A total of 32 different isolates of mesophilic bacteria were isolated and selected from the locations of companies as eight representative isolates for each location (Supplementary Table S1). These isolates were biochemically analyzed. Microscopically, the examination of fresh cultures of these bacterial isolates divided them into two major groups: bacilli (21 isolates) and cocci (11 isolates). Moreover, 17 of them were Gram-positive, and the rest were Gram-negative.

All collected bacterial isolates (32 isolates) were identified using MicroScan Panels (Table 1). The results of MicroScan Panels confirmed that all strains were negative for H2S production, and the remaining tests showed variable results among bacterial isolates. The results of MicroScan revealed that six isolates were identified as Bacillus subtilis, seven as Leuconostoc mesenteroides, four as Pseudomonas fluorescens, four as Escherichia coli, two as Acinetobacter baumannii, four as Staphylococcus xylosus, and three as Enterobacter amnigenus, in addition to one unidentified species called Aeromonas with two isolates. Acinetobacter baumannii was present in Dakahlia and Kafr El-Sheikh regions, as were Aeromonas species isolated from samples collected from Sharkia and Alexandria regions, while Staphylococcus xylosus, L. mesenteroides, P. fluorescens, and B. subtilis were common in the four study regions. Escherichia coli were found in samples from Dakahlia, Alexandria, and Kafr El-Sheikh regions, but Enterobacter amnigenus was isolated in Dakahlia, Sharkia, and Kafr El-Sheikh regions.

Brief biochemical descriptions of the eight identified bacterial species based on the MicroScan panels' results are shown in Table 1. Ten thermophilic isolates were isolated from beetroot samples collected from four companies after growing on nutrient agar at temperatures ranging from 55°C to 70°C. All of them were bacilli in shape and were Gram-positive and spore-forming. The biochemical and physiological characteristics of these isolates indicated that all of them could grow on casein and could not grow in Voges–Proskauer broth medium (Table 2). Moreover, they could reduce nitrate and convert hydrogen peroxide. All isolates could hydrolyze starch except isolate 2.

An amplified fragment of 350 bp of the 16S rRNA gene was amplified from the selected thermophilc bacterial isolate as a representative one, which was previously identified as Bacillus themophilus using biochemical analysis. This fragment was sequenced using the forward primer previously used for amplification, and the generated sequences were submitted to BLASTN to find homology with other bacterial 16S rRNA sequences from the Gene Bank. The homology of sequence analysis identified this strain as Bacillus themophilus HMA 220 with an accession number (MW577708). The phylogenetic tree with the closest species based on the UPGMA method is shown in Figure 1 using MEGA 10.

A total of 20 different plant extracts were examined as antibacterial substances to inhibit the growth of Bacillus thermophilus because it was a major microorganism causing sugar losses. The results revealed that 18 plant extracts effectively inhibited the growth of Bacillus thermophilus (Table 3), and this was shown by forming an inhibition zone around the growth of the microbe; however, two plant extracts (Mentha longifolia L. and Nerium oleander L.) were not effective at all. Data on plant extracts showed that the most effective plant extracts against the growth of Bacillus thermophilus were recorded with extracts of Geranium gruinum (40 mm), followed by Datura stramonium (39 mm), and Mentha spicata (35 mm) (Figure 2 and Table 3). On the contrary, the lowest inhibition effect was observed with the extract from Cymbopogon proximus (hochst) staps (15 mm) (Table 3). The rest of the other plant extracts exhibited moderate inhibition of the growth of the thermal bacterial strain (B. thermophilus), and their inhibition zones ranged from 30 to 34 mm (Table 3).

A total of 13 different oils were screened for their ability to inhibit the growth of B. thermophilus (Table 3). The results indicated that oils of Mentha spicata showed the highest inhibitory effect against B. thermophilus with an inhibition zone of 50 mm, followed by oils of Ocimum bacilicum with an inhibition zone of 45 mm (Figure 2). The other five oils (Cinnamomum zeylanicum, Origanum majorana, Citrus limon, Eugenia caryophyllus, and Eucalyptus rostrata) had a moderate effect on the tested strain with inhibition zones of 25, 23, 22, 20, and 20 mm, respectively, while oils from Pimpeniella anisum, Olea europaea, Thymus vulgaris, and Allium sativum were very less effective against B. thermophilus with inhibition zones of 13–19 mm (Table 3). The remaining two oils, Apium graveolens and Prunus dulcisvar, had no effect on the tested organism (Table 3).

