Production of Xanthan Gum by Xanthomonas Campestris CCT 13951 Submerged Fermentation on hydrolysed agroindustrial by-products


 Xanthan is a natural polymer often obtained through fermentative processes using Xanthomonas, which remarkable properties - including biocompatibility, biodegradability, and non-toxicity – fit a myriad of industrial applications. The aim of this study was to evaluate the influence of different alternative fermentable substrates derived from agro-industrial waste (cheese whey and passion fruit peel), the production and rheological properties of the gum. The fermented substrate composition had a strong impact on production (8.15–14.81 g∙L-1) and apparent viscosity (31.9–510 mPa.s). The maximum xanthan gum production was observed after 72 h using cheese whey and passion fruit peel acid hydrolysate supplemented with K2HPO4 (AH:W – Phosphate). However, the maximum viscosity was found for medium without supplementation (AH:W), twice the value of the samples supplemented with phosphate. All solutions were highly pseudoplastic. This study provides a cost-effective solution for the reusing of cheese whey and passion fruit peel and possible low-cost approach for xanthan production. .


Introduction
Xanthan, a biopolymer with outstanding rheological properties, is the most commercially produced industrial gum and usually obtained through fermentation by Xanthomonas campestris. It was discovered in the 1950s at the Northern Regional Research Laboratories (NRRL) of the United States Department of Agricultures (28).
The biopolymer is extensively used as thickening or stabilizing agent in the food, pharmaceutical, and oil recovery industries. Xanthan gum (XG) is known for its unique rheological propertiespseudoplastic behaviourand physico-chemical characteristics, such as high viscosity at low shear, shear thinning, stability over a broad range of temperature and pH, high resistance to shear degradation in aqueous solutions, as well as its biodegradable and non-toxicity (22).
Commercial XG is currently produced through fermentation using glucose and sucrose, and available in the market at high prices (US$ 4000-5000/ton) (23). Thus, investigations regarding low-cost substrates that could be used to produce XG through fermentation are especially important because they may support the development of a bioprocess with the potential to reduce drastically its production costs and replace synthetic chemicals as standard feedstock.
Several different agroindustrial residues have been used as source of nutrients for the fermentation medium to produce xanthan, such as citrus waste (3); olive-mill waters (1), cheese whey (15), raw cassava starch (19), tapioca pulp (14), residue of apple juice (11), orange peel (26), palm date (29), shrimp shell (10) and sugarcane (15). However, additional nutrients are still required, as shown by Salah et al. (29). In their research, K2H2PO4, MgSO4, (NH4)2SO4, citric acid, H3BO3, ZnCl2, FeCl3 e CaCO3 were added to the medium when palm fruit waste was used as the alternative substrate, which improved the xanthan gum production but unfortunately increased its cost. Thus, the utilization of low-cost material as sole substrate is desirable, once their use can help to reduce the environmental impacted associated with their disposal (2).
Whey is the aqueous fraction of milk generated as by-product of cheese manufacturing.
This by-product represents about 90% of the milk volume and retains 55% of milk nutrients.
Approximately 50% of worldwide whey production is not treated and as a result a very large amount of material with potential value is wasted (8).
In Brazil, another agroindustrial residue with excellent fermentative potential is the passion fruit peel, a material rich in macronutrients and micronutrients (21). During the fruit processing for juice extraction, thousands of tons of peels and seeds are produced, accounting for approximately 70% of the total fruit weight (6). As Brazil is the world leader in the production of yellow passion fruit, exceeding 694,000 tonnes in 2015 (17), the disposal of this residue becomes both an environmental problem and a colossal waste of resources, since it is basically composed by carbohydrates, proteins and pectin (5).
Therefore, the use of agroindustrial wastes, from both economic and environmental point-of-views (2) is highly desirable, since it contribute towards the development of sustainable processes and reduce the production cost of several valuable bioproducts.
In this context, this study aimed to evaluate the use of agroindustrial residues (cheese whey and passion fruit peel) as alternative substrates in the production of XG by Xanthomonas campestris CCT 13951 and its effects on the gum's rheological properties.

Chemicals
All chemicals used in this work were of analytical grade and were obtained from Merck (Darmstadt, Germany) and Sigma (St. Louis, MO, USA).

