“A team of researchers led by the Yale School of Public Health has found that the Lyme disease bacterium is ancient in North America, circulating silently in forests for at least 60,000 years—long before the disease was first described in Lyme, Connecticut, in 1976 and long before the arrival of humans”, (Caccone, 2017). Lyme disease has been an endemic a long time before we even knew it existed. It is believed that it traveled across the world through mammals. There are four kinds of bacteria that cause Lyme disease. In the United States we see Borrelia burgdorferi and Borrelia mayonii. In Europe and Asia there is Borrelia afzelii and Borrelia garinii. “The most common vector-borne disease in North America, bacterium Borrelia burgdorferi, is spread through the bite of infected black-legged tick, commonly known as a deer tick”, (Center of Disease Control, n.d).
Borrelia burgdorferi is a gram-negative, spirochete bacteria. Since it is gram-negative the outer membrane is composed of lipopolysaccharide and the inner membrane has a layer of peptidoglycan. Even though this is a gram-negative bacteria, in gram-staining tests it shows up like a gram-positive bacteria, which is very deceiving. B. Burgdorferi can replicate to survive in its host with bacterial products. This is unlike other virulence factors, especially since this is primarily what the bacterium needs to cause disease inside its host.
Many people who own dogs know tick season is during the warmer months of the year, but ticks are around all year. Nymphs are immature ticks that are tiny (less than 2nm), which makes them harder to see. Most humans become infected through the bite of a nymph. Adult ticks are much larger, so they are easier to spot quicker, thus they get removed quicker before the bacteria is transmitted. Adult ticks carry Lyme disease too, but because of their size not as many humans become infected through their bite. Ticks attach to any skin, but they like warm and dark places, so they most commonly attach to places that are harder to see like the groin, armpits, and scalps. “It takes about 36 to 48 hours for a trick to transmit the Lyme disease bacteria into its host. The disease can be hard to diagnose because many of its symptoms are like those of the flu and other diseases, and the host probably did not notice a tick bite”. (Medline-plus, 2019)
Early Signs and Symptoms occur 3 to 30 days after the infected tick bite. These symptoms include fever, chills, headache, fatigue, muscle and joint aches, and swollen lymph nodes may occur in the absence of rash; and the most visually prevalent, Erythema migrans rash (Occurs in approximately 70 to 80 percent of infected persons about 7 days post contact, expanding gradually over several days reaching up to 12 inches plus, warm to the touch, rarely itchy or painful, sometimes clears as it enlarges showing as a target or “bulls-eye”, (Mayo clinic, 2019).
Later Signs and Symptoms typically occur many months after the infected tick bite. These symptoms include severe headaches and neck stiffness, additional rashes on other areas of the body, facial palsy or the loss of muscle tone causing droop to one or both sides of the face, arthritis with severe joint pain and swelling, intermittent pain (tendons, muscles, joints, and bones), heart palpitations and/or an irregular heartbeat, dizziness, shortness of breath, inflammation of the brain and spinal cord, and even nerve pain resulting in numbness and tingling in the hands or feet, (Mayo clinic, 2019).
There is no evidence that Lyme disease can be transmitted from person-to-person, meaning no transmission from touching, kissing, or having sex with an infected person. If it is acquired during pregnancy, it can lead to an infection within the placenta and possible stillbirth, however, no negative effects on the fetus have been found when the mother receives appropriate antibiotic treatment. There are no reports of Lyme disease transmission from breast milk as well, (Center of Disease Control, 2019).
