Increased sedimentation of a Pseudomonas-Saccharomyces microbial consortium producing medium chain length polyhydroxyalkanoates☆

Chang Liu ,Lin Qi ,Songyuan Yang ,Yun He ,Xiaoqiang Jia,2,3,*

1 Department of Biochemical Engineering,School of Chemical Engineering and Technology,Tianjin University,Tianjin 300072,China

2 Key Laboratory of Systems Bioengineering (Tianjin University),Ministry of Education,Tianjin 300072,China

3 SynBio Research Platform,Collaborative Innovation Center of Chemical Science and Engineering,Tianjin 300072,China

ABSTRACT Concerns about feasibility,separability,settleability,efficiency once hampered studies on polyhydroxyalkanoates(PHAs)production,which mainly focused on single strain microorganism or activated sludge rather than artificial microbial consortia.Here,a medium chain length PHAs (mcl-PHAs) producing Pseudomonas-Saccharomyces consortium with xylose as the main substrate was studied.Mcl-PHAs accumulation increased from 12.69 mg·L-1 to 152.3 mg·L-1 without any optimization method.The presence of Saccharomyces cerevisiae,though in a relatively low concentration,improved the sedimentation of cell mass of the mixed culture by 60%.Reasons for better sedimentation of the consortium were complex:first,the length of Pseudomonas putida increased two to three times in the consortium;second,the positive surface charge of P.putida was neutralized by S.cerevisiae;third,the adhesion proteins on the surface of S.cerevisiae interacted with the P.putida.

Keywords:Medium chain length polyhydroxyalkanoates Pseudomonas-Saccharomyces consortium Sedimentation Xylose

1.Introduction

Polyhydroxyalkanoates (PHAs),intracellular polyesters serving as carbon storage granules mostly under unfavorable conditions,have become a promising environmentally-friendly substitute to traditional chemical plastics [1].Generation of PHAs is gaining increasing importance [2],but the high production costs make PHAs production less successful in large-scale industrial settings[3].To cut the cost of substrates,which accounts for nearly 50%of the total cost of PHAs production [4],researchers have shifted their attention from expensive substrates such as fatty acids to non-staple raw materials like xylose [5].Studies of employing xylose as substrate during PHAs production are no longer rare,wild type Cupriavidus sp.[6],engineered Ralstonia eutropha [7],and microbial mixture [8]are successful examples.In instances above,the use of unmodified single strain microorganism requires special optimization of the cultivation progress during PHAs production,while strains with genetic engineering bear a metabolic burden.As for natural microbial consortia,they usually demand too long domestication period (about tens to hundreds of days) to obtain a relatively high content.Distinct from the above,artificial microbial consortia are not subject to these disadvantages.However,although benefits of employing artificial microbial consortia are obvious,i.e.better metabolic load distribution,increased resistance to system disturbance [9],there have been seldom reports for employing such consortia to generate PHAs.In one of the few examples,Ganduri et al.co-cultivated Lactobacillus delbrueckii and R.eutropha by controlling the mixing intensity,and showed that equal high yields of PHB were produced as comparable as the cultivation of single-strain R.eutropha [10].

It should be pointed out that PHAs’yields vary widely depending on their types.Specifically,short chain length PHAs(scl-PHAs)can reach to tens of g·L-1easily(a representative producer is R.eutropha)[7].However,mcl-PHAs are hard to achieve high production as scl-PHAs,and such PHAs are usually several hundred mg·L-1(a representative producer is P.putida) [11].A comparison among some mcl-PHAs producers is presented in Table 1.It is obvious that production of mcl-PHAs is low in general,hence,improving producing efficiency and lowering costs are significant for mcl-PHAs[12].

Usually,the first step in cell recovery is sedimentation,which is one of the important steps in downstream PHAs processing,and isgenerally performed by centrifugation [18].Effective bacterial cell sedimentation has been suggested as a promising way to minimize overall energy demand for cell recovery[19].According to Ryu et al.,energy consumption accounts for only 3%-11%of that without adding flocculants,suggesting that better sedimentation could significantly reduce the separation cost in centrifugal separation [20].However,cells with small,light and lower concentration make separation via continuous centrifugation being a long and energy intensive process,and thus it is crucial to enhance settleability by enlarging cell sizes,adding flocculants [21],changing hydrophobic/hydrophilic properties [22],producing bioflocculant [23],or even displaying surface adhesive proteins [3].In contrast to the rarely-reported studies on sedimentation in PHAs-producing strains,studies of flocculation in S.cerevisiae are much more systematic[24].In brief,flocculation in S.cerevisiae is based on cell-cell and cell-surface Flo adhesion proteins.Of these proteins,Flo1p is directly related to the degree of flocculation due to its highly repetitive sequence [25].In addition,transcriptional activators encoded by the flo8 and mss11[26]are essential for expression of flo1[27].

Table 1 Comparison of mcl-PHAs production in several producers

As previous studies have shown that P.putida and S.cerevisiae have positive interactions[28],we aimed to use these species to construct an artificial two-species microbial consortium,and to test whether this consortium has superior performances as compared to P.putida pure culture.Maybe there is a concern about application that whether the consortium would perform in real cheap mix,lignocellulosic materials containing xylose,for S.cerevisiae is susceptible to toxic compounds in hydrolysates.However,S.cerevisiae can be genetically modified to make use of xylose[29];more importantly,the tolerance of S.cerevisiae could be increased by many methods,like upregulating genes involved in cofactors generations or membrane robustness [30],as well as adopting genome engineering[31]to inactivate or disrupt some specific genes[32,33].

