Saccharomyces cerevisiae as a Model for Reprogramming of Eukaryotic Cells: Implications for the Study of the Relationship Between Metabolism and Inflammation in Chronic Disease.

Our aim was to determine if fasting of S. cerevisiae would change their mitochondrial response to a high glucose environment (which they are typically grown in) and to an LPS challenge. In these initial experiments being present here, we investigated whether key respiratory parameters were set to respond differently to the same growth environment. Specifically, after growing in a high glucose (2%) broth, we looked at State 3 and State 4 O respiration as well as their ratio known as the Respiratory Control Ratio (RCR) as an indicator of coupling efficiency [ 24 ], and the difference between Uncoupled Maximum OCR and State 3 (often referred to as Reserve Capacity). Importantly, we also looked at the ADP/O ratio, to determine how much oxygen was consumed to respond to a given amount of additional ADP, as a different indicator of coupling efficiency [ 24 ]. We hypothesized that the progeny of the fasted yeast would have been reprogrammed to be more tightly coupled (higher efficiency) and more metabolically flexible, ‘tilting’ more toward oxidative phosphorylation than toward fermentation, even in the same environment. We therefore also set out to determine if reprogramming altered PDC activity, facilitating additional oxidative phosphorylation. Theoretically, this would be akin to restoring health after a threatening experience. It was clearly evidenced that there were changes in the way the mitochondria utilized fuel in relation to the rate at which they consumed oxygen. Comparing the progeny of the fasted yeast with the progeny of the non-fasted HG yeast (both progenies having grown in normal high glucose broth), there was a trend toward a lower OCR in the progeny of the fasted yeast than in the ‘normal’ high glucose progeny. When exposed to LPS, the differences between the two groups were most convincing for State 4 O , reserve capacity, and uncoupled max OCR (Fig. 1 ). Fig. 1 Variations in oxygen consumption rate (OCR) characteristics of the mitochondria isolated from the progeny of the pre-conditioned yeast S. cerevisiae . A State 3 respiration determined from the maximal oxygen consumption slope immediately post ADP addition. B State 4 O respiration determined measuring the slope after the addition of Oligomycin A (known to block ATP synthase). C Reserve Capacity, determined as the difference between the Uncoupled Max and State 3. D Uncoupled Maximum respiration, determined using the uncoupler FCCP at a concentration pre-determined to achieve maximum uncoupling without inhibiting respiration. E Coupled respiration, determined by deducting State 4 O respiration from State 3 respiration. F Yield of Oxidative Phosphorylation (ADP/O ratio). Calculated by dividing the added 125 nmol ADP by the nmols of oxygen consumed, excluding oxygen accounted for by State 4 O , for the duration of added ADP consumption. G RCR, determined by dividing State 3 by State 4 O . Good coupling of oxidative phosphorylation is reflected by a high RCR [ 24 ]. (1-tailed Student T tests were performed to test for significance, paired when comparing data from the same conditioning and unpaired, assuming equal variance, when comparing across the two conditions). Each sample was assayed in duplicate and the results are the mean of three independent experiments ±S.E.M Variations in oxygen consumption rate (OCR) characteristics of the mitochondria isolated from the progeny of the pre-conditioned yeast S. cerevisiae . A State 3 respiration determined from the maximal oxygen consumption slope immediately post ADP addition. B State 4 O respiration determined measuring the slope after the addition of Oligomycin A (known to block ATP synthase). C Reserve Capacity, determined as the difference between the Uncoupled Max and State 3. D Uncoupled Maximum respiration, determined using the uncoupler FCCP at a concentration pre-determined to achieve maximum uncoupling without inhibiting respiration. E Coupled respiration, determined by deducting State 4 O respiration from State 3 respiration. F Yield of Oxidative Phosphorylation (ADP/O ratio). Calculated by dividing the added 125 nmol ADP by the nmols of oxygen consumed, excluding oxygen accounted for by State 4 O , for the duration of added ADP consumption. G RCR, determined by dividing State 3 by State 4 O . Good coupling of oxidative phosphorylation is reflected by a high RCR [ 24 ]. (1-tailed Student T tests were performed to test for significance, paired when comparing data from the same conditioning and unpaired, assuming equal variance, when comparing across the two conditions). Each sample was assayed