Based on our findings, the most active antibacterial plant extracts are Geranium gruinum and oils from Mentha spicata and Ocimum basilicum against the growth of B. thermophilus. Subsequently, it was necessary to test whether the addition of these inhibitory products would affect the properties of the juice. Interestingly, it was noted that the pH of the juice increased (up to pH 5.75) compared to the control (pH 4.2), which was accompanied by an increase in the amount of sucrose (18.4%) compared to the control (15.2%). On the contrary, treatment with either Ocimum basilicum oil or the plant extract of Geranium gruinum showed no significant effect on the pH value (pH 4.2) and, subsequently, did not induce any considerable increase in the percentage of sucrose compared to the control.

The decrease in lactic acid concentration due to the growth of thermophilic bacteria was considered to be an indication of sucrose losses because of degradation. The highest ranges of lactic acid concentration (660 mg.L⁻¹) were noted after 72 h of incubation. However, adding oils from Mentha spicata, Ocimum bacilicum, and Cinnamomum zeylanicum reduced lactic acid by 47.7%, 26.5%, and 35.6%, respectively.

Based on these findings, Mentha spicata and Ocimum basilicum oils were selected to further investigate their phytochemical characteristics. To obtain insights into the bioactive contents of the most active oils, we conducted GC/MS analyses for Mentha spicata oil and Ocimum basilicum oils. The results of GC/MS analysis indicated that 28 compounds were recognized in Mentha spicata oil Figure 3A. The most abundant components found in the Mentha spicata oil were cis-isopulegone (12.29%), followed by 3methyl-6-(1-methylethyl)-2-cyclohexen-1- one (8.41%), α-myrcene (6.69%), limonene (6.22%), methyl acetate (6.15%), γ-terpinene (5.43%), α-farnesene (4.49%), 5-methyl-2-isopropyl-2-hexenal (4.29%), and 5-methyl-3-heptanol (3.44%) (Table 4). On the other hand, 23 constituents were detected in Ocimum basilicum, representing 98.02% of the oil (Figure 3B). Cis-13-octadecenoic acid (81.53%), followed by hexadecanoic acid (6.66%) and L-linalool (2.22%), were found as major compounds. The whole chemical composition of the oil of Ocimum basilicum, as analyzed by GC and MS, is listed in Table 5.

Random amplified polymorphic DNA-polymers chain reactions (RAPD-PCR) were performed to know the genetic changes that can be present for the thermophilic strain (B. thermophilus) after treating it with different concentrations of Mentha spicata, Ocimum basilicum, and Cinnamomum zeylanicum oils in beet juice (1/25, 1/50, 1/100, and 1/200). The data presented in Figure 4A show that primer A1 provided 15 different band patterns with Mentha spicata oil treatment. Among these, 12 patterns showed polymorphic bands, while three bands were monomorphic. When treated with Ocimum basilicum oil, primer A1 generated 13 band patterns, all of which exhibited polymorphic bands except for two monomorphic bands.

Bacterial cells treated with the oil of Cinnamomum zeylanicum produced seven polymorphic bands and three monomorphic bands. The data presented in Figure 4B show that when primer EZ351 was used with the microbe and treated with Mentha spicata oil, it resulted in 18 band patterns. Out of these patterns, 11 displayed polymorphic bands, indicating genetic variation, while seven patterns exhibited monomorphic bands that remained consistent across all samples. Similarly, treating the thermal microorganism with Ocimum basilicum oil produced eight polymorphic and seven monomorphic bands. Furthermore, the application of Cinnamomum zeylanicum yielded five polymorphic and seven monomorphic bands.

The data presented in Figure 5A show that primer RAPD 2 induced nine monomorphic and four polymorphic bands after treating B. thearothermophilus with Mentha spicata oil while using the oil of Ocimum basilicum as an antibacterial against B. thearothermophilus produced seven monomorphic and three polymorphic bands. The addition of Cinnamomum zeylanicum oil to inhibit the growth of the thermophilic Bacillus strain produced seven monomorphic and three polymorphic bands. The data presented in Figure 5B show that primer A7A10 succeeded in matching six polymorphic bands when the tested strain was treated with oil of Mentha spicata. In the case of using Ocimum basilicum oil, the primer produced seven polymorphic and only one monomorphic band.

