Effects of heterologous genome microinjection on the enlargement of Enterococcus faecalis protoplasts.

The lactic acid bacterium Enterococcus faecalis genomic DNA and seven phylogenetically distant bacterial genomic DNAs were microinjected into 126 enlarged protoplasts of E. faecalis. After the microinjection, a time-lapse observation was performed on how the cells enlarged. Most cells did not stop enlarging. The enlargement patterns were compared with the enlargement of E. faecalis protoplasts not treated by microinjection (control). They were clustered into three groups, with different levels and speeds of protoplast enlargement. The statistical analyses showed that the protoplasts injected by E. faecalis and four of the seven phylogenetically different bacterial genomic DNAs had enlargement patterns significantly different from those of the control. Thus, injected genomic DNAs affected the protoplast enlargement. Most of the affected cells, including the E. faecalis genome, had weakened enlargement.

Introduction

The horizontal transfer of genetic elements greatly influences on bacterial evolution (Garcia-Vallvé et al., 2000; Koonin et al., 2001; Nishida and Oshima, 2019). However, the mechanism of the horizontal transfer is uncertain, except for plasmids and viruses. Furthermore, to our knowledge, any horizontal transfers of bacterial chromosomes have not been detected and reported. Thus, it has been uncertain whether bacterial cells can control the genetic information when incorporating heterologous chromosomes.

Horizontally transferred plasmids and viruses have lower guanine-cytosine (GC) content than their host chromosomes (Nishida, 2012; Rocha and Danchin, 2002). However, it is uncertain whether such a phenomenon also occurs in horizontally transferred chromosomes.

The present genome manipulation technology is based on the repetition of existing gene manipulation (Itaya et al., 2005; Lartigue et al., 2007). The host bacteria cells used in genome manipulation, such as Bacillus subtilis and Mycoplasma mycoides, are limited (Itaya et al., 2005; Lartigue et al., 2007). In addition, such an experiment is time-consuming, labor-intensive, and costly.

Thus, we focused on microinjection technology. Unfortunately, normal bacterial cells are too small to microinject. Instead, bacterial protoplasts are generated by the cell wall lyses, which can enlarge without cell division under suitable culture conditions where cell wall synthesis is inhibited (Kami et al., 2019; Kuroda et al., 1998; Kusaka, 1967; Lederberg, 1956; Nakamura et al., 2011; Nishino et al., 2018; Takahashi et al., 2016). We succeeded in enlarging the cells of the lactic acid bacterium Enterococcus faecalis and in microinjecting the fluorescent proteins into their enlarged cells (Takahashi et al., 2020).

The E. faecalis protoplast enlargement requires DNA replication (Kami et al., 2019; Tsuchikado et al., 2020). Thus, when DNA replication stops, cell enlargement also stops. There are two possible causes. One is that the replicated genomic DNA is directly involved in the plasma membrane enlargement. It has been reported that the DNA is bound to the plasma membrane of bacteria (Earhart et al., 1968; Leibowitz and Schaechter, 1975; Marvin, 1968; Sueoka and Quinn, 1968). Therefore, if the attachment is required only for cell division, it may not be essential for protoplast enlargement. The other reason is that the replication and plasma membrane synthesis systems are linked; i.e., both stand side by side and do not function independently. Thus, in the former case, the injection of the E. faecalis genomic DNAs may promote plasma membrane synthesis. In the latter case, the effects of the genomic DNA injection on the replication system may affect the cell enlargement.

This study microinjected the E. faecalis genomic DNA and various heterologous genomic DNAs into the E. faecalis enlarged protoplasts and investigated the effects of the injected DNAs on protoplast enlargement.