in duplicate and the results are the mean of three independent experiments ±S.E.M LPS was not as clearly seen to affect State 4 O of the F progeny as it was seen to increase State 4 O of the HG progeny. If considering the RCR however as an indicator of coupling efficiency (and by extension an inexact indicator of membrane potential), when the F progeny were exposed to LPS, their State 4 O respiration was possibly achieving roughly the same OCR at a lower membrane potential (Fig. 2B ). That could mean that in response to LPS, the membrane integrity was maintained, and proton leak was decreased (for example possibly due to closing or possible turnover of a potential uncoupling protein (e.g., SC MUC [ 25 , 26 ]) turnover or the block in supply reduced ROS production diminishing activation of adenine nucleotide translocators [ 10 , 27 ]). While the difference in State 3 of the F progeny after LPS exposure was not as great as the difference observed in the HG progeny after LPS exposure, it was also not to the same degree proportionally because the mitochondria needed more oxygen to make the same amount of ATP (Fig. 2F ) This could for example possibly be due to a change in the electron transport chain (ETC). State 4 O is when ATP synthase is inhibited and is often thought to be an indicator of membrane integrity. However, the possibility that different programs can achieve the same OCR at different membrane potentials could reveal a new understanding of how the regulation of coupling efficiency is programmed. Fig. 2 Data shown in Fig. 1 relative to the RCR (State 3/State 4 O ). A State 3 respiration, determined from the maximal oxygen consumption rate immediately post ADP addition, B State 4 O respiration determined measuring the slope after the addition of Oligomycin A (known to block ATP synthase), C Reserve Capacity determined as the difference between the Uncoupled Max and State 3, D Uncoupled Maximum respiration, determined using the uncoupler FCCP at a concentration pre-determined to achieve maximum uncoupling without inhibiting respiration, E Coupled Respiration, determined by deducting State 4 O respiration from State 3 respiration, F Yield of Oxidative Phosphorylation (ADP/O ratio). Calculated by dividing the added 125 nmol ADP by the nmols of oxygen consumed, excluding oxygen accounted for by State 4 O , for the duration of added ADP consumption. Data shown in Fig. 1 relative to the RCR (State 3/State 4 O ). A State 3 respiration, determined from the maximal oxygen consumption rate immediately post ADP addition, B State 4 O respiration determined measuring the slope after the addition of Oligomycin A (known to block ATP synthase), C Reserve Capacity determined as the difference between the Uncoupled Max and State 3, D Uncoupled Maximum respiration, determined using the uncoupler FCCP at a concentration pre-determined to achieve maximum uncoupling without inhibiting respiration, E Coupled Respiration, determined by deducting State 4 O respiration from State 3 respiration, F Yield of Oxidative Phosphorylation (ADP/O ratio). Calculated by dividing the added 125 nmol ADP by the nmols of oxygen consumed, excluding oxygen accounted for by State 4 O , for the duration of added ADP consumption. LPS caused Reserve Capacity (i.e., Uncoupled Max – State 3) to be higher in both progenies, but more so in the HG Progeny (Fig. 1C ). Being higher in the HG progeny possibly suggests that the F progeny were programmed to be more tightly coupled, and the higher RCR in relation to the higher OCR due to LPS exposure for the F progeny is more clearly seen in Fig. 2C . Based on how high the LPS caused the Uncoupled Max to be in the HG progeny (Fig. 2D ), it appears the inefficiency that is possibly built into the ETC of the HG progeny is the main contributor to the observed high Reserve Capacity (which might be thought of as greater ‘slippage’ which could have implications for what Reserve Capacity really tells us). A key to making sense of this is that Coupled Respiration (Fig. 2E ) had virtually the same OCR for all conditions, i.e., reprogramming did not affect OCR via a change in the difference between State 3 and State 4 O (of a particular cell population) when dealing with a high glucose environment. Nor did it clearly affect this difference when challenged with a PAMP (although there was a trend toward LPS exposed cells to have a greater difference). Yet, what is clearly different is that programming possibly caused them to make adjustments such that they maintained the same difference between State 3 and State 4 O OCR with different RCR’s (State 3/State 4 O , which is thought to indicate different membrane potentials). We believe this is fundamental to understanding what is happening. And will