Senior supervisor: Benno Ter Kuile
Supervisor: Tania Darphorn
SILS
Eva Kozanli
Student ID:11612770
Bachelor Biomedical Sciences
University of Amsterdam
Incompatibility and transfer genes can
affect transfer rates in conjugation of
extended
spectrum
β-lactamase
1
Conjugation is a form of horizontal transfer in bacteria that contributes to the
spread of antibiotic resistance, among others. For conjugation to happen, contact
between donor and recipient cells is necessary for the cell-to-cell transfer of
plasmids. Several genes are involved in this process, including Incompatibility
genes and transfer genes. There are indications that these genes are essential for
the transfer rate that differs greatly among plasmids. Because this circular DNA
can evolve rapidly, complex evolutionary aspects and genetic features are
associated with them. The mechanisms that drive these features are yet to be
fully understood. Experiments are conducted in Escherichia coli using a strain that
contains antibiotic resistance, ESBL2082. Determination of the minimal inhibitory
concentration, sequence analyses and literature led to the conclusion that
Incompatibility genes and transfer genes can influence the transfer rate in
conjugation.
Introduction
Antibiotic resistance plasmids can be spread among bacteria via three different paths. Transformation, transduction and conjugation are processes of bacteria for sharing DNA. (1) This can be established between bacteria of the same species and between species that differ in ecologically or evolutionary origins (2, 3). Conjugation works through contact between bacteria. When bacteria that contain a specific plasmid get in contact with a recipient, the pilus is generated. This forms the bridge between the donor cell and its recipient through which the plasmid is able to be transported. For this to happen, the circular plasmid DNA needs to be unfolded by helicases that are encoded on the plasmid (1). Known genes that are necessary for the mobility in the transfer of plasmids are Transfer (tra) genes. Tra genes on the plasmid DNA can encode proteins for pilus assembly, unwind the DNA and act for example as a promoter (1). Other genes that are located in plasmids are accessory genes that hold resistance against antibiotics and toxins for example that can improve the survival rate of bacteria (4).
One way to increase the survival rate can be antibiotic resistance. Extended-spectrum β-lactamases (ESBL) are a known example of plasmid-associated antibiotic resistance and these are genes that can inactivate β-lactam antibiotics and have an even wider resistance range (5). That is why they are referred to as extended spectrum. Antibiotics, such as penicillin and ampicillin, can be broken down by the plasmids encoding enzymes. β-lactam antibiotics normally bind to transpeptidases that reside in bacteria. Their function is to form crosslinks between alanine and lysine in the bacterial peptidoglycan layer. Without these, cells are not able to maintain a stable cell wall. This means that the cell wall will collapse and cells will die. These β-lactam antibiotics are therefore bactericidal antibiotics, meaning that they kill bacterial cells that they attack. Other antibiotics are categorized as bacteriostatic, meaning they inhibit the growth of the bacteria they target.
There are several β-lactamase subtypes discovered, such as TEM, SHV and CTX-M (6, 7). In this study the detected β-lactamases on plasmid isolates are originating from cattle and poultry. Since these are part of the food chain, chances are that the existence of these plasmids forms a thread when the meat of these animals is used for consumption. As transfer of the plasmids from bacteria to bacteria is possible, human consumption of meat could potentially lead to resistance in humans (7). This resistance is harmful as β-lactam antibiotics are widely prescribed in human healthcare and can obstruct a broad
2 spectrum of bacteria (7). Therefore, knowledge of mechanisms underpinning the transfer of this plasmid can aid in prevention of the spread of these resistance genes.
This study will focus on transfer rates in conjugation, essential genetic features within the transferable plasmids and variable regions and mobile elements in it. Parts of these aspects are discussed in other studies which will be referred to in the literature study embedded in this report. However, many details are still not comprehended. What is known from previous work is that Incompatibility (Inc) genes are a set of genes that are represented in plasmids and exist in different subtypes (8). The Inc genes are regulatory but cannot regulate plasmids of another subtype, hence, the name incompatibility (8). Another previously mentioned group of genes that play a role in transfer, are the tra genes. Although it is suggested that both of these groups are involved in transfer of plasmids, its actual function without these genes is not demonstrated. Studies tried to group the different types of plasmids (9). In the literature of this study there will be looked into the plasticity of plasmids which complicates grouping of the plasmids. Besides the transfer of plasmids that causes different species to share traits, plasmids themselves do also contain variable genes. Parts of plasmids can be changed by the introduction of mobile elements (10, 11). Multiple genes are known for their ability to move a part of the DNA from one plasmid to another (10, 11, 12). Beside these embedded genes, factors from the outside are able to invade the circular and chromosomal DNA, like bacteriophages (12). Previous research showed that for E. coli only 10% of the genome is intergenic, meaning that it is present in all of the genomes of E. coli (13). The other 90% of the genetic features in bacteria are variable within species. The existence of mobile elements that can be inserted to the genome and the presence of plasmids in the bacteria add to their variability. These aspects can even help to change the cells in their surroundings (14). Conjugation is a strong example for this, but not every plasmid can be transferred by themselves (14). The question is what makes them prone to being transferred in conjugation while other plasmids stay in place. To get grip on the complexity of plasmids, more insights regarding the evolutionary aspects of plasmids can be meaningful. Since plasmids cost bacteria a lot of resources and energy, the benefits should outweigh these costs (10). Plasmids are less stable than chromosomal DNA and there are mobile elements that can invade chromosomal DNA (10). So exactly why plasmids and other mobile elements are beneficial to bacteria and why plasmids are able to evolve alongside chromosomal DNA, remains unclear. In addition to this, changes in plasmid sequences are observed within one transfer from cell to cell (13, 15). This phenomenon called gene editing seems illogical and it is unclear what underpins these DNA changes.
