3.1. Investigation of the role of filtration method on pectin yield
The acid digestion technique is the most common procedure of pectin isolation. In this method, sugar beet pulp was mixed in hot acidified water for several hours, and the pulp released its pectin content to water. Following this process, pectin solution is separated from residual pulp via filtration. In previous studies, filtration step was performed with nylon cloth, centrifugation [18-20], vacuum filtration. The pectin product efficiency is known to be mainly reaction condition-dependent, whereas the filtering method is one of the critical contributors of the biopolymer isolation process. In this study, the potential effects of the filtering method on such isolation processes were investigated by comparing various separation applications. According to current investigation, the filtration method has a critical influence on the yield as it is anticipated. Filtration of digested pulp solution through cellulose filter paper (Whatman No.1 Cat.No: 1001-125) (11 µm) resulted in a low amount of filtrate since some of the pectin content is absorbed by filter paper. Therefore, a faster and better way of filtering through a vacuum filter (0.2 µM Polyethersulfone-PES-membrane) was applied for the next isolations. Despite the smaller pore size of PES filtration, the pectin yield increased almost two-fold from 18% to 33% (Table 1. trials 3 and 5). Although both cellulose and PES are hydrophilic membranes, PES get wet quickly and had superior flow rate compared medium flow cellulose filter paper. These results indicated that the material of the filtration apparatus and vacuum has a significant role in the biopolymer isolation process, the filtration method has to be carefully selected by considering the content of the extract.
3.2 Investigation of importance of pH arrangements on pectin yield and DE
Soluble pectin in filtrated isolate is recovered via ethanol precipitation. Different treatments can be applied before pectin retrieval; according to previous methods, the isolate's pH was either increased to 3.5 for kept constant . To understand the pH dependency of the resulting extract, the effect of pH adjustments on pectin yield and esterification degree were investigated. Firstly, the pectin sample was alkalized to pH 3.5, and this pH treatment was found to result in a significant improvement in product yield (up to 3-fold) compared to non-treated groups (Table 1. trials 4, 5 and trials 7, 9). Increasing pH to 2.5 has also improved the yield, whereas optimum pectin yield was acquired from pH 3.5 treated groups (Table 1. trials 7,9,10). The pH arrangements affected to esterification degree of pectin as well. In addition to the higher yields, increasing pH to 2.5 or 3.5 results in a higher esterification degree (DE) but the resulting pectin is still in the low-ester pectin range. It is also valid that pectin isolated in all trials is in low DE since the highest DE was 20%. It was clearly observed that in all trials, increasing the pH of filtrate to 2.5 or to 3.5 after the vacuum filtration, enabled improved precipitation in ethanol (Fig. S2) and better pectin yield than the groups in which pH was not changed (Table 1).
3.3 Investigation of the relationship between the type of agent used for pH arrangements and pectin yield
After observing the importance of pH for pectin yield, it has been questioned whether the type of agent used to adjust pH conditions of isolate influences the yield as well. In previous studies, before alcohol precipitation, pH adjustments were made with salts like K2HPO4  or bases like NH3·H2O [20, 21]. In this study, the possible effects of salt or base use on pectin isolation, K2HPO4 and NaOH were separately investigated. Using salt solution (K2HPO4) resulted in higher pectin yield than base solution (NaOH) (Table 1. trials 10 and 11).
3.4 Investigation ethanol precipitation step for pectin yield and DE
Following the treatments indicated above, pectin can be extracted from the solution via alcohol precipitation, and this step can have a possible influence on the yield quantity. The administered volume of alcohol (absolute ethanol) varies among the studies. Previous studies reported that two volumes (2X) of ethanol to one volume of extract [18, 22] or four volumes (4X) of ethanol [18, 23]can be applied to precipitate pectin. To optimize the required ethanol volume for precipitation, three different volumes of ethanol (4X, 3X and 2X) were applied (Table 1. trials 6, 7, 8). When the filtrate was precipitated in ethanol, pectin yield became 47.6%, 35.8% and 38.89%, respectively, with 4X, 3X and 2X ethanol volumes. Esterification degree of pectin was 11%, 9.1% and 14.3%, respectively. There is no direct correlation between ethanol volume and pectin yield, and the DE of pectin was obtained; the increasing volume of ethanol led to a slight increase in the quantity of obtained yield.