Ticks are found in grassy and heavily wooded areas. The more ticks an area has, the more likely it is that they are carrying Lyme disease. Those who spend time in those areas are more likely to acquire a tick bite that has Lyme disease. It is important to take precautions and check yourself after spending time in tick-infested areas, so even if a tick has latched on, it can be removed quickly. If Lyme disease does occur the sooner treatment begins, the better the outcomes of complete recovery are possible. Most treatments include a regimen of antibiotics. “Antibiotics commonly used for oral treatment include doxycycline and amoxicillin. People with certain neurological or cardiac forms of illness may require intravenous treatment with antibiotics such as ceftriaxone or penicillin”, (Mayo clinic, 2019). This can be adjusted depending on the persons current and past medical history. Sometimes patients experience post-treatment Lyme disease syndrome (PTLDS). This is when patient experience muscular joint aches, and nervous system symptoms after they have finished treatment. Many people believe this means they still need antibiotics, but long-term antibiotics have been proven to not help with these symptoms. The only thing patients can do is treat the pain until it is gone.
In conclusion; the spirochete bacteria, Borrelia burgdorferi, is an endemic disease known simply as Lyme disease due to its first recorded location, even though it has been discovered to be around much longer. It for now is transmitted via vector borne infected ticks, and not from any other known species, and is more often found in grassy and heavily wooded areas. The rate of infection is slow but if the host is not known to have been infected it can be hard to treat as the symptoms can be displayed as other diseases. With the sooner the application of antibiotics the faster the recovery and the healthier the life of the host will be. Until a vaccine can be discovered, humans will just need to be more careful and check each other for ticks.
Caccone, A. (2017). Ancient History of Lyme Disease in North America Revealed with Bacterial Genomes. Yale School of Public Health. Retrieved from https://publichealth.yale.edu/news-article/15651/
Lyme Disease Treatment.(n.d.). Center for Disease Control and Prevention. Retrieved from https://www.cdc.gov/lyme/treatment/index.html
A Role for Cell Free ERVs in Propagating IFN Type-I Responses
As previously mentioned, OC is the most lethal of the gynaecologic malignancy in the western world and this is due to poor prognosis and limited treatment efficacy (Coretz et al., 2018). In a meta-analysis Immunotherapy alone has been proven to not yield significant results in OC patients, requiring the need for a combinatorial therapy (Alipour et al., 2016). Until now, research has mostly focused on mechanisms of DNMTi including abnormal DNA methylation of tumor suppressor genes. However, focus has now shifted to ERVs suppression by hypermethylation, predominantly in somatic cells (Phillips, 2018). Work by Jones et al., 2019, has focused on viral mimicry mechanisms showing DNMTi treatment inducing transcription of ERVs within the cell. However, considering other mechanisms of ERVs apart from ‘viral mimicry’ is crucial as OC tumours may rely on other mechanisms apart from DNA methylation to downregulate tumour suppressor genes and antigen presentation via the MHC-complex.
Interestingly, high plasma levels of the RNA of various HERV ENV genes (HERV-K, HERV-R, and HERV-H) have been found in primary cancer patients from various cancers, before and after treatment with Guadecitabine. These experiments conducted in Professor Ghaem-Maghami lab have hinted the possibility of ERVs in the plasma. Therefore, the presence of cell free HERV-H, the top candidate ERV from RNA-seq data, was investigated following Guadecitabine treatment of OC cell lines. Kuramochi and Ovsaho cell lines were utilised as they genetically resemble high grade serous OC which represents 80% of cases (Lisio et al., 2019). Data obtained demonstrated an increase in RNA expression of HERV-H in both Kuramochi and Ovsaho 5 days after treatment with SGI-110 as seen in figure 3.1A. 1 uM and Kuramochi yielded the highest HERV-H expression therefore was the most optimal effective dose to use for consecutive experiments. Although there was an observed