explore this further with real time quantitative measures of membrane potentials. The RCR on its own suggested tighter coupling in the F progeny, and this is in agreement with studies on fasted trout [ 24 ]. And in both F and HG progenies LPS caused a decrease in the RCR, but it was a greater decrease in the F progeny (19% compared to 15%; Supplementary Table 1 ), This is also quite striking when considered with the ADP/O ratio. Both RCR and ADP/O say something complex about coupling efficiency, and therefore, looking at how they differ can be quite revealing. LPS caused ADP/O to be slightly higher in F progeny, yet slightly lower in HG progeny. This once again supports the fact that the protocol has in fact reprogrammed the yeast. This also supports the conclusion that this reprogramming resulted in the F progeny being more tightly coupled. However, the response with regard to ADP/O unlike the RCR suggests the F progeny became more efficient, using less oxygen to utilize the ADP in response to LPS. And this becomes very clear when ADP/O is plotted against RCR (Fig. 2F ), where the LPS appears to make the F progeny more like the non-LPS HG Progeny. But then the increase in oxygen consumption in the HG progeny after LPS is not matched to the ADP utilization comparably to the F progeny (i.e. ATP production to meet demand may be hampered). Based on our previous work in mammalian tissues [ 13 ] where we first proposed that programing of an organism is the adjusting of the regulation of coupling efficiency, we might speculate that the HG progeny are programmed to not be able to maintain their ability to utilize ADP from an increased oxygen consumption at lower membrane potential. We have called this ‘coping capacity’, and it appears that the reprogrammed F progeny are possibly able to better deal with the toxic high glucose environment and still be able to adjust to the toxic high glucose + LPS environment possibly by becoming less efficient with regard to ETC and membrane integrity dynamics but more efficient with regard to ATP synthesis; perhaps simultaneously lowering membrane potential and increasing OCR (which is classic uncoupling), while the HG progeny were already operating at a lower efficiency, their membrane potential was possibly even lower after LPS exposure. From a practical sense, whereas in our animal models we were able to predict relative proximity to death, this data does not reveal which progeny would be better off if exposed to the E. coli that the LPS came from. But from what we have seen in this study, it will now be exciting to move to the next phase and use the model to actually relate these changes to membrane potential. The PDC data (Fig. 3 ) supports the idea that the reprogramming could be affecting fuel supply, to direct the glucose metabolite, pyruvate, towards the Krebs cycle and oxidative phosphorylation (more ‘opened valve’), whereas cells that had evolved from progenitors that had always been in high glucose were more restrictive on allowing glucose metabolites access to the mitochondria, tilting toward fermentation (more ‘closed valve’). In humans it is thought that being too heavily geared toward fermentation could contribute to cytokine storms (e.g. in conditions like sepsis or Covid19 [ 28 ]) or just chronic inflammation (e.g. in conditions like atherosclerosis or asthma), as the cells cannot shift back to oxidative phosphorylation in response to subtle signals by nutrient/energy sensors. Tilting towards fermentation would theoretically also lead epithelial cells to proliferate, like in cancer or endometriosis. Fig. 3 Pyruvate Dehydrogenase Complex (PDC) activity based on NADH production. High Glucose (blue) and Fasted (green) mitochondria isolated from cells regrown in high glucose after nutritional interventions. The PDC activity levels are represented as moles of NADH produced per milligram of protein per minute (nmol/mg/min). 1 unit of PDC is considered to form 1µmol of NADH per minute at pH 7 and 30 °C for high glucose and fasted mitochondria. (1nmol/mg/min = 1 unit/mg). Data represent the single measurements from three independent experiments ±S.E.M. Mitochondria were frozen at −70 °C and the assay was carried out on a single day. Negative controls with no NAD + in the assay buffer were used which showed no increase in absorbance readings over the reaction time. The reaction wells were blanked with negative control wells to find the actual rise in absorbance due to NADH levels Pyruvate Dehydrogenase Complex (PDC) activity based on NADH production. High Glucose (blue) and Fasted (green) mitochondria isolated from cells regrown in high glucose after nutritional interventions. The PDC activity levels are represented as moles of NADH produced per milligram of protein per minute (nmol/mg/min). 