It is important to provide a more coherent view on this field of research. That is why this report will collect relevant proposals from previous work and adding own experiments contributes to this. Providing a closer look to the sequences of the plasmids of these bacteria before and after conjugation, can contribute to understanding the mechanisms that drive conjugation. For various genes carried on the transferable plasmids, this study tried to find what their influence in conjugation is on the transfer rate of plasmids carrying extended-spectrum β-lactamase resistance in E. coli. It is thought that some genes are essential in transfer and can influence it by making conjugation more efficient by different means. In order to recognize the role of the plasmids’ genes, literature is studied and compared to the detected genes in the sequences of the transferred plasmids in these experiments. The experiments that are conducted should display variation in plasmids, the transfer rate of different plasmids and the elements that reside on these plasmids. With this, a more complete picture of the roles of genes on the plasmids is sought.
3
Materials and method
Bacterial strain and antibiotics
In this study the E. coli strain MG1655 served as acceptor (kindly provided by the stock collection of the University of Amsterdam). This strain contained a chloramphenicol resistance gene in its bacterial chromosomal DNA. E. coli strains containing the ESBL2082 isolates from poultry functioned as donor (kindly provided by the stock collection of the University of Amsterdam). It was known that this strain contained ampicillin and tetracycline resistance genes on one of its plasmids. Ampicillin, chloramphenicol, tetracycline and kanamycin stock solutions (10µg/ml) were sterilized using a 0,2 µm filter.
Bacterial cultivation
Batch cultivation was applied to grow the two bacterial strains at 37˚C. The bacteria were grown overnight in Evans medium (16, supplementary data) and plated out on Luria-Bertani (LB) medium plates with selective antibiotics (supplementary data). All strains provided were stored in a 60% glycerol solution (supplementary data) at -80 ˚C.
The colony experiments
The first part of this study was conducted to examine the transfer rate and gene transfer in conjugation of the strains. The MG1655 acceptor strain was combined with the ESBL2082 for conjugation. To execute the conjugation experiment, the strains were first grown overnight in Evans medium with glucose separately. Antibiotics were added to this medium for growth (64 µg/ml). For MG1655 chloramphenicol was added and for ESBL2082 ampicillin. The succession of conjugation was determined with these antibiotics, in a later phase of the experiments. After overnight culturing, the cells were depleted from glucose by centrifuging for 15 minutes on 4400 rpm. Evans medium without glucose was added to the pellet and the cells were incubated at 37˚C for 4 hours. After starvation at an OD600 of 0.25 of both solutions is added for transfer in Evans without glucose. The bacteria in transfer are placed in a 37 ˚C stove at 200 rpm.
The transfer is plated out after 1 hour and again after 24 hours in different dilutions. After this, the transfer was terminated. Selective plates that were worked with to distinguish transconjugants from donors and acceptors. To visualize the transconjugants, plates containing LB medium, ampicillin and chloramphenicol were required. Plates containing LB medium and either ampicillin or chloramphenicol displayed respectively for the acceptor and donor strain.
4
MIC measurement
The colonies that were located on the selective transconjugant plate were compared to establish variation. All colonies on the selective plates with transconjugants were grown separately in Evans medium with both ampicillin and chloramphenicol and stored in -80˚C with the 60% glycerol solution. To verify a rough difference in conjugation, minimal inhibitory concentrations (MIC) were determined. Differences in resistance for ampicillin, chloramphenicol, tetracycline and kanamycin were assessed for each colony. For MIC determination, these antibiotic resistances were measured in a starting concentration of 2048 µg/ml and two times diluted in every new well. Only tetracycline had a deviating starting concentration of 512 µg/ml, because earlier experiments showed that tetracycline is effective in lower doses (17). All wells contained an OD600 of 1,55 of the bacterial suspension except for the negative control, which existed of only Evans medium to check for contamination. The positive control contained bacterial suspension as well as Evans medium to compare normal growth with growth under influence of antibiotics. The wells plates were incubated in the Multiskan™ FC Microplate Photometer plate reader at 37 ˚C to measure bacterial growth. This measurement had 138 readings of 10 minutes at a wavelength of 595 nm. The wells plates were shaken between every reading to prevent the bacteria from sinking to the bottom of the wells. A minimal absorbance of 0,15 was considered as bacterial growth.
Sequence analysis
The plasmids of the ESBL2082 strain and transconjugant were isolated using the plasmid isolation kit of QIAGEN with ethanol precipitation. Isolation was performed at the Swammerdam Institute of Life Sciences (SILS, Amsterdam). This resulted in 10000 ng isolated plasmid DNA of the donor strains. These samples were sequenced (BaseClear, Leiden) using Illumina NovaSeq for short read sequences as well as using PacBio for long read sequences. The results were combined for analysis creating a hybrid assembly. The received sequences of the ESBL2082 donor strain were compared to their transconjugants. Alignment of the sequences showed the parts of the plasmid DNA that were overlapping in the two.