increase, no statistical test had been applied due to limited technical repeats, thus it may not be significantly different. Figure 3.1A indicates the presence of HERV-H in the supernatant in a dose-dependent manner as would be expected. Although the mechanism in which cell free RNA leaves the cell is unknown, Jones et al., 2019 established that ERVs are reactivated into the cytosol as dsRNA following epigenetic drug exposure via MDA5, and translocate back into the nucleus through MAVS aggregation and IRF7 activation. However, research to date has not specifically explored the potential of a dsRNA escaping sensing by PRRs like MDA5, TLR3, RIG-1 and leaving the cell. The data mentioned above suggests Guadecitabine may activate ERVs, escape sensing and leave the cell either freely or through a mechanism of action not yet established. Although this data seems promising, expression fold change levels are below 5 and very variable. In order to optimise the HERV-H expression and reduce variability between replicates, RNA integrity needs to be established and improved, as RNA quantification results stated in Table 3.1, showed that although there was sufficient RNA quantity, the A260-A280 values were controversial, potentially causing this issue. This can be rectified by measuring RNA integrity using an agarose gel electrophoresis on 1.0% agarose gel containing GelRed™ (Biotium, Inc., Hayward, CA, USA) as performed in literature, in order to QC RNA quality used for qPCR. Treating OC cell lines with 5uM has shown minimal HERV-H expression potential as DNMTi at a high concentration may be cytotoxic and cause DNA damage in the cells. This elucidates why lower doses must be used considered clinically to avoid cytotoxicity in OC patients. One can argue that HERV-H from inside the cell pellet is being detected as opposed to HERV-H within the supernatant. We attempted to decrease this possibility by spinning the media collected and removing the visible cell pellet, however, due to technical pipetting errors, there is the possibility of lcarry-over. This may have occurred during handling of the cells in tissue culture prior to RNA extraction. When calculating log fold change, CT values for PPIA housekeeping gene were quite variable, although standard deviation values were quite low. PPIA as opposed to GAPDH, has a higher stability and a good candidate for OC cell lines as stated by Jacobsen et al.,2014; however, the setup of the experiment and the SGI-110 treatment at 2 days may have interfered with PPIA causing instability. Additionally, RNA quantity may have been variable well to well, such as inconsistency in diluting RNA as cDNA enzymes and primers may have inhibited the reaction all due to technical errors.
Housekeeping genes involved in different biological processes may have varied stabilities so the DNMTi-treatment of OC cell lines could have influenced its stability as Yauhen et al., 2017 states that the enzymatic activity of PPIA may affect its stability. In addition, a study investigating reference genes in OC, PPIAs statistic stability remained the same despite different OC tissue conditions, however, other genes including RPL4, RPLP0, and HSPCB proved to be the most stable for qPCR in OC tissue, thus must be considered in future experiments in which DNMTi treatment does not affect its stability (Fu et al., 2010). Excluding variability, this data may indicate the presence of HERV-H in the supernatant although the mechanisms of action leaving the cell remains unknown.
Enhanced type I IFN signalling
HERVs have previously been reported to promote tumour growth (Lemaître et al., 2017). However, current research focuses on to a new role of HERVs which include their ability to produce an anti-tumour response. As mentioned previously, DNMTi treatment induces an IFN response in OC cells following activation of MDA5, RIG-I and TLR-3, in which de-methylation of ERV expression, aids IRF3, NF-kB and IRF7 transcription leading to type- I IFN production. In this project, it was explored whether exposing untreated OC lines with cell free dsRNA, presumably HERV-H, upregulates IFN type-I response in OC cell lines. In figure 3.1B and C, there is evident data possibly indicating the upregulation of IFN-α and IFN- β release in a dose-dependent manner, following exposure of supernatant from DNMTi-treated OC cells.
