Introduction
λ has been widely studied for decades as a model bacteriophage, providing invaluable information on the inner workings of phages and their potential use in the modern world. In spite of this thorough investigation, there are still many unanswered questions regarding the exact method of DNA injection into gram negative cells by λ phage – such as the driving force behind the phage genome entry into the cytoplasm. In order to fully understand the significance of this, it is imperative to review the first few steps of λ infection among its host organism Escherichia coli.
There are three general steps at the beginning of the infection cycle for all phages: adsorption, breaching of the outer membrane, and DNA entry into the cytoplasm. Considering the morphology of the victim cell can complicate matters however, as infection procedures necessarily differ between gram positive and gram negative cells due to their difference in external cell layers. The focus of this paper is on λ phage, and its host Escherichia coli falls into the latter category. Gram negative cells are defined by their results during the Gram straining process, which is due to their smaller proportion of peptidoglycan relative to gram positive bacteria. It is also worth noting that said layer of peptidoglycan is located between the outer and inner membrane of these gram negatives, as opposed to being directly exposed to the environment such as in gram positive bacteria.
So, with consideration of gram negative cell morphology, DNA must pass from the λ phage through the outer membrane into the periplasm, past the thin layer of peptidoglycan, and through the inner membrane in order to reproduce and spread. The appeal of membranes being their selective permeability, λ phage depends on integral membrane proteins of both the inner and outer membrane for successful infection. However, the consequence of said layer of peptidoglycan, and what method is required to transverse it, is as of yet unknown.
λ Phage Adsorption to the λ Receptor
Before injection of DNA can occur, the first step of phage injection must be attachment to some sort of receptor along the surface of the cell. In λ phage’s case, this receptor is the maltoporin LamB, used for transporting small hydrophilic molecules, most notably the sugars maltose and maltodextrin (1). The LamB protein itself is a homotrimer, composed of three identical polypeptides, that form a beta-barrel that acts as the transport medium in question. λ phage interacts with this protein via the phage J protein portion of the tail fiber of the phage, forming an initial reversible interaction. This attachment is referred to as a Type 1 complex, which is defined by a gap of about 20 nm between the phage and the liposome covered in LamB receptors used for the study and notable interaction between it and the λ tail fiber. Type I complexes also significantly still retain the phage genome and can potentially detach and reattach at a different site (2). This reversible binding has been correlated with rapid dissociation and re-association, often taking multiple instances before irreversible adsorption takes place (2).
Once λ has attached to LamB, in vitro studies have shown that upon meeting temperatures of around 37oC, and in the presence of chloroform, the phage will undergo a structural shift (2). This structural change results in the banded tail portion directly attaching to the surface of the membrane, and the loss of DNA from inside the phage capsid heads. Importance of the LamB protein as the sole λ receptor on the outer surface is confirmed by attempting infection on cells that do not express the protein. When a plasmid containing the relevant gene and ampicillin resistance is inserted into the cell, phage susceptibility is restored (3).
Temperature plays a role in adsorption of the phage as well. Measurement of the rate of adsorption at different temperatures shows that adsorption rates are significantly lower as temperature decreases (3). Further investigation via fluorescent imaging of phages surrounding bacterial cells at higher and lower temperatures demonstrates an aggregation of phages in patches around the cell when temperature is low, implying that the LamB protein demonstrates clumping formation, leading to the decreasing rate of adsorption relative to warmer environments (3). In fact, comparison of the rate constant of phage desorption (detachment after adsorption) is double the rate constant of irreversible adsorption, so it can be assumed that it is twice as likely for the phage to attach and detach compared to attaching and staying permanently (3). This further emphasizes the significance of host temperature ranges for successful λ infection.
LamB Mutants
There are three types of mutants for the maltoporin that confer different level of resistances to λ phage. The first are Class 1 mutants, which are defined as resistant to λ phages with a normal host range, but susceptible to one and two-step host range mutants λh and λhh* (1). Class II mutants are resistant to λ wild-type host range and one-step host range mutants, but susceptible to two-step host range mutants (1). Finally, Class III mutants are resistant to all types of λ phage but lose maltoporin function entirely (4). The relevant section of the host genome is of particular note, because they are clustered around a specific region of the lamB protein, such that 12 amino acid residues were found to be absolutely essential for λ phage adsorption (1). Similarly, the host range mutations for λ phage are clustered around the last 4-10% of the J protein genome. When comparing the mutations of the λh mutants, correlations can be found among the J proteins. Specifically, for the Class I mutations Ser-247 -> Leu, Gly-245 -> Arg, and Glu-148 -> Lys, the associated λ phage on host mutants all had the mutation Leu-1127 -> Pro on the J protein gene. Although simply speculation on the author’s part, it is worth considering that the relevant genes may be related to an “accessibility gate” of the LamB receptor that shrinks the entrance, thus rendering the phage unable to adsorb properly to the protein (1). This would also explain why the host range mutants, which presumably shrink their J protein tip to an appropriate size, are also able to infect wild type cells with a normal maltoporin (1).