The same isolation for ESBL2082 was conducted before. These isolates were sequenced with only Illumina sequencing. This sequence data was analyzed apart from the hybrid data to look at differences between the results.
Firstly, resistance genes and Inc genes were identified using ResFinder and PlasmidFinder. Identifying the Inc genes helped in determining the contigs that form a separate plasmid, since these are rarely located on one plasmid. In this analysis, sequences were aligned (NCBI BLAST, SnapGene 5.1.1) to see whether the transconjugant originated from only one plasmid of the donor or multiple. The traits on these plasmids were inspected and research was performed to collect more information on genes that could affect transfer through conjugation. To be able to fully explain these results, several alignments were made: one between the hybrid data and transconjugant, one between Illumina data and transconjugant and one between hybrid and Illumina data.
5
Results
The colony experiments
Transfer of plasmids through conjugation was realized in 1 attempt for ESBL2082. This was shown on the selective LB agar plates after 1-hour transfer as well as 24 hours of transfer. The plates that included ampicillin and chloramphenicol showed transconjugants. The plates containing one of these antibiotics were examined to calculate the rate of transconjugants in contrast to the donor and acceptor colonies. Dilutions revealed a variating count for the transconjugants per ml (see table 1 in supplementary data). The mean of two dilutions is therefore used to provide a more precise calculation of the conjugation efficiency.
Table 1. Conjugation efficiency: The conjugation efficiency is shown together with relevant calculations. A calculation is made to change the number of colonies for every plate into the number of colonies per mL by multiplying for the dilution and dividing by the volume that is pipetted on each plate. From this number, resulted the ratio between donor and recipient. Because transconjugant plates showed colonies on two differently diluted plates, both together form the mean transconjugants for calculating the mean conjugation efficiency, which is explained by the number of transconjugants per donor cell.
Number of donors/mL Number of recipients/mL Donor to recipient ratio Number of transconjugants/mL Mean conjugation efficiency 6,40 ·107 4,00 · 106 16 : 1 1,15·104 1,77 · 10-4 6,20·103
MIC measurement
Variation of resistance levels between the colonies on the transconjugant plates occurred, according to the MIC results. All colonies after 1-hour transfer (n=15) and after 24 hours of transfer (n=9) are used in MIC measurement and presented variable results in resistance against antibiotics (see figure 2). To process the results, the most frequently appeared minimal inhibitory concentration at both time points in the colonies is compared. To indicate variation, the percentage of the most frequently observed value is displayed. Ampicillin resistance occurred in high concentrations in all of the colonies regardless of the duration of transfer. This was even higher than the highest concentration of 2048 µg/ml with no observed variation. For chloramphenicol the most frequently detected concentration was 256 µg/ml (for 60% of the colonies). After 24 hours this was still 256 µg/ml (55,6%). Tetracycline showed low resistance after 1 hour with a value of 4 µg/ml (40%). After 24 hours this was at a much higher value of 256 µg/ml (77,8%). Kanamycin colonies mostly started at 32 µg/ml (60%) and after 24 hours this moved towards 128 µg/ml (66,7%). For some antibiotics concentrations and therefore also resistance levels varied, while others remained more or less constant.
6
Figure 2. antibiotic resistance: This graph shows the minimal inhibitory concentration that was observed for respectively ampicillin (A), chloramphenicol (B), tetracycline (C) and kanamycin (D) concentrations added to the different transfer colonies. The blue bars show the percentage of total colonies after the 1-hour conjugation (n=15) and their resistance level. The green bars show the percentage of total colonies after 24 hours conjugation (n=9) and their resistance level.
0 20 40 60 80 100 4 8 16 32 64 128 256 512 1024 2048
Percentage of total colonies (%)
Mi ni m al inh ibi tory c on cent rat ion ( µg/m l)
Kanamycin resistance
24 hours 1 hour 0 20 40 60 80 100 1 2 4 8 16 32 64 128 256 512Percentage of total colonies (%)
Mi ni m al inh ibi tory c on cent rat ion ( µg/m l)
Tetracycline resistance
24 hours 1 hour 0 20 40 60 80 100 4 8 16 32 64 128 256 512 1024 2048Percentage of total colonies (%)
Mi ni m al inhi bi tor y conc ent rat ion (µg/m l)
Chloramphenicol resistance
24 hours 1 hour 0 20 40 60 80 100 4 8 16 32 64 128 256 512 1024 2048Percentage of total colonies (%)
Mi ni m al inhi bi tor y conc ent rat ion (µg/m l)
Ampicillin resistance
24 hours 1 hour7
Sequence analysis
The sequence analyses started with investigating the circular DNA results apart from each other in the SnapGene, PlasmidFinder and ResFinder database (see figure 2AB in supplementary data). This showed one plasmid for the transconjugant containing an IncX4 gene, four plasmids in the hybrid assembly of the donor strain with one of them also containing the IncX4. This hybrid possessed a total of six different Inc genes where the first and second plasmid both contained two Inc genes, the third contained the IncX4 gene and the fourth IncFIB, which is a subtype originating from bacteriophages. The Illumina sequences gave less extensive results with also four plasmids but with only four of the Inc genes being detected (see table 3).
All tra genes were noted to compare in further literature. In every plasmid, except for the bacteriophage plasmid, tra genes were detected (see table 4ABC). The number of tra genes varied from plasmid to plasmid. The overview shown in table 4 provides all identified plasmids which are the ones that did contain Inc genes. Because Illumina data failed to assembly all parts of the sequences, information regarding the plasmids’ genes can be absent.