Materials and methods

Preparation and culture of protoplasts

E. faecalis NBRC 100480 was cultivated and protoplasts were prepared as previously described (Kami et al., 2019). The protoplasts were centrifuged at 7000 r.p.m. for 5 min and resuspended in Difco Marine Broth 2216 (DMB; 5 g/L peptone, 1 g/L yeast extract, 0.1 g/L ferric citrate, 19.45 g/L NaCl, 5.9 g/L MgCl2, 3.24 g/L MgSO4, 1.8 g/L CaCl2, 0.55 g/L KCl, 0.16 g/L NaHCO3, 0.08 g/L KBr, 34 mg/L SrCl2, 22 mg/L H3BO3, 8 mg/L Na2HPO4, 4 mg/L Na2SiO3, 2.4 mg/L NaF, and 1.6 mg/L NH4NO3 [BD, Franklin Lakes, NJ]) containing 300 µg/mL penicillin G. 2 mL of DMB containing 300 µg/mL of penicillin G (Wako, Osaka) was added to the resulting suspension (10 µL) and incubated at 24 °C.

Microinjection workstation

The microinjection workstation consisted of an Olympus IX73 (Olympus, Japan) differential interference microscope with a UPlanFL N objective 4×, a UPlanFL N objective 10×, a LUCPlan FL N objective 40×, and a UPlanFL N objective 60× . In addition, the microscope was equipped with a TransferMan 4r micromanipulator set (Eppendorf, Germany) with a FemtoJet 4i microinjector (Eppendorf, Germany).

Preparation of DNA injection solution

Heterologous genomic DNAs were selected in consideration of the evolutionary relationship and the DNA composition (GC content). Genomic DNAs of D. grandis ATCC 43672, E. faecalis NBRC 100480, E. litoralis NBRC 102620, E. coli MG 1655, L. amnigena NBRC 105700, L. curvatus NBRC 15884 and L. lactis NBRC 100933 were prepared using a DNeasy Blood and Tissue Kit (Qiagen, Germany) and were eluted with 25 mM Tris–HCl (pH 7.4) containing 300 mM sucrose. In addition, the genomic DNA of B. subtilis 168 was prepared using a modified method of Saito and Miura (1963) and was eluted with 25 mM Tris–HCl (pH 7.4) containing 300 mM sucrose. Finally, all bacterial genomic DNA was dissolved at a concentration of 100 ng/µL in 25 mM Tris–HCl (pH 7.4) containing 300 mM sucrose.

Loading the DNA injection solution into a microinjection needle

The DNA injection solution was mixed with ATP (Thermo Fisher Scientific Baltics UAB, Lithuania) at a final concentration of 5 mM (Moriizumi et al., 2018) and centrifuged at 11,000 × g for 15 min at 4 °C. Next, a 2 µL aliquot was loaded into the tip of a Femtotip II microinjection needle (pore size: 0.5 µm, Eppendorf, Germany) using a microloader (Eppendorf, Germany). Next, the needle was tightly mounted in the capillary holder of the FemtoJet 4i microinjector, and then fixed onto the micromanipulator.

Microinjection of E. faecalis enlarged cells

The E. faecalis enlarged protoplast culture (400 µL) at 48 h of incubation in DMB containing penicillin G was mounted on a cover glass (40 × 50, Matsunami, Japan) and placed on the microscope stage. First, the DNA solution was released into enlarged cells using the FemtoJet 4i microinjector. Next, Femtotip II was inserted into enlarged cells for approximately 10–60 s while the DNA solution was released at 50–100 hPa. An intracellular change confirmed successful injection. Following the microinjection, the cover glass was put in a petri dish, and a time-lapse observation was performed under a BZ-X710 microscope (Keyence, Japan).

Cell size measurement

Bright-field microscopy images of the protoplasts were obtained using the BZ-X710 microscope (Keyence, Japan). The cell sizes were measured using the cellSens Standard 1.11 imaging software (Olympus, Tokyo).

Computer analyses

Each profile of cell diameters from 0 to 20 h was used for cluster analysis. The programs “dist” and “hclust” were used in the statistical software R (http://www.R-project.org/). Two statistical tests, namely, the chi-square test and Fisher's exact test, were also performed using R.