Extracted mRNA from the bacterial cells of a selected isolate (B. thermophilus) treated with the effective oil of Mentha spicata was used to synthesize cDNA using primers RAPD1, RAPD2, RAPD3, and RAPD4 (DD-PCR). Variation in gene occurrence and density of the selected strain due to treatment with antibacterial products was carefully examined after the amplification of cDNA by comparing bands in a gel at different times of treatment (0, 8, 16, and 24 h). A number of such unique bands that were only present in treated samples and absent in non-treated samples were observed. Data presented in Figure 6A show that primer RAPD1 produced 17 different bands, primer RAPD 2 (Figure 6B) generated 16 different band patterns, and primer RAPD 3 produced 14 different band patterns (Figure 7A). Data in Figure 7B show the generated bands produced from primer RAPD4, which was produced for differentiation. The molecular weights of these bands produced from the four examined primers were between 150 bp and 950 bp. These bands were selected, cut, purified, and sequenced. Five specific genes (bands) from strain B. thermophilus, which were induced when treated with Mentha spicata oil, were selected for sequencing. The sequence analysis revealed that these genes corresponded to lipid kinase, extracellular solute-binding protein, naphthoate synthase, major facilitator superfamily, and transglycosylase (Table 6). Further search on the Internet regarding the function of each gene confirmed that they are all involved in the defense pathway mechanism.

To count mesophilic bacteria, 1 mL of juice was spread on the surface of tryptone soya agar medium (Scotter et al., 2000), and then, plates were incubated for three days at 37°C. Growing colonies were picked up and used to inoculate the tryptone soy broth medium. Afterward, 1 mL of this inoculated broth medium was distributed on the surface of the blood agar medium to isolate G+ve bacteria (DIFCO Laboratories Dehydrated, 1984), while the G-ve bacteria were obtained by distributing 1 mL of this culture on the surface of the MacConkey agar medium. To count and isolate the thermophilic bacteria, adopting the procedure of Sahm and Washington (Sahm et al., 1991), 1 mL of the serial dilutions was spread on the surface of the nutrient agar medium, and plates were incubated at 55°C for 48 h. Both isolated mesophilic and thermophilic bacteria were examined microscopically, and the common isolates were purified and subjected to complete identification through biochemical tests and molecular analysis.

Mesophilic bacterial isolates were subjected to complete identification using the MicroScan Dried system in the Microbiology lab at Mansoura Children's Hospital University, which is designed to identify both Gram-negative and positive bacteria using fluorogenic substrates as a pH indicator for a bacterial enzymatic reaction. Conventional MicroScan Negative and positive Combo panels were inoculated with the bacterial isolates using the standard turbidity technique. The panels were incubated for 24 h at 35°C within the MicroScanWalkAway system. All procedures were performed according to the manufacturer's instructions (Abdel-Wahab et al., 2022). It is also used for measuring several biochemical tests, including carbohydrate fermentation tests such as glucose, xylose, mannitol, sucrose, galactose, rhamnose, maltose, raffinose, sorbitol, arabinose, inulin, mannose, and lactose, in addition to other specific tests such as Voges Proskauer (VP), nitrate reduction (NIT), indole test, urease test, hydrogen sulfide production test, citrate utilization, and oxidase test.

Isolates were tested using classical microbiological methods such as Gram stain, spore-forming using a phase contrast microscope (Sala et al., 1995), and motility as described by Priest (Priest et al., 1988), in addition to examining their biochemical reactions such as the ability to grow under anaerobic conditions, starch hydrolysis, catalase test, Voges–Proskauer test, casein hydrolysis, and nitrate reduction. To examine the ability of strains to grow anaerobically, isolates were inoculated on nutrient agar tubes containing the following ingredients per liter: 20 g of Trypticase, 5 g of NaCl, 10 g of glucose, 15 g of agar, 2 g of sodium thioglycolate, and 1 g of formaldehyde sulfoxylate). Additionally, 1 ml of nutrient broth culture was added to each tube. Then, the growth of the strains was evaluated by the naked eye after incubating the tubes for 4 days at 55°C. The ability of strains to hydrolyze starch was tested by growing them on nutrient agar plates containing 1% starch (pH 6.5) and incubating them at 55°C for 3 days. The appearance of a clear zone around the colonies upon the addition of Lugol's iodine indicated the presence of amylase activity and the ability of starch hydrolysis. Catalase reaction strains were subjected to tests to determine whether or not they could hydrolyze hydrogen peroxide by adding drops of 3% H₂O₂ on the surface of the bacterial colony, and positive strains that produced bubbles were recorded against those negative strains that could not produce bubbles.