Analysis of the isolated plasmids of the ESBL2082 strains resulted in differences when comparing the Illumina sequences to the transconjugant, contrasted to the results of the hybrid data and the transconjugant. This led to the expectation that the two would confirm the results of conjugation as both are isolates from the same donor strain. For that reason the Illumina and hybrid data were compared as well. Differences in the Illumina and hybrid data sequences were discovered as well. The alignment between the hybrid sequence and the sequence the transconjugant’s plasmid, presented no relevant genetic differences and resulted in similar sequences. The start of the sequences was adapted to show the alignment of the two initiating at the same gene (see supplementary data figure2AB).
In the overview in table 4, the resistance genes are also shown. These genes were relevant to determine which plasmids had been transferred on the selective plates and in MIC measurement. Several types of resistance genes that were discovered, were: β-lactams (blaSHV-12, blaTEM-1B), aminoglycosides (aadA1, aadA2b, aadA24), tetracyclines (tet(A)), sulphonamides (sul3), phenicols (cmlA1) and macrolides (Inu(G)).
Table 3. plasmid analysis: this table shows the number of plasmids and their Inc genes per sequence method and strain.
ESBL2082 ILLUMINA ESBL2082 HYBRID ESBL2082
TRANSCONJUGANT
NUMBER OF PLASMIDS 4 4 1
NUMBER OF REPLICONS IncI1-I(), IncX1, IncX4, IncFIB (phage plasmid)
IncI1-I(), IncX1, IncFII, IncFIB, IncX4, IncFIB (phage plasmid)
IncX4
NUMBER OF RESISTANCE GENES
8
Table 4ABC. Overview of Inc, tra and resistance genes in plasmids: The different genes from the hybrid data (A), Illumina data (B) and transconjugant (C) are shown. For the Illumina data not all information can be shown, since assembly information is missing in this method.
A. Hybrid ESBL2082
PLASMID NUMBER INC GENE SUBTYPES TRA GENE SUBTYPES RESISTANCE GENES
1 IncI1-I
IncX1
traA, traB, traC, traG, traH, traI, traJ, traR, traS, traU, traW
aadA1, aadA2b, blaSHV-12, blaTEM-1B, sul3, tet(A), cmlA1 2 IncFII IncFIB
traA, traB, traC, traD, traE, traF, traG, traH, traI, traJ, traK, traL, traM, traN, traP, traQ, traR, traS, traT, traU, traV, traW, traX, traY
none
3 IncX4 traE, traR aadA24, blaTEM-1B,
Inu(G)
4 IncFIB (phage) none none
B. Illumina ESBL2082
C. Transconjugant ESBL2082
PLASMID NUMBER INC GENE SUBTYPES TRA GENE SUBTYPES RESISTANCE GENES
1 IncX4 traE, traG, traR none
2 IncX1 traG, traR blaTEM-1B, tet(A)
3 IncI1-I traA, traB, traC none
4 IncFIB (phage) traA, traB, traC, traG,
traH, traI, traJ, traR, traS, traU, traW
none
PLASMID NUMBER INC GENE SUBTYPES TRA GENE SUBTYPES RESISTANCE GENES
1 IncX4 traE, traG, traR aadA24, blaTEM-1B,
9
Literature study
Process of conjugation
Conjugation is an important mechanism for bacteria to transfer genes and it even seems to be the most frequently occurring form of horizontal gene transfer (18). There are multiple conjugation transfer systems known. For example, for the regularly present F-like (fertility-like) system at the start of conjugation, the DNA that is going to be transferred, is replicated in the donor cell. In most cases, the DNA is nicked on a single strand on the origin of transfer (1). The other strand will then be replicated before transfer to the other, recipient cell. Conjugation is mediated by several proteins that are able to form the pilus through which the plasmid is able to travel between cells. Essential to bacteria in conjugation are DNA transfer proteins, genes that encode for proteins in the pilus and proteins that link these two together (1). The set of proteins that a plasmid, capable of conjugation, possesses, can differ. Having these genes, partly determines whether conjugation is possible, but there are plasmids known that do not have these genetic features and yet are able to be transferred when another plasmid initiates the forming of a pilus (14). This indicates the complexity of conjugation.
For some bacteria, initiation of the conjugative complex happens due to pheromones, small peptides that are sensed by the recipient cells. For gram-negative bacteria such as E. coli it was thought for a long time that conjugation could only occur when an F-plasmid (fertility-plasmid) was present in the cell. The F-plasmid contains a factor that is able to induce the forming of the pilus. However, more recently formation of other pili was observed in bacteria (9). This categorization of plasmids on the presence of the fertilization factor or other genes such as resistance genes, resulting in a R-plasmid (resistance-plasmid), did not seem sufficient anymore (9).
In all sequenced plasmids in this study, a diversity of Inc and tra genes are detected. Since the inhibitory concentration differed greatly across tested colony, various resistance genes may have been present in the colonies in MIC determination. This could be an indication that plasmids were only in some cases able to be transferred and were more often transferred after a longer period of time. This could suggest that different plasmids within the cells were able to be transferred but not all at the same rate. Since various Inc and tra genes were encoded in these plasmids, this could possibly aid the transfer rate. If this is in fact the case it can be proposed that multiple plasmids can be transferred within one cell and then the set of incompatibility genes might not exclude other plasmids from being transferred.
