range and efficacy. PLoS ONE, 10, e0118557.
•  Nakakoshi,  M.,  Nishioka,  H., and  Katayama,  E. (2011). New versatile staining reagents for biological transmission electron microscopy that substitute for uranyl acetate. J. Electron Microsc., 60, 401-417.
•  O'Flynn,  G.,  Ross,  R.,  Fitzgerald,  G., and  Coffey,  A. (2004). Evaluation of a cocktail of three bacteriophages for biocontrol of Escherichia coli O157:H7. Appl. Environ. Microbiol., 70, 3417-3424.
•  Oliveira,  H.,  Pinto,  G.,  Oliveira,  A.,  Noben,  J.-P.,  Hendrix,  H.,  Lavigne,  R.,  Łobocka,  M.,  Kropinski,  A.M., and  Azeredo,  J. (2017). Characterization and genomic analyses of two newly isolated Morganella phages define distant members among Tevenvirinae and Autographivirinae subfamilies. Sci. Rep., 7, 46157.
•  Paez-Espino,  D.,  Sharon,  I.,  Morovic,  W.,  Stahl,  B.,  Thomas,  B.C.,  Barrangou,  R., and  Banfield,  J.F. (2015). CRISPR immunity drives rapid phage genome evolution in Streptococcus thermophilus. mBio, 6, e00262-15.
•  Sulakvelidze,  A. (2013). Using lytic bacteriophages to eliminate or significantly reduce contamination of food by foodborne bacterial pathogens. J. Sci. Food Agric., 93, 3137-3146.
•  Tanji,  Y.,  Shimada,  T.,  Yoichi,  M.,  Miyanaga,  K.,  Hori,  K., and  Unno,  H. (2004). Toward rational control of Escherichia coli O157:H7 by a phage cocktail. Appl. Microbiol. Biotechnol., 64, 270-274.
•  Torido,  Y.,  Ohshima,  C.,  Takahashi,  H.,  Miya,  S.,  Iwakawa,  A.,  Kuda  T., and  Kimura,  B. (2014). Distribution of psychrophilic and mesophilic histamine-producing bacteria in retailed fish in Japan. Food Cont., 46, 338-342.
•  Wang,  D.,  Yamaki,  S.,  Kawai,  Y., and  Yamazaki,  K. (2020). Histamine production behaviors of a psychrotolerant histamine-producer, Morganella psychrotolerans, in various environmental conditions. Curr. Microbiol., 77, 460-467.
•  Yamaki,  S.,  Kawai,  Y., and  Yamazaki,  K. (2015). Characterization of a novel bacteriophage, Phda1, infecting the histamine-producing Photobacterium damselae subsp. damselae. J. Appl. Microbiol., 118, 1541-1550.
•  Yamaki,  S.,  Kawai,  Y., and  Yamazaki,  K. (2018). Biocontrol of Morganella morganii subsp. morganii and histamine accumulation in tuna meat by treatment with a lytic bacteriophage. Food Sci. Tec. Res., 24, 329-337.
•  Yamaki,  S.,  Kuronuma,  S.,  Kawai,  Y., and  Yamazaki,  K. (2020). Inhibitory effect of a combination with novel jumbo bacteriophages ΦMV-1 and ΦMV-4 on Morganella morganii subsp. morganii growth and histamine accumulation. Int. J. Food Microbiol., 317, 108457.
•  Yamaki,  S.,  Omachi,  T.,  Kawai,  Y., and  Yamazaki,  K. (2014). Characterization of a novel Morganella morganii bacteriophage FSP1 isolated from river water. FEMS Microbiol. Lett., 359, 166-172.
•  Yamaki,  S. and  Yamazaki,  K. (2019). Histamine food Poisoning associated with fisheries produces and histamine-producing bacteria. Jpn. J. Food Microbiol. (Nippon Shokuhin Biseibutsu Gakkai Zasshi), 36, 75-83, 2019. (in Japanese)
•  Yamaki,  S.,  Yamazaki,  K., and  Kawai,  Y. (2022). Broad host range bacteriophage, EscoHU1, infecting Escherichia coli O157:H7 and Salmonella enterica: Characterization, comparative genomics, and applications in food safety. Int. J. Food Microbiol., 372, 109680.
•  Yuan,  Y.,  Wang,  L.,  Li,  X.,  Tan,  D.,  Cong,  C., and  Xu,  Y. (2019). Efficacy of a phage cocktail in controlling phage resistance development in multidrug resistant Acinetobacter baumannii. Virus Res., 272, 197734.
•  Zhu,  J.,  Rao,  X.,  Tan,  Y.,  Xiong,  K.,  Hu,  Z.,  Chen,  Z.,  Jin,  X.,  Li,  S.,  Chen,  Y., and  Hu,  F. (2010). Identification of lytic bacteriophage MmP1, assigned to a new member of T7-like phages infecting Morganella morganii. Genomics, 96, 167-172.