The Voges–Proskauer (VP) test is used to detect the presence of acetoin in bacterial broth cultures. The test was conducted by adding alpha-naphthol and potassium hydroxide to the Voges-Proskauer broth, which had been previously inoculated with a bacterial strain. A cherry-red color indicates a positive result, while a yellow-brown color indicates a negative result (MacFaddin, 2000). Additionally, the ability of these bacteria to hydrolyze casein was demonstrated by the production of a clearing halo zone around their colonies after they were grown on skimmed milk agar plates. The composition of the agar plates included 1.0% skimmed milk, 0.2 % yeast extract, 0.01 % KH₂PO₄, 0.03 % K₂HPO₄, 0.5% NaCl, and 2% agar (pH 6.5). The plates were incubated at 55°C for 72 h. This hydrolysis ability was described by Fujio and Kume (Fujio and Kume, 1991). A bacterial colony that gives off a red color after cultivation in a broth medium containing 5g peptone, 3g beef extract, and 1g KNO₃, supplemented with naphthylamine and B sulfanilic acid, indicates positive nitrate reduction.

Bacterial genomic DNA was obtained from a bacterial isolate that was previously identified as B. thermophilius using a biochemical assay using the QIAGEN Genomic DNA Purification Kit according to the manufacturing procedure. Then, the pure DNA was evaluated under UV light after electrophoresis on a 0.8% agarose gel that had been supplemented with ethidium bromide in TBE buffer. 1 μl of pure bacterial DNA from this strain was subjected to RAPD-PCR. The RAPD-PCR method employed was described by Heikal et al. (Heikal et al., 2008). Four different primers, namely A1 (5,-GGACTGGAGTsGTGATCGC-3), Ez351 (5,-AGGAGGTGATCCAACCGCA-3), Rapid2 (5,-GTTACGCTCC-3), and A7a10 (5,-GAAACGGGTGGTGATCGCAG-3), were used. The RAPD-PCR reaction was conducted in a total volume of 25 μl, comprising 10 x buffer, 2 μl of 25 mM MgCl2, 2 μl of 2.5 mMdNTPs, 1 μl of each 10 pmol primer, 0.2 μl of Taq DNA polymerase (5 unites/μl), and 1 μl of 50 ng DNA as the template. The PCR reaction was conducted using the following program cycles; cycle at 95°C for 5 min; then, 40 cycles were performed as follows: For 1 min at 95°C for denaturation, 1 min at 30°C for annealing, 1 min at 72°C for elongation, and 10 min at 72°C for a final extension. Reaction mixtures were held at 4°C. The generated fragments were assessed on a 2% agarose gel containing ethidium bromide after electrophoresis in TBE buffer for 1 h.

To amplify the gene coding for 16S rRNA to identify the strain, primers 350F (5,-AGGTGATCCAACCGCA-3) and 350R (5,-AATGGAGGAAGGTGGGGAT-3) corresponding to the polymorphic region and conserved gene sequence of bacterial 16S rRNA (El-Hanafy et al., 2007; Mohamed et al., 2021, 2022a,b) were used to amplify ~350bp. The amplified fragment of 350 bp was assessed on a 0.8% agarose gel plus ethidium bromide after running in TBE for 30 min. After evaluating the targeted fragment of 16S rRNA, the fragment was purified using a PCR clean-up column kit (Maxim Biotech, Inc., USA).

The targeted and purified fragment of 16S rRNA was sent with the forward primer that was previously used for amplification to the Macrogen Company (South Korea) for sequencing. Sequences obtained from 16Sr RNA by the previously mentioned company were compared with the available sequences in the Gene Bank database using BLASTN searches at the National Center for Biotechnology Information site (http://ncbi.nlm.nih.gov).

Multiple DNA sequence alignments were carried out using the Clustal w program version 1.82 (http://www.ebi.ac.uk/clustalw) (Thompson et al., 1997; Abdel-Wahab et al., 2022).

Bacterial cells were lyzed in 1 mL of GS1 buffer after adjusting the cell number at a rate of 10⁶–10⁷ in sterile water. Then, total RNA was extracted from the treated and non-treated thermophilic bacterial strains using the GStract™ RNA Isolation kit II (GuanidiumThiocyanate) according to the manufacture's instructions, after taking into account that all the disposable plastics used in this method should be RNase-free, decontaminated, and treated with 0.1% v/v diethyl pyrocarbonate (DEPC) in water.