Chemical Composition of the Substrate
Substrate humidity, total protein, and ash contents were determined according to Horwitz (16), while the total lipid content was determined as described by Bligh and Dyer (4).
Total carbohydrate content was calculated by difference, subtracting the humidity, protein, lipid, and ash fractions from the sample total mass. Total reducing sugars were quantified by the IAL method (24).

Microorganism
Xanthomonas campestris CCT 13951 was acquired from the Collection of the Tropical Cultures André Tosello Foundation (Campinas -São Paulo -Brazil). The strain was cultured on yeast malt (YM) medium containing (m/v) 0.3% yeast extract, 0.3% malt extract, 0.5% bacteriological peptone, 1.0% glucose. The maintaining culture was held in YMA, being added 2.0% agar to the standard YM. The mediums were sterilized at 121ºC for 15 minutes. The inoculum was prepared by growing the strain in a yeast malt (YM) medium at 28ºC and 150 rpm for 24 hours.

Substrate preparation
The cheese whey was provided by the Federal Institute of Sergipe (São Cristóvão, SE, Brazil). Upon collection, the cheese whey was pretreated at 90°C for 5 minutes, followed by filtration using cheese cloth and stored at -5°C until its use.
Yellow passion fruit (Passiflora edulisf. flavicarpa L.) was purchased from the local fruit market in Aracaju, Sergipe state, Brazil. The fruits were washed, cut and the pulp and seeds were removed. The peels were washed in tap water to remove the adhering pulp and soluble sugars. The passion fruit peels with mesocarp were ground by a mechanical mixer.
The obtained passion fruit extract was subjected to either acid or basic hydrolysis in a 1:10 ratio (grams of residue per gram of water). Acid hydrolysis was performed by adjusting the pH to 2.0 using a 2 • −1 H2SO4 solution, and the basic hydrolysis by adjusting the pH to 7.0 using a 2 • −1 NaOH solution. Both acid and basic suspensions were autoclaved at 121ºC for 15 minutes and subsequently filtered using a 102 qualitative filter paper, to remove insoluble solids. The acid and basic hydrolysed extracts were labelled as AH and BH, respectively.

Fermentation
Both hydrolysed extracts -AH and BHwere diluted using cheese whey (W) in a 1:3 The medium was inoculated with 20% (v/v) X. campestris culture. The fermentation was carried out for 96 hours at 30ºC and 250 rpm on an orbital shaker. Samples were collected every 24 hours to observe XG production and the conversion of total reducing sugars.

Recovery of Xanthan Gum
Samples collected during the fermentation were autoclaved and then centrifuged at 9,625xG for 15 min to remove the Xanthomonas campestris biomass. The XG was precipitated from the supernatants by adding ethanol 96% (3:1 v/v) to the fermentation broth. The solutions were maintained at 4ºC during 24 h and re-centrifuged. The precipitate was collected and dried at 50°C until constant weight, which allowed to calculate the XG concentration (in grams of gum per litre of culture medium) (10).

Rheology of Xanthan Gums
The rheological behaviour of the XGs produced by the X. campestris ATCC 13951 was evaluated by apparent viscosity ( • ) using a rheometer (Rheomat RM 300, Lamy Rheology, France) equipped with a concentric cylinder geometry coupled to a water bath. The apparent viscosities of the samples were determined at 25ºC and shear rates from 25 s -1 to 1000 s -1 . The XG solutions were prepared by adding 3% (m/v) XG in distilled water. The relationship between the shear rate and the viscosities for the XG solutions was described by the Ostwaldde-Waele kinetic model, as shown below.
Where µ is the apparent viscosity, γ is the shear rate, k is the consistency indexi.e. the viscosity measured at a shear rate of 1 s -1 , and n is the flow behaviour index. For shear thinning or pseudoplastic media, n<1. Eq. (1) assumes an absence of yield stress.
The pseudoplastic behaviour of the XG solution was confirmed by fitting the experimental data to the Oswald-de-Waele model, and thus estimating the values of the consistency index (K), flow index (n), and correlation coefficients (R 2 ) for each XG solution.