Another study focusing on lamB mutants that denote λ phage resistance found that none of the mutations were found in regions I through IV, out of XI, spread out sequentially along the genome from N-terminal to C-terminal (4). This means that mutations that affect λ infection are not present on the first third of the N-terminal side of the genome (4). To elaborate upon the clustering observed by the previous experiment, the Class I mutations are found in intervals X-XI, which is towards the C-terminus of the genome. This is notable because the C-terminus is not responsible for reversible (Type 1 complex) binding of λ phage, which implies the mutation affects Type 2 complex formation (4). Class II mutations are entirely focused in interval V, much earlier than the Class I mutations; the only other notable mutants were those resistant to phage K10 are found on interval VI, but these were λ sensitive – mutations in intervals X-XI, on the other hand, were resistant to λ but did not affect K10 sensitivity. Finally, none of the 18 λ resistant mutations significantly affected dextrin consumption, meaning that the mutations relevant to λ adsorption are not necessarily related to normal maltodextrin functionality. What is important about this study is the understanding that, within the limits of the mutations that were found, each mutation was found within intervals V and VIII-XI, implying that these areas are on the outer part of the outer membrane (4).
DNA Ejection
A true understanding of the mechanics behind λ DNA ejection is frankly lacking. However, there is some research that allows for educated extrapolation. Study into the relationship between DNA ejection from the phage capsid and surrounding temperature reveals that DNA packed inside the capsid undergoes a structural change itself at 33oC (5). The phage capsid is packed to the brim with DNA, at 55% by volume, otherwise filled with mostly water. The problem with this enormous potential energy inside the head is the risk of premature genome ejection, which introduces the concept of viral metastability, in which the virus must be able to detect the right moment to release its DNA. By examining DNA structure as a function of temperature, SAXS measured diffraction peaks directly related to DNA ordering was shown to decrease as temperature increased, showing that DNA disorders inside the capsid head at the right temperature, priming it for ejection. Otherwise, the DNA remains highly ordered at colder temperatures, correlated with much less intrastrand repulsion, and therefore a greatly reduced chance of DNA ejection (5).
Osmotic pressure of the bacterial cell also plays a role in DNA ejection. E. coli has a putative osmotic internal cell pressure of around 2 atm, far below the minimum inhibitory pressure required to prevent DNA ejection, which can vary between 15-25 atm depending on genome length (6). This suggests that stronger ejection forces are desirable for phages – however, too high of ejection forces can lead to premature genome ejection, which corresponds with increased DNA ordering (and therefore decreased ejection force) at colder temperatures (6). Also notable is that pressure inside the capsid cannot be the only mechanism for DNA ejection from the capsid head. When measuring DNA ejection rates, via spectrophotometer, it was found that the initial 50% or so of DNA was ejected relatively quickly. The remaining though, was removed at a remarkably slower rate. If DNA ejection depended entirely on capsid pressure, ejection rate should in theory be more consistent (6). This means that there is some unexplained mechanism that is contributing to final removal of remaining DNA (6).
David Valen, among others, provided unique insight into some of the possibilities of translocation mechanisms by measuring rates of DNA ejection compared to amount of DNA inside and outside the viral capsid. By using SYTOX Orange to stain phage DNA, and measuring absorbance within the capsid, DNA ejection velocity could be measured and then compared to the remaining fluorescence within the capsid head in vivo (7). DNA ejection velocity showed great variety between cells, at a rate of 5.4 ± 4.1 min in λ strain c160, and when mean velocity at given percentages of ejected DNA between phage genomes of different length were compared, significant overlap was found. This is significant because it means that velocity does not correlate with amount of genome inside of the capsid, and instead with the DNA now inside of the cell (7). What is also particularly interesting about this experiment is that the shorter phage genome of strain b221 encountered ‘stalls’, or ejections that did not finish during the course of the recording. This is significant because similar stalls have been associated with DNA-based motors, which have been considered as potential translocation mechanisms for λ phage (7). The pauses observed were much longer than those seen for motors, however, and were permanent, so this is not especially conclusive.
DNA Translocation Through the Inner Membrane Generally Among Phages
DNA movement through the cell is still missing one layer at this point – the inner membrane. In order to transverse it, λ phage must utilize the IICMan and IIDMan integral membrane protein of the mannose phosphotransferase system (8). This protein spans the periplasm and cytoplasm, providing access after transfer of the DNA through the outer membrane. What is known is that the relevant proteins are subunits of a larger structure that receive phosphoryl groups, and that both proteins have three membrane spanning sections in total (8). The ability to transfer particles through the inner membrane is clearly an important aspect of the mannose transport system’s utility for λ phage, but there remains the question of how the phage interacts with the protein without adsorbing directly to it as with the maltoporin.