Sixty samples of sugar beet roots (10 beet roots for each sample) were collected from different companies in Dakahlia, Alexandria, Kafr El-Sheikh, and Sharkia and stored in mild conditions at Dakahlia Sugar Company for 29 days. Leaf petioles were trimmed at the base, and the terminal buds were removed. Then, the levels of sugar, Na, K, α amino-N, best quality, and inverted sugar, as well as temperature, were determined every 2 days during the storage period. Isolation, collection, and identification of the most common bacteria from all samples were achieved.

Juice samples were collected from the middle extraction tower of Dakahlia Sugar Company at different time intervals (4 h), transferred to the laboratory in sterilized bottles, and kept at 4°C until further analysis. Sugar percentage cossets, sugar losses, and sugar yield were estimated for every sample.

The levels of each apparent sugar (Na, K, α amino-N, beet quality, inverted sugar, and temperature) were determined every 2 days during storage. Sucrose was determined with a polar meter (sucromate), sodium (Na), potassium (K), and α-amino nitrogen were measured using the cold digestion method, while inverted sugars were detected using the Benzamidine method (Edye and Clarke, 1998; Larrahondo et al., 2006). Invert sugar percent was used as an indicator for sucrose losses. The Benzamidine method was used to estimate the reducing carbohydrate; to determine the optimal conditions of the benzamidine reaction, the reaction time, temperature, medium pH, reagent concentration, and the fluorescence spectrum have been investigated.

During beet sugar manufacturing, there are two types of losses: determined and undetermined. Determined losses were measured in the factory laboratory, and undetermined losses were calculated. Samples from pulp, mud, and molasses every day (during the campaign) were collected, and then the sugar level was determined by routine daily work. The average was determined every 10 days (period), and the undetermined losses and total losses were calculated.

Juice samples were collected from the BMA (Braunschweigerische Maschinenbauanstalt AG) extraction tower at Dakahalia Sugar Company. Juice samples were collected from the diffuser and incubated at 55°C. The lactic acid and pH were measured and collected using the lactic acid analyzer and pH meter (052-50910080, model 910/8) at different times. The drop in pH and increased lactic acid indicated high bacterial activity and sucrose destruction.

All cultivated plants were collected from the Botanical Gardens and Ornamental Plants Department, Horticulture Research Institute (HRI) Institute, Agricultural Research Center, Egypt, while wild plants were collected and identified in the Botany Department, Mansoura University, Egypt, by Prof. Ibrahim A. Mashaly (Professor of Flora and Plant Ecology), stored in the herbarium of the Plant Department, Mansoura University, Egypt, with their herbarium ID as shown in Supplementary Table S1.

The leaves of plants Labiatae (Ocimum basilicum L.; Mentha spicata L. & Mentha longifolia L.), Rutaceae (Citrus limon L.), Lauraceae (Cinnamomum zeylanicum L.), Umbelliferae (Cuminum cyminum L.), Myrtaceae (Eugenia caryophyllus L. & Eucalyptus rostrata Schlecht.), Geraniaceae (Geranium gruinum L.), Verbenaceae (Lantana camara L.), Gramineae (Cymbopogon proximus (hochst) staps), Solanaceae (Datura stramonium L. & Nicotiana glauca R.C. Graham), Asteraceae (Silybum marianum Gaertn L. & Calendula officinalis L.), Cannabaceae (Humulus lupulus L.), Anacardiaceae (Schinus terebenthifolius Radd), Apocynaceae (Nerium oleander L.), and Zingibiraceae (Zingiber officinale L.) were collected from cultivated fields to use them for excreting the active gradients to be used as antibacterial substances against bacteria that grow on sugar beet during storage or manufacturing. Plant materials were dried at room temperature for 15 days, ground using an electric mill to obtain a fine powder, and extracted by soaking in 1:1 (w/v) methanol and shaking for 3 days. The soaker was filtered through a specific filter under strong hand pressure, and the solvent was removed under vacuum at 60–65°C to produce a crude extract using a rotary evaporator (Alade and Irobi, 1993). The crude extract was preserved in the fridge for further analysis, dissolved in dimethyl sulfoxide DMSO (1mg/mL), and diluted extract solutions were used for different analyses (Taylor et al., 1996; El-Sherbieny et al., 2002).