3.5 Investigation of the effect of drying method on pectin properties
In the last step of pectin isolation, the precipitated pectin is retrieved from the solution and dried. According to our observations and literature examples , drying conditions directly influence the physical characteristics of pectin. In this study, pectin samples were dried with the freeze-drying method or at 60 °C in a heater overnight. The most demonstrable characteristic of the resulting pectin was its dark brown color which is dried at 60°C. The quality and quantity of resulting pectin yield were poor, the brown-colored product is comparably different from commercial pectin, and the yield was insufficient to disable any further analysis (Fig. S3 A). The quantity of the isolated pectin was not sufficient to calculate the esterification degree. Therefore, wet pectin was dried with the freeze-drying method and this technique considerable increased the quality and the quantity of the obtained pectin. As indicated in previous work  on soy pectin, drying conditions affected the appearance of isolated pectin. It has been shown that pectin produced by oven drying has the lowest Hunter values (lower Hunter L values indicate dark color) for color compared to freeze-dried and spray dried pectin. It was also mentioned in another work that dehydration and drying with hot air dryer results in degradation of pectin backbone because of β-elimination reaction. Temperatures above 50 °C can also increase the activity of enzymes that degrade pectin.
3.6 Fourier Transform Infrared Spectrophotometer (FTIR) Analysis
Figure 2 shows the FT-IR spectra of pectin extracted from sugar beet pulp, using the various conditions presented in Table 1. Carbohydrate specific spectral regions of four selected samples were compared to elucidate the relationship between the isolation method and the quality of the product. In most cases, improvement in the yield quantity results in a decrease in quality [25, 26]. Therefore, two high yield and two low yield extracts were selected for comparison.
In comparison with previous works on the infrared spectra of pectin extracts [27, 28] for the four samples that were analyzed in this study, ν (O-H) peaks were found in the region of 3600–3000 cm–1. The peak at ̴1620 cm–1 was attributed to carboxylate; since the esterification degree of all samples is low, strong carboxylate ion stretching was observed in each spectrum. For the trials 7, 9, 10 and 11 carboxylate specific absorption peaks were observed at 1606 cm–1, 1618 cm–1, 1618 cm–1, 1624 cm–1 and 1640 cm–1, respectively (Fig.2).
The strong asymmetrical stretching at 1620 cm–1 is accompanied by another carbohydrate-specific weak symmetric stretching band around 1441 cm–1, followed by strong absorption patterns between 1300 and 800 cm–1. This set of absorption peaks enables identifying the major chemical moieties in polysaccharides, and this 'fingerprint 'region is unique to the carbohydrates . In this region, galacturonic acid in pectin molecules was identified using spectral region between 1120–990 cm–1 (Fig.2). The band at 1019 cm–1 indicated that pectin extracts contain pyranose. The week peaks characterized the presence of D-glucopyranosyl and α-D-mannopyranose at 917 cm–1 and 825 cm–1, respectively[30, 31]. In addition to this, the most characteristic bands of L-arabinose were also observed at 1445 and 1260 cm-1 . Trials 7 and 10 have shown stronger carbohydrate specific peak profiles indicating better preservation of saccharide content during isolation. This result can be attributed to use of salt (K2HPO4) during pectin isolation in trial 7 and 10. Salt might have acted as stabilizer to prevent saccharide degradation from pectin chain.
3.7 Protein content of pectin
Like low DE low protein content is also indicates the success of the pectin isolation procedure and low protein content is desirable for biomedical applications. Therefore, protein content of the pectin isolates was measured for each trial. It was found that protein content ranged from 1.02- 3.17% among trials. As it was already declared by Li et al.  and shown in the study of Yapo et al.  longer extraction time and higher temperatures assist to breakage of covalent linkage between protein residues and pectin chain. Hence, low protein content of pectins isolated in all trials seems to be associated with these harsh isolation conditions used in this study.