Essential genetic features
Since all plasmids do contain a set of Inc genes, they can give an indication of the number of plasmids that are present in a cell. Online databases use this information to distinguish the plasmids in plasmid finders. Many different Inc genes are discovered and most of their plasmids are assumed to be incompatible with each other. Incompatibility here does not imply that the genes do compete with each other in a way that is lethal to the other plasmids, but that the genes that are included in the set of Inc genes encode for regulatory proteins needed for that plasmid, which other plasmids cannot operate with (8). Competition until it is lethal to other plasmids may not be the case, since plasmids of other Inc groups are existing in one cell on a regular basis (9, 18). However, research does argue that Inc genes compete for fitness (4). Some Inc groups are more likely to be transferred than others, giving them a benefit with respect to other plasmids (18). The plasmids can be grouped in Inc categories based on the mechanisms of replication that they have (9). These replication mechanisms are linked to the transfer of these plasmids as well and therefore can be a strong indicator for how their conjugative systems operates (9).
10 The Inc genes found in this study in ESBL2082 are IncX1, IncX4, IncFIB and IncI1-I. This mix of different Inc genes is useful for analyzing the transfer of plasmids. IncX is a commonly encountered group in plasmids, particularly in cattle, such as chicken (18). Of the IncX group, IncX4 is the most prevalent (18, 19). IncX4 is known for transferring antibiotic resistance from one cell to another (18, 19). The resistance gene CTX-M β-lactamase is one gene that is frequently carried by the IncX4 plasmid (18, 19). In this study the transconjugant contained this IncX4 plasmid together with a β-lactamase of the TEM type. According to literature, transfer of this plasmid is not rare and this can cause high prevalence of the plasmid (19). As it is easy to transfer this type, its popularity in bacteria increases more rapidly.
As stated earlier, there may also have been transfer of other plasmids in the MIC experiment, since the resistance between different colonies was fluctuant. Literature provides little clarity about the IncFIB subtype, but research showed it being spread in bacteria in clinical settings (20). These IncFIB plasmids were also containing ESBL genes, which posed great danger for the hospital where it has been localized (20). More is known about the IncF type itself and these IncF subtypes can vary a lot from each other. Usually the copy number of these plasmids is low and the plasmids are typically longer than 100 kilobases, which is rather large (2). These findings suggest that these are not easily transferred since longer plasmids can take longer to be transferred and with low copy numbers the chances on conjugation seem lower. Since the IncFIB plasmid of ESBL2082 did not contain resistance genes, results of the MIC could not tell whether this plasmid is transferred or not.
IncX1 and IncI1-I are located together on the first plasmid of ESBL2082. This might be due to a mistake in assembly, since a combination of these genes together is rarely observed and the Illumina data suggests that these genes are on two separate plasmids. Because the sequences of these two in the Illumina data seem to contain sequences that are similar, a fusion could have taken place. To remain concrete, the plasmid is reviewed as one as it was identified as one. This plasmid contains most other resistance genes of the ESBL2082 strain and is therefore thought to be transferred in some colonies in the MIC experiment. The IncI1 gene set is one that has a narrow host range, but it is seen commonly in transfer and often includes resistance genes (18).
In addition to Inc genes, tra genes play an important role in conjugation. Various subtypes of these genes are found in the plasmids of ESBL2082 in this study. The genes are thought aid in certain processes of conjugation such as pilus assembly and production and DNA transfer, where the majority is involved with the firstly mentioned (1).
Tra genes that were located in the investigated plasmids are showed previously in the overview of table 4. What stands out is that the number of tra genes in the plasmids, does not determine how easily it is transferred. It is expected that with a high number of tra genes, it becomes less complex to be transferred, since all needed instruments to promote transfer, seem present. The tra genes in the transconjugant plasmid are traE and traG, which aid in pilus assembly, and traR, which is a transcription activator (1). Two of these are also represented in the hybrid and Illumina data before conjugation as well. Tra genes are thought to be helpful in conjugation, but this seems not to be measured by the number of tra genes they possess. It is possible that their significance differs per plasmid and other genes that reside on it.
Gene editing
Gene editing is an event that is observed in some recent studies about plasmids (22). The plasmids’ sequences vary sometimes before and after conjugation. How this occurs cannot yet be explained. Whether this is induced by the plasmid or one of the cells involved, is unclear. What is known is that both can benefit from this change because this phenomenon can either be a loss or gain in genes. As discussed earlier, a loss of genes can be beneficial for the transfer of the plasmid and less costs for the
11 cell, but gaining a gene can give the plasmid as well as the bacteria beneficial traits (4). The study by Modi et al. (1992) showed that parts of plasmids changed as a consequence of transfers in competition. The loss of tetracycline genes led in this study to an high increase in fitness with respect to surrounding plasmids (22). On the long term and in the presence of tetracycline this could of course mean the opposite.
In the MIC experiment that is conducted, similar events could have occurred. Since only the resistance cannot tell what plasmid has been transferred, conjugation of multiple plasmids and gene editing are both still open for option. Gene editing can improve plasmid and bacterial survival, but this can become a disadvantage for people and animals. When genes can evolve as rapidly as one single transfer, this could be a quick way of gaining resistance genes or other genes that favor pathogenic bacteria.