Spectroscopy of Fourier transform infrared (FTIR)
FT-IR analysis were performed with a Thermo Scientific Nicolet iS10 for all samplescommercial and produced by the authors -in the 400 • −1 to 4000 • −1 wavelength range using a 4 −1 step and 64 scans per sample. Three spectra were acquired sequentially for each sample. These spectra were processed and analysed using OriginPro 8.0 software.
3 Results and discussion 8

Agroindustrial Waste Composition and Proximate
As shown in Table 1, both cheese whey residue and passion fruit peels residues have similar protein and carbohydrate composition but vary significatively in the lipid and inorganic fractions. The centesimal composition of the fermentable substrates is essential to study the XG production, given that nutrients and micronutrients influence both cell growth of Xanthomonas bacteria and XG production during the fermentation (9). Druzian and Pagliarini (11) have proved that most of the carbohydrates present on passion fruit peels are in fact pectin, a galacturonic acid polymer that can be broken down by Xanthomonas to its monomers and then be used as a source of carbon during fermentation. Bilanovic et al. (3) investigated the use of citrus waste as a low-cost substrate for XG production by converting it to pectin, hemicelluloses, and cellulose fractions. The results also proved that the pectin present in the fermentation medium was consumed by the microorganism and subsequently converted to XG.
Kalogiannis et al. (18) studied the XG production by X. campestris ATCC 1395 using pre-treated molasses from sugar beet as carbon source, supplemented with K2HPO4, yeast extract, Triton 80, and water. Addition of K2HPO4 to the medium had a significant positive effect on XG production. These studies indicated that the addition of phosphates enhanced XG production both because the medium was deficient in phosphates and because they act as a buffering agent, reducing the pH fluctuations. Figure 1 shows the production of XG over time by X. campestris ATCC 13951 for both pre-treatment methods (acid hydrolysis and basic hydrolysis), as well as for mediums supplemented and with no supplementation. The supplementation of the medium and the pretreatment of extract had a strong effect on XG production, which ranged from 8.15 • −1 to 14.81 • −1 .

Xanthan Gum Production
The acid-hydrolysis method reached higher XG production, even without supplementation, when compared to the other substrates. The maximum XG production -15 • −1was observed at 72h of fermentation for an acid-hydrolysed pre-treated substrate supplemented with K2HPO4 (AH:W -Phosphate) (Fig.1A).
[  proportionally correlated to the xanthan production. In all cases, Xanthomonas campestris CCT 13951 was able to convert over 70% of reducing sugars, which associated to XG production curves indicates that most of the reducing sugar was assimilated for the xanthan production.

[FIGURE 2]
Li et al. (23) reported that the reducing sugars conversion rate was rather low (53.88%) and likely used for cell reproduction and respiration, whereas higher concentrations (higher than 50 • −1 ) of a carbon source could inhibit the formation of xanthan product.

Rheological properties
The Figure 3 shows the viscosity behaviour of the produced XGs according to the range of shear rate used during the rheological assay. The shear rate dependence of viscosity was fitted by the Ostwald-de-Walle model, according to Eq. (1). The parameters estimated by the Levenberg-Marquardt algorithm in Statistic 6.0 (Statoil Inc) are presented in Table 2.

[TABLE 2]
According to Garcia-Ochoa et al. (12) xanthan solutions have a non-Newtonian rheology, in which the apparent viscosity decreases as shear rate increases. One important property of XGs is the ability to modify the rheological behaviour of solutions. The rheological properties of XGs are mainly related to its chemical structure, molar mass, molecular arrangement and molecular bonding, which will vary according to the chosen microorganismand respective strainas well as with the substrate used.
The highest viscosity value (510 mPa.s) was obtained for the gum from acid pretreatment and without supplementation (AH:W) at shear rate of 25 s -1 , which is twice of those values obtained for samples supplemented with phosphate.