After obtaining the diluted extracts from 20 medicinal plants that were dissolved in DMSO, which proved its neutral effect on the growth of tested microorganisms, as it did not make an inhibition zone for growing B. thermophilus, aliquots of these extracts with a concentration of 50 μg/ml were used as an antimicrobial product to reduce the growth of the tested microbe using the agar diffusion technique, as previously explained by Hili (Hili et al., 1997). To test each extract, 50 μl containing 50 μg/ml of active ingredients (in nutrient broth inoculated with microbes) was added to the well made by a cork borer on the surface of nutrient agar medium, and plates were incubated for 18h at 55°C (Burt, 2004; Holley and Patel, 2004). The inhibition of growth due to the addition of antibacterial products was recorded by eye compared to the control, and the diameters were measured. Similarly, 13 oils from Labiatae (Origanum majorana L.; Ocimum bacilicum L.; Thymus vulgaris L.&Mentha spicata L.), Myrtaceae (Eucalyptus rostratas chlecht; Eugenia caryophyllus L.&Cinnamomum zeylanicum Blume), Alliaceae (Allium sativum L.), Oleaceae (Olea europaea L.), Umbelliferae (Apium graveolens L.), Umbelliferae (Pimpinella schweinfurthii Asch), Asteraceae (Prunus dulcisvar L.), and Rutaceae (Citrus limon L.) that purchased from local Egyptian Market were examined against the growth of the tested bacteria.

The two effective oils of Mentha spicata and Ocimum basilicum against the growth of thermophilic bacteria were analyzed using Gas Chromatography and Mass Spectrometry (GC/MS) at the National Research Center, Dokki, Cairo, Egypt, using a Varian 3400 GC equipped with a DB-5 fused silica capillary column (30m X 0.25 mm; i.d. μm film thickness). The multi-step temperature program used for analysis was raised from 60°C (held for 3 min) to 260°C (held for 10 min) every 5°C min⁻¹. Helium was the carrier gas used at a flow rate of 1 ml min⁻¹, and the sample size was 1 μL (injector temperature was 250°C). A mass spectrometer version of the Varian-Finnigan SSQ7000 was used to determine each peak's mass with an ionization voltage of 70 eV. Scan time and mass range were 5 s and 40–400 m/z, respectively. Different oil peaks were nominated by matching their recorded mass spectra with those stored in the Wiley/NBS mass spectral library of the GC/MS data system and other published mass spectra (Adams, 2001).

Reverse transcription (RT), or first-strand DNA, was obtained after converting the mRNA to complementary DNA (cDNA) in the presence of dNTPs and reverse transcriptase. The components were combined with a DNA primer in a reverse transcriptase buffer for 1 h at 42°C. The exponential amplification via reverse transcription was conducted using a polymerase chain reaction (a highly sensitive technique), where a very low copy number of RNA molecules can be detected. A reverse transcription reaction was performed using an oligodT primer (5′-TTTTTTT -TTTTTTTT-3′). Each 25 μl reaction mixture contained 2.5 μl (5x) buffers with MgCl2, 2.5 μl (2.5 mM) dNTPs, 1 μl (10 pmol) primer, 2.5 μl RNA (2 mg/ml), and 0.5 units of reverse transcriptase enzyme. The PCR reaction was conducted using PCR programs at 42°C for 1 hr. and 72°C for 10 min (enzyme killing), and the product was stored at 4°C until use.

Primers RAPD1 (5′-TGCCGAGCTG-3′), RAPD2 (5′-ATGCCCCTGT-3′), RAPD3 (5′-AGCCACCGAA-3′), and RAPD4 (5′-CCTTGACGCA-3′) were included to conduct differential display analysis to examine genetic changes that occur due to the antibacterial effect (obtained as a result of oil addition) against the common microorganism that attacks sugar beet during storage or manufacturing (Bacillus thermophilus). The reaction of DD-PCR was carried out in 25μl containing 2.5 μl 10x buffer with MgCl2, 2μl (2.5 mM) dNTPS, 1 μl of 10 pmol primer, 1.5μl cDNA, and 0.2 μl (5 units/μl) Taq DNA polymerase. PCR is programmed as follows: 1 cycle at 95°C for 5 min, then 40 cycles for 30 sec at 95°C, 1 min at 30°C for annealing, and 1min at 72°C for elongation. The reaction was incubated at 72°C for 10 min for a final extension before being stored at 4°C until further use. Afterward, 10 μl of the differential PCR products were mixed with 2 μl of loading dye and electrophoresed at 80 volts in 0.5 x TBE running buffer. The gel was stained with 0.5 μg/cm3 (w/v) ethidium bromide and checked under UV light. Separated DNA bands of DD-PCR representing upregulated and downregulated genes were cut from the gel, purified using a gel extraction kit from Promega, USA, and sent with the primer to the same company mentioned before for sequencing.