1 unit of PDC is considered to form 1µmol of NADH per minute at pH 7 and 30 °C for high glucose and fasted mitochondria. (1nmol/mg/min = 1 unit/mg). Data represent the single measurements from three independent experiments ±S.E.M. Mitochondria were frozen at −70 °C and the assay was carried out on a single day. Negative controls with no NAD + in the assay buffer were used which showed no increase in absorbance readings over the reaction time. The reaction wells were blanked with negative control wells to find the actual rise in absorbance due to NADH levels The data from the PDC experiments (Fig. 3 ) showed a great deal of variation due to one of the preparations producing anomalously lower absorbance readings. We were unable to determine what was different, so the data was included in the results and although it increased the variance, it still did not eliminate the average difference between the two groups which is why we are reporting it as we are ( p  = 0.32). However, all three independent experiments resulted in the PDC activity being higher in the progeny of the fasted yeast. And when the anomalous data was removed from analysis, there was no overlap between the data sets, further suggesting the difference was real. Further research needs to be done to confirm this. However, the assay contains triton-x and due to technique, this sometimes led to the formation of bubbles in the wells, which was eventually resolved by using a small gauge needle. It is possible that this had not yet been perfected, and some bubbles were missed. Yeast, as a model of immune response, is a challenging idea. It has been proposed that yeast started producing ethanol when plants started producing fruit that fell to the ground with toxic levels of sugar [ 29 ]. The history of ethanol fermentation positions the process as having the core parts of our immune system built into our single cell ancestors that we share with fungi, with ethanol secretion in the yeast, and fermentation in general, as a defensive (immune/pathogen) response conjoined to glycolysis as a detoxifying (immune/poison) response, and the pentose phosphate pathway as a fortifying (immune/damage/pathogen/poison) response. Variation between progeny of the F vs HG in this experiment is thought to show reprogramming of how coupling efficeincy is managed. And the variation in LPS treated samples is thought to indicate reprograming of a cellular immune response. Programming, as adjustments of the regulation of coupling efficiency [ 13 ], determines how much fuel and/or how much oxygen is required to meet a specific demand and at what membrane potential the mitochondria make that transaction. This supply and demand adjustment programming then plays out in all cellular processes. Based on work in Martin Brand’s group over more than three decades and more recently in very small mammals [ 30 , 31 ], the ability to manage the matching of supply and demand at a lower membrane potential could facilitate a longer life. In summary, we first demonstrated that S. cerevisiae could be reprogrammed to metabolically respond differently to a specific environment, compared to ‘normally’ programmed yeast. To our knowledge, this is the first time it has been shown that programming, at the level of coupling efficiency, could be altered over generations by dietary changes, with reprogramming the regulation of coupling efficiency being the novel contribution. Furthermore, the reprogrammed yeast were also shown to respond differently to a PAMP (LPS), with regard to bioenergetic changes associated in mammalian cells with the switch to a proinflammatory and proliferative metabolic state. This switch from ‘tilting’ toward oxidative phosphorylation to ‘tilting’ toward fermentation, is analogous to that of M1 macrophages or endothelial cells seen in atherosclerosis [ 32 ], as well as possibly all mitochondria containing mammalian cells when exposed to excess glucose. Research that better delineates glucose poisoning at the cellular and organismal level could also have a foundation here. This data clearly supports the use of this model for further investigation of inflammatory processes and potential interventions to restore proper regulation of immune responses, and thereby restore coherence, the true basis of health. A key issue highlighted by this initial study is that accurate quantitative measures of membrane potential will be of immense value in future studies. At this stage, what this study contributes to the longstanding, even ancient, concept of energy imbalance as a cause of pathology and death, is a clear experimental model and the language needed to formalize the idea and address the problem of how to reprogram health.