Evolutionary aspects
Old classification systems for plasmids tried to divide plasmids into groups depending on the genetic features these plasmids had (9). Because of the evolutionary aspects that come with plasmids, it is hard to divide them into categories. Traits, like resistance, can be taken from one plasmid to another and as mentioned earlier, gene editing can change the plasmids, what leads to spreading of groupable traits. Multiple studies have looked into the backbone genes (23, 24), because most of these genes will not be easily transferred from one plasmid to another, since regulatory proteins do not present great advantages compared to their costs.
Classification by Inc genes provides a new method of grouping that several studies looked into (23, 24). Since most of these gene sets possess their own qualities, this seems more suitable. Nevertheless, differences between Inc types and subtypes exist and therefore groups will not entirely explain the features of plasmids that belong to this group (1).
It is hard to explain whether introduction of plasmids and other mobile elements is a strictly evolutionary process (4). Multiple theories have been formulated, suggesting that mobile elements act strictly evolutionary (25). One poses that it is selection driven and only genes that are favorable manage to be transferred and conserved (26). Other theories state that this is a random process, where the elements are seen as selfish entities that only act for their own survival (27). Both theories can be argued (25). When it is claimed that only favorable genes will be conserved, this would mean that all non-beneficial traits will be disposed if not needed. Since mobile elements are sometimes characterized by the ability of integrating the genome and if all beneficial features were selection driven (25), this would mean an end to plasmids as relevant genes can be integrated into the chromosomal DNA. Embedding into chromosomal DNA would be a logical step, because this would lead to loss of the regulatory genes in the plasmid DNA and would in this way be less costly to the bacteria. These arguments would lean more towards the selfish entity theory, but there are flaws to this theory too. If the plasmids would selfishly transfer at rates as high as possible, this would mean loss of many accessory genes. The plasmid would get rid of accessory genes, because these lead to longer DNA sequences. These accessory genes would drive the need for more materials of the cell to replicate the plasmid, which would only interfere with a high transfer rate.
Research discovered that even without positive selection costly plasmids are not lost efficiently, which indicates that the plasmids can be somewhat selfishly evolving (4). However, maintenance of plasmids is costly and in many situations plasmids benefit their host. The relation between bacteria and plasmids can therefore be seen as symbiosis, where coevolution seems a more suitable term than evolution.
12
Discussion
In this study’s experiments, LB plates containing multiple colonies were used in MIC measurement to look at differences. MIC results showed variating resistance levels in the transfer time spans of E. coli ESBL2082. When sequences were analyzed, these changes in resistance were not genetically visible for the transconjugant. The one plasmid that was transferred, did not contain tetracyclineresistance, while resistance to this was detected in several colonies in MIC measurement. The transferred IncX4 plasmid did also contain three tra genes: traE, traG and traR. All of these were also observed in other plasmids that were not transferred to the sequenced transconjugant. The MIC data suggests the possibility that after a longer period of time, other plasmids are transferred as well. Another option is that a selection of genes originating from other plasmids transported to the transconjugant by gene editing. A conclusion is drawn from the research outcomes and the additional literature study.
Genes are able to influence transfer via conjugation. The set of incompatibility genes seem to be crucial for conjugation. The importance of transfer genes is stated in literature but not yet shown by the experiments. Literature acknowledges that certain Inc genes are found more frequently involved in conjugation than others. The experiments conducted for this study, supported this. Transfer of the IncX4 plasmid occurred more rapidly than the transfer of other plasmids.
Methods to determine the transfer of plasmids or gene editing did not entirely suffice. With a higher number of colonies on the transconjugant plates present, more can be concluded from the results. For calculating the efficiency, it is common to use higher numbers of transconjugants per plate (around 100 to 300). To be able to draw full conclusions including a transfer in triple can be useful. Measuring resistance with the use of MIC determination is not precise enough to conclude this from but can be a strong indicator. In addition, it is cost and time efficient to determine what further research should focus on. The sequence analysis is interesting, since this can give exact locations for where changes occurred. However, to get a better overview on whether one plasmid was transferred or gene editing took place, data of one colony is not enough. Because this is a costly analysis, MIC and Illumina results are collected before sequencing the plasmid isolates. However, the Illumina sequence data gave dubious results that needed to be confirmed with the hybrid sequence data. Afterwards this additional data was very helpful as the Illumina data alone gave different results that were not reliable. However, the Illumina did help when assembly of the hybrid data failed in splitting two different plasmids. Both methods provided different information. This underlines yet again the relevance of using good methods to come to findings.
Earlier research already suggested involvement of tra genes, which explains the chosen name (1). Inc genes were linked to conjugation as well, but definite proof is still hard to find on this. Although awareness of plasmids and their variety exists for a long time, their role in antibiotic resistance raised interest in this subject again. For example, a better understanding of the genes involved in conjugation can help in improving prevention against the spread of antibiotic resistance. Furthermore, understanding bacteria is necessary since humans live with bacteria, pathogenic as well as commensal, on a daily basis and these can impact human health heavily.
Further experiments can help explaining when and how gene editing occurs, what factors do make plasmids prone to be transferred and what not and the extent to which genes are of importance to this. The methods used here do not provide complete answers yet but form a step towards understanding the behavior of bacteria and plasmids. This can help in treatment and this is what should be aimed for in further research.