This is addressed by the study of certain E. coli strains deficient in a protein pel are shown to be resistant to λ DNA penetration (9). Mapping of these mutations have found they are in the same place as the ptsM gene, which is a component of the mannose phosphotransferase system, demonstrating at the very least very high correlation and affinity, if not outright showing that they are the same gene (9). Further mutational analysis has shown that pel– mutants can be accessed by λhp mutants that have altered tail components pV and pH, located along the shaft of the tail, which suggests these two proteins must interact with the mannose phosphotransferase system during injection (10). Specifically, adsorption was still allowed, but DNA injection was inhibited, meaning that the mannose transport system must be correlated with DNA transfer into the cytoplasm, not initial adsorption, which makes sense because it is an integral membrane protein that doesn’t reach the outside of the cell (10). Gathering from this that the tail genes pV and pH are involved in some channel mediating DNA transport, an additional protein gpGT has been demonstrated to interact with gpV by radio-labeling proteins and examining correlation via SDS-PAGE, a type of gel electrophoresis (11). Furthermore, gpH – which has been identified as a tape-measure protein that extends a channel to control DNA movement – expression was tested using plasmids mixing and matching gpGT and gpG functionality, and gpH function was shown to only occur when all three were expressed together. This demonstrates strong correlation between gpGT, gpG, gpH, and gpV proteins to collectively mediate DNA transfer into the cytoplasm (11). Since the gpH and gpV proteins were correlated with interaction with the pel protein, it can be deduced that there is some interaction between these channel proteins of λ and the mannose phosphotransferase system, the latter very likely acting as a portal at the end of a constructed channel for DNA entry into the cytoplasm.
With regard to the thin layer of peptidoglycan that remains within the periplasm, study of other phages can potentially provide insight into possibilities for λ. Three proteins that are connected with murein hydrolase activity, which degrades peptidoglycan, have been identified as T7gp16, P22gp4 and gp3, all three of which are found among phages that can penetrate gram negative cells, and phages o6, PRD1, and PM2 are known to have hydrolase activity as well (12). This was verified by experiments similar to one that shows P22 phage virions lacking in gp4 lack the ability to degrade peptidoglycan (12). Both T7 and P22 infect E. coli and Salmonella thypimurium respectively, which suggests peptidoglycan degradation is a vital part of gram-negative cell infection. However, phage λ has not been shown to exhibit homologous hydrolase genes, which means that their mechanism could be significantly different from other phages that infect similar targets (12). The λ phage’s method for traversing the layer of peptidoglycan therefore remains unknown. While exact methods of insertion vary drastically between bacteriophages, some insight might be obtained upon comparison to a very different method of DNA insertion.
Mechanisms of entry are much better defined in other bacteriophages, such as the dsDNA phage T7. Initially, upon adsorption to the cell wall, T7 appears to eject some sort of channel proteins to allow for DNA to have a clear pathway into the cytoplasm of the host cell, which is similar to λ (13). The DNA travels around 850 base pairs forward and is then is clamped by the gp16 protein to prevent further expulsion from the capsid, and this is where major differences began to pop up. This strip of DNA has three promoters for host RNA polymerase, which then attaches to this initial strip and propels about 7 kbp of the genome into the cell – this second strip then codes for a T7 RNA polymerase, which propels the rest of the DNA into the cytoplasm (13). When gp16 mutants are studied however, the entire phage genome has been shown to inject into the cell at a constant rate (14). Study of this mutant reveals that this protein acts as a clamp that prevents constant DNA transcription, which turns out to be integral to the T7 infection strategy that avoids type 1 restriction enzymes, which are bacterial defense systems that would otherwise invalidate phage infection. This necessary process stretches out what would normally be less than a minute to around 20, making up two thirds of the thirty-minute latent period of wild-type T7 (14). This could potentially be of some relevance to understanding plausible injection mechanics for λ phage, because gp16 mutants translocate their DNA at a constant rate, which is inconsistent with the notion of a purely pressure based DNA injection mechanism, which would have a constant negative acceleration after the initial ejection (14). It is also worth noting how T7 uses channel proteins to overcome the length between the outside of the cell and the cytoplasm, just as λ phage has – this highlights the importance of recognizing common problems each phage must overcome, even if they do so differently.