3.8 Investigating toxicity of pectin in vitro
Anti-proliferative and antimetastatic  activity of pectin extracted from various sources became the subject of previous studies. For instance, pectin was shown to have an anticancer activity on prostate cancer [9, 10]and colon cancer. However, these studies mainly focused on citrus pectin, modified citrus pectin and apple pectin. There are very few studies investigating the anticancer activity of sugar beet pectin[15, 19]. Pectin isolated from various sources with different methods can result in specific metabolic responses on different cancer derivatives. This potential activity is possibly gained after pectin modifications like heat and pH treatments, resulting in depolymerization and de-esterification of pectin. In this study, by optimizing the isolation technique, it was aimed to isolate pectin from sugar beet pulp not just with improved yield and quality but also with already acquired bioactivity, which dismiss the requirement of post isolation treatments.
Hence, we investigated the toxic effect of pectin on cancerous SaOS-2 osteosarcoma. The osteosarcoma cell line was selected as being relatively aggressive , which is advantageous for tracing the anticancer activity of pectin. In addition, the effect of pectin isolates from different sources on cell lines was also investigated by isolating citrus (orange peel) pectin (CP). The isolates obtained from high yield methods with low esterification degree (Table 1, trial 6) were used for cell culture analysis.
SaOS-2 cells were incubated with different concentrations of sugar beet pulp pectin (SBPP) and CP for up to three days. 2%, 1%, 0.5% 0.25% of SBPP [8, 15] and %0.66, %0.33, %0.16 and %0.083 CP were applied to investigate toxicity profile of pectin isolates. According to our preliminary studies, the viscosity of the CP solution were much higher than the same concentrations of SBPP; therefore, 3-fold lower concentrations of CP was used, which have similar activity level with 1-fold SBPP. Together with an obvious decrease in cell viability after 24 and 48 hours of treatment with 1% SBPP, the viability of SaOS-2 cells decreased considerably (69.3%) for 72 hours of treatment with 2% SBPP. The highest anti-proliferative activity of CP was observed after 24 hours of treatment with the highest concentration (%0.66). Lower concentrations of SBPP and CP were not effective to disable cell growth.
Although after 24 hours of treatment activity of SBPP and CP on cell viability rather close to each other, for longer durations SBPP was more effective to hinder cell viability (Fig. 3). This outstanding property of SBPP was observed in HUVEC cells as well. SBPP revealed better antiproliferative profile compared to CP, which can be critical parameter to suppress metastasis of aggressive cancer lines. Therefore, for longer treatments SBPP could be selected as the first choice over CP.
3.9 Measuring increase in Reactive Oxygen Species (ROS) Level in cells treated with pectin
The viability tests have shown that SBPP and CP have toxic activity for osteosarcoma and healthy HUVEC cell lines in determined concentrations. At this point, to get an insight into the cellular death mechanism and understand whether/how cells undergo apoptosis, we measured cellular reactive oxygen species (ROS) and level of anti-apoptotic gene (Bcl-2) expression upon pectin treatment.
Although low levels of ROS are essential for the maintenance of biological functions, excess amount of ROS production upon toxic signals may induce apoptosis . ROS regulates apoptosis via the intrinsic mitochondrial pathway and by regulating pro-apoptotic or anti-apoptotic genes' expression, including Bcl-2. However, in some cases, excessive accumulation of ROS can provoke other cell death mechanisms. Since most toxic concentrations of pectin disable proper cell growth, for ROS measurement, second-most and least toxic concentrations of pectin were selected for comparison. SaOS-2 and HUVEC cells were incubated with %1 and %0.25 SBPP and %0.33 and % 0.083 CP for 24 hours and 72 hours. After 24 hours of treatment, ROS level increased more than 4-fold (407%) in 0.25% SBPP treated SaOS-2 cells. The increase in ROS level was comparatively low (39.86%) in the higher concentration of SBPP. 0.33% CP treated SaOS-2 cells experience an 84.3% rise in the ROS level after 24 hours (Fig. 4A). These results exhibited that SBPP can be used as ROS inducing agent, and the effective dose was determined as 0.25%.
Cellular ROS level decreased after 72 hours, compared to 24 hours of treatment in all concentrations. SBPP is more effective in boosting cellular ROS levels than CP in shorter durations. 0.25% SBPP and 0.083% CP were nontoxic for SaOS-2 cells. However, these concentrations were the most influential ones in ROS rise in SaOS-2 cells (Fig. 4A). Intracellular ROS inducing function of citrus pectin was previously shown in the study of , our results have found consistent with previous studies and SBPP induced significant enhancement in ROS production compared to CP.