BY4742 yeast (Invitrogen), from the original stock were streaked on a YPG agar plate, (Agar [OXID], 1%w/v, yeast extract [OXID], 2%w/v bacterial peptone [OXID], 3% v/v glycerol) and incubated at 30 °C for 24-h. A single robust colony was selected and added to 2 ml YPD broth, referred to here as high glucose (HG) broth (1%w/v yeast extract [OXID], 2%w/v bacterial peptone [OXID], 2% w/v α D-Glucose [Aldrich]) and incubated (INCU-Shake FL18-750R) for 24-h at 30 °C, rotated at 200 rpm. After which 20 μl was aliquoted on to a YPG agar plate which was incubated at 30 °C for 24 h before being placed at 4 °C. All experimental cultures were started using a single colony of SC selected from this plate. A single yeast colony was transferred into 2 ml HG broth in a 15 ml Falcon tube and warmed for three hours in a shaker at 30 °C/200 rpm. This was then transferred to a 500 ml glass flask containing 100 ml HG broth and returned to the incubator for 24 h at 30 °C 200 rpm. After 24 h the yeast were centrifuged (Eppendorf, 5804 R ) at 4 °C/1000xg/5 min then resuspended in 30 ml sterile cold d H 2 0 and re-pelleted by centrifugation at 4 °C/1000xg/5 min before being resuspended in 5 ml sterile d H 2 0. A cell count reflecting greater than 8 × 10 9 live cells at this stage was seen to produce sufficient mitochondria preparations. Equal amounts of yeast were added to either fasted in 100 ml sterile d H 2 O (F) or to 100 ml HG (HG) and incubated at 30 °C 200 rpm for 23 h. Both conditions were pelleted by centrifugation at 4 °C/1000 g/5 min before being resuspended in 7 ml HG broth and placed on ice. It is important to recognize that a matching number of progenitor cells from each condition was seeded into 500   ml HG broth (1 × 10 9 was found to be the minimum number of cells needed for each intervention to result in a sufficient mitochondrial yield after isolation). Each 500 ml was divided equally (100 ml each) into 5 × 500 ml flasks and incubated at 30 °C/200 rpm for 20 h. Yeast progeny were pelleted, washed with d H 2 O, and combined to produce a single pellet for each condition. Yeast challenged with LPS were conducted on separate cultures to those that remained unchallenged (due to timing constraints). Unchallenged cultures progressed directly to isolation. Yeast from both conditions (F and HG) were washed in 4 °C d H 2 O, resuspended in 50 ml HG broth with LPS (100 ng/ml), and placed into 250 ml flasks with LPS, concentration based on the work of Marques et al. [ 18 ]. After 90 min the yeast were pelleted, washed and re-pelleted the same as the unchallenged yeast. Yeast isolation followed the protocol for isolating yeast mitochondria, as described by Gregg et al. [ 19 ], with a few key notable exceptions. As all pellets were combined, we picked up the original protocol at step 5. All pelleting of yeast prior to step 17 was done at 4 °C/1000  g /5 min. At step 14 rather than powdered Zymolyase, the enzyme was dissolved in d H 2 O, aliquoted, and stored at −70 °C. At the final step, the isolated mitochondria were suspended in 0.1 M MES/NaOH buffer, pH 6.5. The protein concentration of the final mitochondrial suspension was assayed using the Biuret protein assay, with BSA as protein standard. Analysis, of respiration rate and proton leak kinetics, of the isolated mitochondria, was conducted using a Rank Brothers Clark Electrode Setup (oxygen sensitive electrode), connected to an AD Instruments PowerLab and LabChart 8 software (AD Instruments). The Clark Electrode chamber (3.5 ml) with recirculating water jacket was regulated at 30 °C and was calibrated using electrode buffer (20 mM Tris/HCl, 450 mM sorbitol, 100 mM KCl, 0.5 mM EGTA, 5 mM MgCl2, 10 mM K2HPO4, 0.1% defatted BSA, pH 6.8), assumed to contain 424.8nmol 0 2 /ml at 30 °C [ 20 , 21 ]. Once calibrated the chamber containing 3.5 ml electrode buffer was carefully closed, making sure to be free of air bubbles, before 60 μl of 100 mM d-Lactate, was added using a micro-syringe through a small aperture. d-Lactate is an electron donor to cytochrome c in yeast mitochondria [ 21 , 22 ]. Once the electrode trace had stabilized, the mitochondrial prep (F, HG, F-LPS or HG-LPS), was added to a final in-chamber concentration of 0.3 mg/ml. Then 5 μl of 25 mM Adenosine Diphosphate (ADP), to establish State 3. Then 2 μl of 0.01 mM Oligomycin A (OMY), an ATP synthase inhibitor, was added to establish State 4 O . 2 × 2 μl titrations of 1 mM Carbonyl cyanide-p-trifluoromethoxy phenylhydrazone (FCCP), an ionophore that acts as an uncoupling agent, used to establish the maximal, uncoupled oxygen consumption rate (OCR). 