Conclusion
What is notable about λ phage is the gap in complete knowledge on its ability to translocate DNA. Other phages, such as T7 and P22, have better understood mechanisms, such as those involving murein hydrolases or DNA polymerase that have not been found among λ phages. In a similar vein, even the process of DNA ejection from the capsid is poorly understood, despite the study of the roles of temperature and turgor pressure of the host cell. Some information can be gleaned from the mechanisms of other phages, but often homologs have not been found in λ. λ phage has been a model for understanding phage mechanics, and research could potentially be beneficial by use in understanding phage lysis as a potential alternative treatment for bacterial infection. Comprehension of λ’s ability to infect cells could also be useful in the development of antibiotics in vitro by elaborating upon the vulnerabilities of gram negative pathogens.
For now, research should focus specifically on all of the requirements for DNA ejection within the capsid, to maximize efficacy of phage therapeutics, especially on the mechanism behind DNA movement into the cell. Specifically, it is still unclear how exactly λ phage ejects its DNA from the capsid, as ejection rates have been inconsistent with a simple ‘uncorking reaction’. It is somewhat how λ guides its DNA into the cytoplasm, in a similar way other phages such as T7 which have specific mechanisms like channel proteins to address this problem. The role of murein hydrolase should be explored among phages in general as well, as it is required for the phages that have it in their genome – so why λ can get away without it is unclear. Despite the information on the role the mannose phosphotransferase system plays in the process of injection, there is still much to be elaborated upon, such as λ’s channel proteins maneuvering around the afore-mentioned cell wall to reach it.
Considering the mystery behind this process, it should be noted that new hypotheses are becoming increasingly refined as data comes in. In particular, Ian Molineux, an expert featured in many of the papers in this review, has come to a particular opinion on the ejection method of single-step DNA injections. Options being considerably limited by the incomplete theory of pressure-based ejection, and the λ injection rate too inconsistent with a motor-based mechanism, Molineux proposes the phage utilization of the osmotic gradient to pull water from the environment through the phage capsid and into the cell, pulling the phage genome along with it (15). This predicts a relatively constant rate that fits somewhat closely to what is observed with λ phage – however previous studies have shown that λ phage ejection is ejected at two different rates, which potentially conflicts with this mechanism (6). Regardless, it is a more viable theory than what has previously been proposed and should be investigated further in the future.
References
1. Werts, Catherine et al. 1994. Adsorption of Bacteriophage Λ on the LamB Protein of Escherichia coli K-12: Point Mutations in Gene J of Λ Responsible for Extended Host Range. Journal of Bacteriology. 176: 941-947.
2. Roessner C.A, Ihhler G. M. 1987. Sequence of Amino Acids in LamB Responsible for Spontaneous Ejection of Bacteriophage Λ DNA. Journal of Molecular Biology. 195: 963-966
3. Moldovan, R. et al. 2006. On Kinetics of Phage Adsorption. Biophysical Journal. 93: 303-315.
4. Hofnung, Maurice et al. 1981. General Method for Fine Mapping of the Escherichia coli K-12 lamB Gene: Localization of Missense Mutations Affecting Bacteriophage Λ Adsorption. Journal of Bacteriology. 148: 853-860
5. Liu, Ting et al. 2013. Solid-to-fluid-like DNA transition in viruses facilitates infection. PNAS. 111:14675-14680.
6. Grayson, Paul et al. 2006. The effect of genome length on ejection forces in bacteriophage λ. Virology. 348: 430 – 436
7. Valen, David Van. 2012. A Single-Molecule Hershey-Chase Experiment. Current Biology. 22: 1339-1343.
8. Huber F, Erni B. 1996. Membrane topology of the mannose transporter of Escherichia coli K12. European Journal of Biochemistry. 239: 810 – 817.
9. Elliot J, Arber W. 1978. E. coli K-12 pel Mutants, which Block Phage λ DNA Injection, Coincide with ptsM, which Determines a Component of a Sugar Transport System. Molecular General Genetics. 161: 1 – 8.
10. Scandella D, Arber W. 1976. Phage λ DNA injection into Escherichia coli pel– mutants is restored by mutations in phage genes V or H. Virology. 69: 206-215.
11. Xu, Jun et al. 2014. Chaperone-protein interactions that mediate assembly of the bacteriophage λ tail to the correct length. Journal of Molecular Biology. 426: 1004-1018.
12. Moak M, Molineux I. J. 2004. Peptidoglycan hydrolytic activities associated with bacteriophage virions. Molecular Microbiology. 51: 1169 – 1183.
13. Kemp, Priscilla et al. 2004. Bacteriophage T7 DNA ejection into cells is initiated by an enzyme-like mechanism. Molecular Microbiology. 53: 1251 – 1265.
14. Struthers-Schlinke, J. S. et al. 2000. The Internal Head Protein Gp16 Controls DNA ejection from the Bacteriophage T7 Virion. Journal of Molecular Biology. 301: 35 – 45.
15. Molineux, I. J. 2006. Fifty-three years since Hershey and Chase; much ado about pressure but which pressure it? Virology. 344: 221 – 229.
The Scenic Climax 