Results and discussion

E. faecalis protoplast enlargement patterns were clustered into three groups

The bacterial genomic DNAs used in this study are shown in Table 1. The E. faecalis protoplasts that the E. faecalis genomic DNA or heterologous genomic DNAs were microinjected into, continued to enlarge (Fig. 1). The 126 protoplasts of E. faecalis into which eight phylogenetically distant bacterial genomic DNAs were injected had been observed by time-lapse microscopy (every hour for 20 h). These enlargement patterns were compared with 20 E. faecalis protoplasts (control, Fig. 1A) that were not treated by microinjection. Thus, most E. faecalis protoplasts possessing injected genomic DNAs can survive.

Table 1

Bacterial genomes used in microinjection into Enterococcus faecalis protoplasts.

Bacterial species	Gram staining	Genome size (Mbp)	GC content (%)
Lactococcus lactis	positive	2.6	35
Enterococcus faecalis	positive	2.8	37.5
Lactobacillus curvatus	positive	1.8	42
Bacillus subtilis	positive	4.2	43.5
Escherichia coli	negative	4.6	50.8
Lelliottia amnigena	negative	4.4	52.9
Erythrobacter litoralis	negative	3.3	65.2
Deinococcus grandis	negative	4.0	69.9

Fig. 1

Change of cell diameter of E. faecalis protoplasts. Control, no DNA injection. Bacterial species name means that its genomic DNA is injected at 0 h. The time-lapse observation was performed every hour for 20 h, using a BZ-X710 microscope (Keyence, Japan).

We compared patterns of increase and decrease in cell diameter of those protoplasts (Fig. 2) and performed cluster analysis. As a result, a total of 146 enlargement patterns were clustered into three groups (Fig. 3). Cluster 1 consists of 18 protoplasts that have the highest enlargement level in the three clusters (Fig. 4A). Cluster 2 consists of 45 protoplasts that have the lowest enlargement level among the three clusters (Fig. 4B). Finally, cluster 3 consists of 83 protoplasts, in which 16 of 20 control protoplasts existed (Fig. 4C).

Fig. 2

Patterns of increase and decrease in cell diameter of E. faecalis protoplasts. Control, no DNA injection. Bacterial species name means that its genomic DNA is injected at 0 h. The time-lapse observation was performed every hour for 20 h, using a BZ-X710 microscope (Keyence, Japan). Red indicates a change of median at each incubation time.

Fig. 3

Phylogenetic relationships among 146 increasing and decreasing patterns in the cell diameter of E. faecalis protoplasts. The cluster analysis was performed using the statistical software R. The programs “dist” and “hclust” were used.

Fig. 4

Increasing and decreasing patterns in cell diameter of each cluster. Red indicates a change of median at each incubation time.

E. faecalis genomic DNA injection weakened the protoplast enlargement

The chi-square and Fisher's exact tests indicated that the protoplast enlargement patterns of the cells into which E. faecalis genomic DNA (Figs. 1C and 2C) was injected differed significantly (p < 0.05) from those of the control E. faecalis protoplasts (Table 2, Fig. 5). Thus, clusters 1 and 3 decreased, and cluster 2 increased (compared between "Control" and "E. faecalis" in Fig. 5), indicating that the protoplast enlargement was weakened by injecting its own genomic DNA into E. faecalis protoplasts.

Table 2

P values in chi-square and Fisher's exact tests.

Control vs.	Chi-square test	Fisher's exact test
Lactococcus lactis	0.599	0.616
Enterococcus faecalis	0.00463	0.00223
Lactobacillus curvatus	0.0489	0.0476
Bacillus subtilis	0.0143	0.0146
Escherichia coli	0.0148	0.0120
Lelliottia amnigena	0.0128	0.0123
Erythrobacter litoralis	0.226	0.309
Deinococcus grandis	0.0710	0.0883

Fig. 5

The number of protoplasts at each cluster of different bacterial genomic DNA injections. Chi-square and Fisher's exact tests were performed using the statistical software R. *, p < 0.05; G(+), Gram-positive; G(−), Gram-negative.