13 Why transfer of plasmids occurs quickly, slowly or sometimes cannot occur is dependent on multiple factors within and outside of the plasmid. This paper mainly focused on the intrinsic factors. Because of the complexity of both intrinsic and extrinsic factors together, much of the transfer via conjugation remains unknown. Several genes that are involved are highlighted in this study, but none of the results are conclusive in displaying the exact effects of the single genes. There are other genes discovered in bacteria that potentially have a role in transfer. These gene groups differ per plasmid from type to type. All of these additional gene groups such as virB in IncX plasmids, can be of interest for further research (28). To identify the effects of single genes, more research needs to be conducted on separate genes involved. In follow-up research, CRISPR-cas9 or knockout genes can reveal the effects of editing single genes within one plasmid. Because the plasmids are equal and will be maintained in a stable environment, impact on the transfer rate can be measured independently. Editing tra genes, Inc genes and other potential genes can cause better understanding in the exact contribution of this set of genes to conjugation. Insights on how transfer rates can be influenced and what type of plasmids need to be paid attention to, can help in stopping dangerous traits from being spread among species and even counteracting them. Great steps have already been made towards understanding and these will contribute to public health enormously.
Want to know more about the proposal for follow-up research? The proposal will be made public at the end of July 2020.
14
Literature
1. Zatyka, M., & Thomas, C. M. (1998). Control of genes for conjugative transfer of plasmids and other mobile elements. FEMS microbiology reviews, 21(4), 291-319.
2. Kruse, H., & Sørum, H. (1994). Transfer of multiple drug resistance plasmids between bacteria of diverse origins in natural microenvironments. Appl. Environ. Microbiol., 60(11), 4015-4021.
3. Cottell, J.L., Webber, M.A., Piddock, L.J. (2012). Persistence of transferable extended-spectrum-beta-lactamase resistance in the absence of antibiotic pressure. Antimicrob Agents Chemother, 56:4703–4706.
4. Harrison, E., & Brockhurst, M. A. (2012). Plasmid-mediated horizontal gene transfer is a coevolutionary process. Trends in microbiology, 20(6), 262-267.
5. Livermore, D. M. (2008). Defining an extended‐spectrum β‐lactamase. Clinical Microbiology and Infection, 14, 3-10.
6. Ali, T., ur Rahman, S., Zhang, L., Shahid, M., Han, D., Gao, J., et al. (2017). Characteristics and genetic diversity of multi-drug resistant extended-spectrum beta-lactamase (ESBL)-producing Escherichia coli isolated from bovine mastitis. Oncotarget, 8(52), 90144.
7. Händel, N., Otte, S., Jonker, M., Brul, S., & ter Kuile, B. H. (2015). Factors that affect transfer of the IncI1 β-lactam resistance plasmid pESBL-283 between E. coli strains. PloS one, 10(4). 8. Novick, R. P. (1987). Plasmid incompatibility. Microbiological reviews, 51(4), 381.
9. Arutyunov, D., & Frost, L. S. (2013). F conjugation: back to the beginning. Plasmid, 70(1), 18-32.
10. Rankin, D. J., Rocha, E. P., & Brown, S. P. (2011). What traits are carried on mobile genetic elements, and why?. Heredity, 106(1), 1-10.
11. Frost, L. S., Leplae, R., Summers, A. O., & Toussaint, A. (2005). Mobile genetic elements: the agents of open source evolution. Nature Reviews Microbiology, 3(9), 722-732.
12. Toussaint, A., & Merlin, C. (2002). Mobile elements as a combination of functional modules. Plasmid, 47(1), 26-35.
13. Touchon M, Hoede C, Tenaillon O, Barbe V, Baeriswyl S, Bidet P et al. (2009). Organised genome dynamics in the Escherichia coli species results in highly diverse adaptive paths. PLoS Genet 5: e1000344.
14. Sørensen, S. J., Bailey, M., Hansen, L. H., Kroer, N., & Wuertz, S. (2005). Studying plasmid horizontal transfer in situ: a critical review. Nature Reviews Microbiology, 3(9), 700-710. 15. Wozniak, R. A., & Waldor, M. K. (2010). Integrative and conjugative elements: mosaic
mobile genetic elements enabling dynamic lateral gene flow. Nature Reviews Microbiology, 8(8), 552-563.
16. Evans, C. G. T., Herbert, D., & Tempest, D. W. (1970). The continuous cultivation of microorganisms. 2. Construction of a chemostat. Methods Microbiol, 2(277), e327.
17. Gibbons, S., & Udo, E. E. (2000). The effect of reserpine, a modulator of multidrug efflux pumps, on the in vitro activity of tetracycline against clinical isolates of methicillin resistant Staphylococcus aureus (MRSA) possessing the tet (K) determinant. Phytotherapy Research: An International Journal Devoted to Pharmacological and Toxicological Evaluation of Natural Product Derivatives, 14(2), 139-140.
18. Lo, W. U., Chow, K. H., Law, P. Y., Ng, K. Y., Cheung, Y. Y., Lai, E. L., & Ho, P. L. (2014). Highly conjugative IncX4 plasmids carrying blaCTX-M in Escherichia coli from humans and food animals. Journal of medical microbiology, 63(6), 835-840.
15 19. Sun, J., Fang, L. X., Wu, Z., Deng, H., Yang, R. S., Li, X. P., ... & Liu, Y. H. (2017). Genetic analysis of the IncX4 plasmids: implications for a unique pattern in the mcr-1 acquisition. Scientific reports, 7(1), 1-9.