SBPP boosted ROS formation in cancer cells compared to normal cells. Since cancer cells are believed to exhibit elevated steady-state fluxes of ROS relative to healthy cells due to the impairment in oxidative metabolism, pectin treatment was caused predominant ROS production in cancer cells. After 24 hours of incubation of HUVEC cells with %1 SBPP pectin resulted in a 34.32% increase in ROS formation while only a 6.65% increase was observed for 0.25% SBPP treated cells compared to the control group. For the 0.33% and 0.083%, CP treated group, 57% and 48.28% increase was observed respectively (Fig. 4B). According to these results, even though an anti-proliferative effect was observed in both cancerous and healthy cell lines, SBPP induced different cellular mechanisms in cancerous SaOS-2 cells, which have found promising to develop different treatment methods.
3.10 Investigating apoptosis-related gene Bcl-2 expression level in pectin treated SaOS-2 and HUVEC cells
Bcl-2 is a member of a group of proteins that regulates cell survival but the mechanisms whereby Bcl-2 functions are not well characterized. It is an anti-apoptotic oncogene whose imperative expression causes cell proliferation by surpassing the effect of oncogenic stress. Bcl-2 prevents apoptosis by regulating various mechanisms  and in one reported mechanism Bcl-2 regulates the protection of the cells against oxidative injury resulting from post-hypoxic reoxygenation. Thus, if pectin exposure leads to apoptosis in cancer cells, it can be because of the decrease in Bcl-2 anti-apoptotic gene expression. To investigate the efficacy of pectin as an anticancer agent, PCR analysis were performed with the reference of GAPDH housekeeping gene. Similar to ROS measurements, cells were treated with 1% and 0.25% SBPP and 0.33% and 0.083% CP for 24 and 72 hours. Since most toxic concentrations of pectin prevent cell growth and thereby disable RNA isolation from cells, second most toxic and non-toxic concentrations of pectin were selected for PCR analysis.
In SaOS-2 cells, Bcl-2 expression decreased compared to control cells only when cells were incubated with 0.33% and 0.083% CP for 24 hours (Fig. 5A) and 72 hours (Fig. 5B) of exposure. In the viability test, 0.083% CP have found almost non-toxic (Fig. 2). However, there is an increase in ROS and a decrease in Bcl-2 expression level in cells treated with 0.083% and 0.33% CP. 24 h treatment of 0.33% CP induced a rise in the cellular ROS level, a decrease in cell viability and in the Bcl-2 expression. These results indicate that CP may induce cellular death via apoptosis.
For the SBPP treatment, even though ROS level increases in 0.25% SBPP, the elevated ROS production did not decrease Bcl-2 expression, unlike CP. It was also suggested that under certain circumstances higher ROS level and high Bcl-2 expression eventuate simultaneously. These results mean that CP and SBPP activate different pathways in the SaOS-2 cell line. Moreover, the cell death induced by SBPP does not directly cause apoptosis, or other factors might overcome the Bcl-2 effect. The observed prolonged cell death and better antiproliferative activity of SBPP could be caused by this action mechanism.
Interestingly, 24 hours of treatment did not cause a difference in Bcl-2 expression for HUVEC cells both for SBPP and CP (Fig. 5C). These results are meaningful since the cellular ROS level also did not change at this concentration in HUVEC cells. Blc-2 expression decreased after 72 hours of treatment for all tested concentrations of SBPP and CP (the highest decrease was observed in 1% SBPP) (Fig. 5D). Regarding the viability results, the proliferation of HUVEC cells was not negatively influenced by these concentrations of pectin after 72 hours (Fig. S4). Even though the expression of Bcl-2 declined, ROS levels did not change, and the viability of cells was not affected by pectin treatments.
All these results indicate that SBPP and CP induce different pathways in different cell lines, and an increase in the ROS level does not always cause apoptosis and a decrease in the Bcl-2 expression. Since many other genes are also responsible for cellular death, further analysis is necessary to better understand the mechanism of anticancer activity of SBPP.