Our previous studies showed that E. faecalis protoplast enlargement requires DNA replication (Kami et al., 2019; Tsuchikado et al., 2020). DNA attachment to the plasma membrane (Earhart et al., 1968; Leibowitz and Schaechter, 1975; Marvin, 1968; Sueoka and Quinn, 1968) might be needed for enlargement. If so, the genomic DNA injection leads to enlargement promotion. In addition, the RNA expression from the E. faecalis genome may require enlargement. Therefore, the E. faecalis genome injection should promote enlargement. However, E. faecalis genomic DNA injection into E. faecalis protoplasts weaken the enlargement. Thus, these are considered unlikely possibilities.

In our opinion, the DNA replication system is associated with plasma membrane biosynthesis. Therefore, the injected E. faecalis genomic DNA may weaken DNA replication, and then plasma membrane biosynthesis may be weakened in conjunction with it. Furthermore, the injected E. faecalis genomic DNA may include fragmented DNA. This fragmented DNA may interfere with the replication of the E. faecalis intact chromosome. More work is needed to elucidate the interference.

Microinjection of four heterologous genomic DNAs weakened the enlargement of E. faecalis protoplasts

Microinjection of the Gram-positive bacterial genomes of Bacillus subtilis and Lactobacillus curvatus (Figs. 1D, 1E, 2D, and 2E) and the Gram-negative bacterial genomes of Escherichia coli and Lelliottia amnigena (Figs. 1F, 1G, 2F, and 2G) into E. faecalis protoplasts led to a weakened enlargement as well as that of the E. faecalis genome (p < 0.05; Table 2, Fig. 5). Conversely, microinjection of the Gram-positive Lactococcus lactis genome (Figs. 1B and 2B) and the Gram-negative Deinococcus grandis and Erythrobacter litoralis (Figs. 1H, 1I, 2H, and 2I) did not lead to enlargement patterns different from that of the control (p > 0.05; Table 2, Fig. 5). Therefore, considering that E. faecalis is Gram-positive, the effects of heterologous genomic DNA injection on protoplast enlargement are independent of bacterial evolutionary relationships.

The GC content of the E. faecalis genome is 37.5%. Only the L. lactis genome has the lowest GC content (35%) among the seven heterologous genomes. The effect of microinjection of the L. lactis genome on enlargement did not differ from that of the control (p > 0.05; Table 2, Fig. 5). The B. subtilis, E. coli, and L. amnigena genomes have a higher GC content than the E. faecalis genome. The effects of microinjection of these heterologous genomes on enlargement differed from those of the control (p < 0.05; Table 2, Fig. 5). These results suggest that genomic DNAs with GC content higher than that of the E. faecalis genome may function in the E. faecalis protoplasts. However, microinjection of the D. grandis and E. litoralis genomes with GC content much higher than that of the E. faecalis genome did not have different effects from that of the control (p > 0.05; Table 2, Fig. 5). Therefore, it is uncertain whether the GC content of microinjected genomic DNA affects protoplast enlargement or not.

When the injected DNAs are aggregated in the protoplasts, the injected DNA may not influence protoplast enlargement. When the injected DNAs are not aggregated, the DNA may function in the protoplasts. However, if the injected DNA hybridizes to the host E. faecalis chromosome, the chromosome replication may be weakened. This replication weakness may influence plasma membrane biosynthesis. Moreover, even if the injected genomic DNAs do not hybridize to the host chromosome, the heterologous genomic DNAs might reduce the replication efficiency of the host E. faecalis chromosome.

Conclusion

We generated more than 100 E. faecalis protoplasts into heterologous genomic DNAs. Protoplast enlargement did not stop and continued after the microinjection. The effect of the microinjected genome was observed, which was species-dependent but did not reflect bacterial phylogeny. The effect is that microinjection of genomic DNA weakened enlargement. We clarified the difference in the effect of the injected genomic DNA on enlargement, but it is uncertain how the injected DNA exists in the host cell. More work is needed to elucidate whether the DNA sequence plays an important role in the weakness of protoplast enlargement.

Credit Author Statement

Sawako Takahashi: Validation, Formal analysis, Data curation, Writing-original draft

Hiromi Nishida: Conceptualization, Formal analysis, Data curation, Writing-original draft, Writing-review and editing, Supervision, Project administration.

Declaration of Competing Interest

The authors declare that they have no competing interests.