20. Gonullu, N., Aktas, Z., Kayacan, C. B., Salcioglu, M., Carattoli, A., Yong, D. E., & Walsh, T. R. (2008). Dissemination of CTX-M-15 β-lactamase genes carried on Inc FI and FII plasmids among clinical isolates of Escherichia coli in a university hospital in Istanbul, Turkey. Journal of clinical microbiology, 46(3), 1110-1112.
21. Villa, L., García-Fernández, A., Fortini, D., & Carattoli, A. (2010). Replicon sequence typing of IncF plasmids carrying virulence and resistance determinants. Journal of antimicrobial chemotherapy, 65(12), 2518-2529.
22. Modi, R.I., et al. Genetic changes accompanying increased fitness in evolving populations of Escherichia coli. Genetics, 130 (1992), pp. 241-249
23. Guglielmini, J., de La Cruz, F., & Rocha, E. P. (2013). Evolution of conjugation and type IV secretion systems. Molecular biology and evolution, 30(2), 315-331.
24. Smillie, C., Garcillán-Barcia, M. P., Francia, M. V., Rocha, E. P., & de la Cruz, F. (2010). Mobility of plasmids. Microbiol. Mol. Biol. Rev., 74(3), 434-452.
25. Syvanen, M. (1984). The evolutionary implications of mobile genetic elements. Annual review of genetics, 18(1), 271-293.
26. Stanley, S. M. (1975). A theory of evolution above the species level. Proceedings of the National Academy of Sciences, 72(2), 646-650.
27. Hickey, D. A. (1982). Selfish DNA: a sexually-transmitted nuclear parasite. Genetics, 101(3-4), 519-531.
28. Christie, P. J., & Vogel, J. P. (2000). Bacterial type IV secretion: conjugation systems adapted to deliver effector molecules to host cells. Trends in microbiology, 8(8), 354-360.
16
Supplementary data
Stock solutions for antibiotics
Stock solutions were made for ampicillin, chloramphenicol, tetracycline and kanamycin. Preparation of the antibiotics stock solutions for ampicillin, tetracycline and kanamycin was done by adding the antibiotic to 10 mL Milli-Q (10mg/mL). After the substance was completely dissolved, the solution was filter sterilized with a 0,2 µM filter into a sterile tube.
Preparation of the chloramphenicol stock solution (10mg/mL) used the same protocol but was added to 96% ethanol instead of Milli-Q. All of the solutions were kept at 4˚C and the tetracycline stock was kept in the dark.
Media compounds
Evans medium
Different media were used for tracking bacterial growth. Sterilized Evans medium, based on literature (1), was used as a minimal medium and could either contain 1M glucose to supply the bacteria for growth or not to create a minimal condition during transfer via conjugation. The Evans minimal medium consisted of 100 mM NAH2PO4.2H2O, 10 mM KCl, 1,2 M MgCl2.6H2O, 100 mM NH4Cl, 2 mM Titriplex
and 0,02 mM CaCl2.2H2O. Per 1 L of Evans medium 5 mL of the trace elements was added to it (for
further explanation see Trace elements). The pH is set on 6.9 by adding NaOH.
Trace elements
Trace elements are added to the medium in minimal concentrations. Some of these elements are not essential for the bacterial growth, but without these, the physiology of the cells could be alternated (16). Because of the low concentrations needed of these compounds a separation solution is made and later a small part of this is added to the Evans medium. The compounds in this solution and their concentrations are 12,6 mM ZnO, 50 mM FeCl3.6H2O, 25 mM MnCl2.4H2O, 3 mM CuCl2.2H2O, 5mM
CoCl2.6H2O, 2,5 mM H3BO3,0,04 mM Na2MoO4.2H2O, 20 ml HCl (37%). These are all added to Milli-Q.
LB medium
Sterilized Luria-Bertani (LB) medium with agar was used to grow the E. coli strains on plates. This nutritionally rich medium is effective in showing growth of bacteria overnight. Together with one or multiple selective antibiotics (64 μg/ml) the growth of either donors, recipients or transconjugants was displayed. 1 Liter of the LB medium consists of 170 mM NaCl, 5g yeast extract (Duchefa Biochemie, Haarlem, The Netherlands), 10g bacto tryptone (Brunschwig Chemie, Amsterdam, The Netherlands) and 20g bacteriological agar (Sphaero Q, Gorinchem, The Netherlands).
Glycerol solution
17
Raw data
Table 1. Colony count: Colonies were counted on the differently diluted plates containing antibiotics. The colonies were multiplied by the dilution to come to a value that presents the number of colonies without dilution.
CONTENT PLATE DILUTION COLONY COUNT COLONY COUNT
UNDILUTED DONOR 105 32 3,2 · 106 ACCEPTOR 105 2 2,0 · 105 TRANSCONJUGANT 100 575 5,8 · 102 TRANSCONJUGANT 101 31 3,1 · 102 TRANSCONJUGANT 102 0 0
Figure 2AB. Sequences corresponding plasmids: The figures show the third plasmid (IncX4) of the hybrid (A), and the transconjugant plasmid (IncX4) (B) which look similar. Purple displays functional genes, yellow mobile elements, pink Inc and tra genes, green the antibiotic resistance genes and black hypothetical genes
A.