1.5 μl of 3.5 mM Antimycin A (Anti A), a Complex III inhibitor, was used to establish residual/non-mitochondrial oxygen consumption (ROX). The electrode chamber was thoroughly cleaned before and after each experimental run. At the end of each day, a few crystals of Sodium Dithionite were sprinkled in the open chamber, removing all oxygen. This was used to re-verify the calibration, by checking if the LabChart trace dropped to zero. Each condition was assayed in duplicate in three independent experiments. The real time electrode data was captured using LabChart 8 analysis software, and the traces later analyzed using the Data Pad functionality. This data was then exported to an Excel spreadsheet, where the data was normalized for the protein concentration and the voltage fluxes converted to OCR measured in nmol of O 2 consumed per minute per mg protein (nmol/min/mg) this was done using the following formula: \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$${\rm{OCR}}({\rm{nMol}}/\min /{\rm{mg}})=\frac{\left({{\rm{O}}}_{2}{\rm{slope}}\left\{{\rm{units}}\right\}\times 60\right)\times 424.8}{{\rm{Oxygen}}\; {\rm{full}}\; {\rm{chart}}\; {\rm{span}}\left\{{\rm{units}}\right\}\times {\rm{Concentration}}\; {\rm{of}}\; {\rm{protein}}\; {\rm{in}}\; {\rm{chamber}}\left\{{\rm{mg}}/{\rm{ml}}\right\}}$$\end{document} OCR ( nMol / min / mg ) = O 2 slope units × 60 × 424.8 Oxygen full chart span units × Concentration of protein in chamber mg / ml OCR was then calculated for State 3, State 4, State 4 O , and Maximum OCR by deducting the ROX value from the values derived. This can be done because the ROX value is assumed to be constant within this setup [ 23 ]. These can be seen as presented in Supplementary Fig. 1 . ADP/O ratio was determined by identifying the duration (s) of ADP consumption from when the slope changed following the addition of ADP until the slope returned to the pre-ADP value. By subtracting the amount of oxygen consumed by leak (calculated from State 4 O OCR) from the amount consumed for ATP production (calculated from State 3 OCR), by dividing each OCR by 60 and multiplying by the duration, we were able to determine the nmol O2 consumed and divide that into the nmols of ADP consumed. The assay was done on a 96-well plate with 18 reaction wells and 18 negative controls with no NAD + . Each reaction well contained 40 micrograms protein of mitochondrial sample brought up to 25 μl with homogenization buffer with no BSA. 25 μl of 0.25% v/v Triton X-100 was added to lyse the mitochondrial membranes and incubated at room temperature for 10 min after a quick mix with plate mixer. 100 μl of assay buffer (50 mM K3PO4, 1 mM MgCl2, 2 mM pyruvate, 2.6mM L-Cysteine, 2 mM Thiamine pyrophosphate, and 2.5 mM NAD+ (SIRT substrate), pH 7.2) was added and incubated at 30 °C for 10 min after a quick mix with plate mixer. 100 μl of coenzyme A (final concentration 0.13 mM) was added to start the reaction and a quick mix was given by reverse pipetting with a multi-channel pipette. Absorbance readings were taken at 340 nm every 60 s for 10 min with an endpoint set up using microplate reader (Bio-Rad 680 XR). All statistical analyses were conducted using Microsoft Excel. Measurements were done in duplicate (with the exception of PDC). Error bars show the Standard Error of the Mean (S.E.M.) of averages from three independent experiments. When examining statistical significance, 1-tailed Paired Student T tests were used when comparing data between F and F-LPS as well as between HG and HG-LPS and 1-tailed unpaired Student T tests were used when comparing F or F-LPS to either HG or HG-LPS, in these cases equal variance was assumed. To determine statistical significance of the percentage changes as a result of the LPS challenge in each of the F and HG conditions, the following steps were taken: first the values from each repeat of F was divided by the mean value of the same data from F-LPS, providing the 1 st value set. Then the value from each of the F-LPS repeats was divided by the mean value of the same data from the F, providing the 2 nd data set. A 1-tailed paired Student T test was then conducted between the two data sets. The same process was followed to determine the significance of this data between HG and HG-LPS. As this is exploratory research to justify further research, p  ≤ 0.10 was considered statistically significant.

Supplementary Material Supplementary Material