[FIGURE 3]
Silva et al. (30) determined the apparent viscosity of XG using cheese whey as substrate by two strains of X. campestris. The viscosity obtained was about 60 mPa.s (25ºC and shear rate 12.5 s -1 , solution 3% (w/v). Being below the one found by this work, except for the sample acid pre-treatment and supplemented with sucrose (AH:W-Sucrose). Nitschke et al. (27) determined the apparent viscosity at 1% aqueous gum synthesized in whole medium cheese whey at a shear rate of 10s -1 at 25 ° C and the final viscosity observed was about 9508 cP. Costa et al. (7) evaluated the yields of XGs produced from shrimp shell by three native strains of X.
The 3% XG solutions obtained from the whey and passion fruit peel followed the model previously described, all the correlation coefficients (R 2 ) were close to 0.99, independently of the fermentative medium. In all cases, the value for was smaller than unityimplying pseudoplastic behaviourand ranging between 0.398 and 0.739 for the gum produced with acid pre-treatment, and between 0.488 and 0.545 for that obtained using alkaline pre-treatment.
While a comparison between XG production or its apparent viscosity could be easily provided, no method has been described in the literature that could consider both parameters during a process design or a product development. Therefore, the result variations observed throughout the literature are expected and a consequence of discrepancies in the strain used, the medium composition from different agroindustrial residues, and the fermentation conditions.

Spectroscopy of Fourier transform infrared (FTIR)
Samples of commercial xanthan (CX), AH:W and BH:W were analysed to identify the functional groups present in the structure of these biopolymers. The region studied included all the spectral bands located in the window between the wave numbers 400 and 4000 cm −1 . Fig. 3 shows the infrared spectra of samples. Xanthan samples showed the presence of hydroxyl (l3200 cm -1 ), carbonyl (1400 cm -1 , carboxyl (1600 cm -1 ) e acetal groups (1050 cm -1 ). Fig. 4 shows that the infrared spectrum of the CX is remarkably like that obtained by X. campestris 13951 using the AH:W and BH:W substrates. The decrease in transmittance intensities observed throughout the spectra means an increase in the amount (per unit volume) of the functional group associated with the molecular bond, while a shift in peak position usually means that the electron distribution or hybridization state in the molecular structure has changed.

[FIGURE 4]
Xanthomonas Campestris CCT 13951 has proven to be able to produce xanthan gum using an agroindustrial hydrolysatecomposed by passion fruit peel and cheese wheyas substrate. Both the substrate pre-treatment method and its supplementation showed significative effects on the xanthan gum production and its apparent viscosity.
The highest concentration of xanthan gum 15 • −1 was obtained after 72h of fermentation and using an acid-hydrolysed substrate supplemented with K2HPO4. The maximum apparent viscosity observed in the study was 510 • at 25°C and using 3% of a gum also produced using the acid-hydrolysed substrate, but with no supplementation, and which gum production was 12.38g.L -1 . The results show that higher production yields are not correlated to the gum quality. The results obtained in this study, although preliminary, provide a projection of the potential of using fermentation substrates composed only by agroindustrial residues in the production of xanthan gum.

Ethical Approval and Consent to Participate
Not applicable. The manuscript does not contain data collected from humans or animals.

Consent to Publish
Not applicable. The manuscript does not contain any individual person's data.

Authors Contributions
JCS and RRS conceptualized the study. JCS performed the main experimental work, evaluated the dataset, and wrote the main text of the manuscript. EBT, GFS and DFC were involved in proofreading of the manuscript.

Funding
Authors would like to thank FAPITEC (Fundação de Apoio à Pesquisa e Inovação for the financial support.

Competing Interests
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Availability of data and materials
The datasets used during the current study are available from the corresponding author on reasonable request.      Table 1: Physicochemical characterization of the cheese whey and passion fruit peel used as a substrate for fermentation in the production of xanthan gum by strains of Xanthomonas campestris.

Analysis Composition
Cheese whey  Figures Figure 1 Effect of the supplementation of the medium and the pre-treatment of extract on xanthan gum production: acid hydrolysis (A) and alkaline hydrolysis (B).

Figure 2
Effect of the supplementation of the medium and the pre-treatment of extract on total reducing sugars and reducing sugar conversion: acid -hydrolysis (A and B) and alkaline -hydrolysis (C and D).  FT-IR analysis of a commercial xanthan sample (CX) and samples produced using the Passion Fruit Peel / Cheese Whey substrate hydrolysed using both the acid (AH:W) and basic (BH:W) methods.