IFN-α and IFN- β both OC cell lines is increased in 24 hours which may due to cell death and debris being detected as an IFN response. This suggest that the OC cell lines may be over-confluent due to the DNMTi and inducing cell death. Cell free HERV-H expression was also assessed alongside type-I IFN responses and showed increase in expression compared to the vehicle, highlighting its influence on type 1 IFN response, as well as the potential ability to induce self-expression. IFN-α and IFN- β provide the first line of host defence against viruses in order to provide immune protection. However, they’ve also been implicated in the crosstalk between immune and tumour cells in which IFNs were previously known to drive inflammation within TME (Crow and Ronnblom, 2019). However, there is now more focus on type I possessing immune-stimulatory features on cells such as T -lymphocytes which will be discussed shortly. Recently, Buoncervello et al., observed in a study focusing on colorectal cancer, that IFN-α potentiates the direct and immune-mediated antitumor responses following epigenetic drugs. Specifically, IFN-α has the potential to induce both anti-proliferative and pro-apoptotic features in vitro. It is now acknowledged to be vital to stimulate immunogenic cell death (ICD), strengthening the anti-tumour specific immune responses, enhancing its suitability as a target for novel therapeutics within epigenetic therapy for OC. Moreover, IFN-signalling has also been shown to cooperate with epigenetic drugs to inhibit tumour cell proliferation in vivo (Buoncervello et al., 2016). Specifically, Chiappinelli et al., 2015, have demonstrated that inhibiting DNA methylation induces IFN-signalling in OC via ERVs. Reversing abnormal DNA methylation on promotor site leads to re-expression of tumour suppressor genes, a key attribute in malignancy, as well as promoting apoptosis and changes in cell-cycle (Tsai et al., 2012).
In solid tumours, IFN- β is appreciated as an essential factor in viral defence signalling by up-regulating antigen presentation and immunogenicity by interacting with IFNAR1/2. This activates the JAK/STAT pathway, as JAK/STAT inhibitor strongly decreased ISG transcription. Following up-regulation of IFN-stimulated genes and production, apoptosis also occurs (Medrano et al., 2017). It was observed that IFN- β and viral defence genes (induced by DNMTi) are not hyper-methylated meaning that pathways upstream of these genes are stimulated following DNMTi treatment.
Data obtained from this project showing dose-dependent trend of IFN type-1 signalling of both IFN-α and IFN- β, by treating untreated tumour cells with cell free supernatant, corresponds with Chiapenelli et al data that acknowledge DNMTi’s inducing cytoplasmic sensing of dsRNA and stimulating type-1 IFN in OC cell lines. This suggests the possibility of cell-free RNA causing the induction of immunogenic type 1 IFN in neighbouring cells. If the mechanism of action of IFNs associated with anti-tumour effects can be established, this may lead to optimised combinatorial epigenetic therapy in the future for OC patients failing to response to monotherapies such as immunotherapy alone.
Cell free RNA, Tumour and T cells
γδ T cells, a key subset of “unconventional” T lymphocytes, have been widely used in immunotherapeutic research. γδ T cells are considered as good targets for therapies, optimising anti-tumour responses due to their distinct features, including their ability to recognise Ag’s expressed by tumour cells without MHC-class presentation, as well as producing IFN-γ and TNF- α which possess strong cytotoxic features including stimulating DC and NK cells against neoplastic cells and inducing ADCC (Lo presti et al., 2018).
Work by Chen et al., 2019 has demonstrated that γδ T cells are implicated in tumorigenesis and metastasis when infiltrated in cancers including breast and colorectal cancers. However,
the ability of neoplastic cells to create immunosuppressive environment to evade immune surveillance is a set-back encountered in immune-oncology therapy. Specifically, there is
limited data in literature regarding γδ T cells in OC patients. It has been shown by Fisher et al., 2014, that they’re a vital part of TILs in OC patients, as well as Presti et al.,2018 working on a transcriptome study, established tumour-infiltrating γδ T cells as the most significant favourable prognostic mark. Their functions are often changed or abrogated by immunosuppressive signals originating from the TME. Due to γδ T cells unique features and their controversial implication in tumour progression, they were utilised in this project to assess the role of cell free RNA on tumour: T cell interaction.
Specifically, IFN-γ levels were measured to assess γδ T cell activation following exposure to cell free supernatant in the presence of tumour cells. Here, results displayed in figure 3.3Bshowed a significant decrease in IFN-γ release in γδ T cells from DNMTi treated Kuramochi cells, in comparison to the vehicle.
IFN-γ is a key moderator or cell-mediated immune responses, specifically pro-inflammatory actions, however a recent publication by Bhat et al.,2017 has shown IFN-γ up-regulating anti-tumour and antiviral effects of CTLs. Therefore, the finding from figure 3.3Bshowing a decrease in IFN-γ expression, was not expected as it was assumed there would be a higher IFN-γ release due to the presence of potential cell free ERV inducing T cell activation. However, this decrease in IFN-γ may be attributed to T cell exhaustion. Normally, a CD8 T cell response is provoked in the presence of neoplastic cells, although a sustained stimulation pushes T cells into a non-functional state which decreases CTLs proliferative ability and hindering cytokine expression (Dogra et al., 2016)
This can be avoided in future experiments by collecting the supernatant of co-culture assay <24 hours to avoid prolonged exposure of γδ T cells with the cell free RNA thus preventing T cell exhaustion. As well measuring IFN-γ, γδ T cells secrete pro-inflammatory cytokines, including TNF- α and IL-7, in response to tumour cells which can be measured by flow cytometry, in the future to establish γδ T cell activation in the presence of cell free RNA.
T-cell proliferation was assessed using the live/dead stain to investigate if DNMTi-treated tumour cells had an effect on γδ T cell proliferation. Indeed, it did as treating tumour cells with cell free RNA led to an increase in live percentage of γδ T cells at 48 hours compared to the vehicle.
Consequently, the data obtained verified T cell activation in the presence of DNMTi-treated tumour cells. Data from this project shows in the presence of DNMTi-treated tumour cells with ZA-stimulated γδ T cells, a significant decrease **, is observed in cell viability compared to the vehicle. This suggests that stimulated γδ T cells are involved in tumour cell killing, specifically in those treated with cell free supernatant. It has been established in the past that γδ T cells are implicated in tumour cell lysis (Handgretinger and Schilbach, 2018). γδ T cells interacts with IPP produced via mevalonate pathway. Bisphosphonates, such as ZA used in this project, inhibits FPPS leading to increase IPP and DMAPP in Kuramochi cells. γδ T cells detect HSPs and MICA/B or ULBP expressed through NKG2D on tumour cells (Gruenbacher and Thurnher, 2017). An increase in perforin is observed, as seen in figure 3.8 in the presence of DNMTi-treated cells, allowing granzyme B to enter through the pores and induce apoptosis in the Kuramochi cell lines. In addition, γδ T cells also releases IFN-γ as a bystander effect, which recruits other immune cells (Jarosz-Bie et al., 2019) and may be responsible for the killing, in which data in figure 3.8 encourages this rationale, as an increase in both Perforin Granzyme B expression is seen in γδ T cells cocultured with DNMTi-treated tumour cells.
CD8 T cell possess cytotoxicity functions very similar to γδ T cells, apart from recognising Ag’s via MHC-class 1 complex in order to induce anti-tumour response (Godfrey et al., 2018). Treating tumour cells with DNMTi and co-culturing with γδ T cells showed no apparent difference in CD8 T cell activation. As previously mentioned, γδ T cells are also involved in priming CD8 T cell responses, although it is not seen in figure 3.6B, as well as inducing DC maturation, NK-cell mediated cytotoxicity, both contributing to indirect antitumor response (Srivastava et al., 2015). In addition, T-cell exhaustion as previously described may be the culprit in why the DNMTi-treated tumour cells are not activating CD8 T cells it has been seen that DNMTi can work by avoiding the onset of exhaustion and reprogramming exhausted CTLs into effector phenotypes (Wherry and Kurachi, 2015).
Further research is necessary to understand how dosing and scheduling of these drugs in the clinical setting will modulate the immune response. It is important to mention that there is a counter balance in which epigenetic therapy can actually be immunosuppressive. As witnessed in in MDS and AML patients, the upregulation of PD1 in T cells by DNMTi has been shown to limit epigenetics therapy efficacy (Sun et al., 2018). Therefore, further work is required to classify the most effective approached to induce pro-immunogenic as opposed to immunosuppressive features.
Additionally, the TME can be classified as 3 states, firstly, inflamed state whereby T cells are actively participating