Aiming at the sedimentation characteristic,we explored it with respect to three factors:P.putida size,zeta-potential measurement,and inactivation of flocculation genes in S.cerevisiae.By analyzing these characteristics of the Pseudomonas-Saccharomyces consortium,we would like to present interactions between two members of a microbial consortium from a physical point of view.

2.Experimental

2.1.Strains and growth conditions

P.putida NBRC14164 was purchased from the China Center of Industrial Culture Collection (No.10298) and it worked as the mcl-PHAs-producing strain.Dr.Yingjin Yuan of Tianjin University,China,donated a strain of S.cerevisiae SyBE_Sc01020078 modified to have excellent xylose-utilizing ability with lactic acid as its main product (data unpublished).

P.putida stored at -80°C was activated in 50 ml Luria Bertani(LB)medium for 24 h and S.cerevisiae was activated in 50 ml Yeast extract Peptone Xylose (YPX) medium for 36 h,respectively.Both strains were cultivated at 30°C and 200 r·min-1.LB medium contained (per liter culture):NaCl 10 g,peptone 20 g,yeast extract 5 g,pH 7.0.YPX medium contained (per liter culture):xylose 20 g,peptone 20 g,yeast extract 10 g,the pH was natural.P.putida and S.cerevisiae was counted on LB plate and YPX plate by diluting to 10-7-10-3according to their concentration.

2.2.Preparation of mixed-cultures of mcl-PHAs-producing Pseudomonas-Saccharomyces

The two harvested strains were centrifuged at 8000 g for 2 min.Then they were resuspended with distilled water and were inoculated into 100 ml Mineral Salt (MS) medium containing (per liter culture):(NH4)2SO43.0 g,MgSO4·7H2O 0.5 g,KH2PO41.5 g,K2-HPO4·3H2O 3.0 g,1 ml of microelement solution,pH 6.0.The microelement solution contained (per liter culture):FeSO4·7H2O 2.78 g,MnCl2·4H2O 1.98 g,CoCl2·6H2O 2.38 g,CaCl2·2H2O 1.47 g,CuCl2·2H2O 0.17 g,and ZnSO4·7H2O 0.29 g [34].We used 20 g xylose·L-1as the carbon source,and 5 g octoate·L-1as the precursor of mcl-PHAs.We kept P.putida as 1% v/v inoculation,while S.cerevisiae was set with 1%,5%,and 10% v/v inoculation.In addition,single P.putida strain and S.cerevisiae strain were cultivated under the same conditions as a control.Strains in the consortia were separated for counting by nutritious plate,or rather,LB enriched P.putida,while YPX enriched S.cerevisiae.All experiments were replicated three times,and were subject to t-Test.

2.3.Xylose and mcl-PHAs detection

Xylose was quantified using a high performance liquid chromatograph (Agilent,USA) equipped with an Aminex HPX 87 H ion exclusion column(300 mm×7.8 mm,Bio-Rad,USA).The chromatograph was operated at 65°C with a flow rate of 0.6 ml·min-15 mM H2SO4using a differential refraction detector.

The extraction and detection of mcl-PHAs were performed as described previously[9].Bacterial cells were harvested by centrifugation at 10000 g for 10 min,then washed with distilled water,after lyophilizing cells were subjected to methanolysis at 100°C for 4 h in the presence of 3%(v/v)H2SO4.Benzoic acid was used as an internal standard.Gas chromatography (GC) (Bruker,Germany) equipped with a Rxi-1HT column (30 m×0.32 mm×0.25 μm,Restek,USA)was used for this study.The resulting 1 μl of the mcl-PHAs methyl esters was injected to the GC.The column temperature was started at 80°C for 1.5 min,then increased to 140°C at the rate of 30°C·min-1without maintaining,next increased to 240°C at the rate of 40°C·min-1and was maintained there for 4 min,the detector was FID and its temperature was 250°C,the carrier gas was nitrogen and the split ratio was 10:1.

2.4.Sedimentation analysis

2.4.1.Bacteria size and zeta potential

At the end of cultivation,strains were washed with distilled water and observed with an optical microscope (Nikon,Japan).To increase legibility and obtain the exact length,strains were also subjected to microscopic characterization using a scanning electron microscope (Hitachi,Japan),the method was described by Bozzola and Russell [35].

Before measuring zeta potential,all of the strains were washed with distilled water to avoid interference form particles in the fermentation broth.Zeta potential was measured with a zeta meter(Malvern,UK) as described by Paquot et al.[36].

2.4.2.Polyacrylamide sedimentation

At the end of cultivation,pH values of all media were detected.Cationic/anionic polyacrylamides (CinaFloc,China) were added to the control strain (P.putida pure culture) as polyelectrolytes with different concentrations in the medium (0.01%-0.10% w/v).No polyacrylamides were added to the consortia.After adding the polyacrylamides,solutions were stirred for 10 min.All suspensions were centrifuged at 10000 g for 1 min.Besides,P.putida and S.cerevisiae cultivated alone were mixed according to their final concentration in the 1:10 consortium to determine the sedimentary efficiency,too.Sedimentary efficiency R was defined according to the formula [37].

where OD1was the OD600of centrifugal supernatant and OD2was the OD600of the well-stirred medium.

2.4.3.Inactivation of flocculation-related genes in S.cerevisiae

Inactivation of the flocculation gene flo1 and the transcriptional activators flo8 and mss11 in S.cerevisiae was conducted individually by replacing part of the genes with exogenous fragments,which were also selection marker genes as shown in Table 2.These genes were inactivated by overlap extension PCR.To be specific,met15,his3 and leu2 was integrated into flo1,flo8,and mss11,respectively,thus created silence genes with substituted fragments and consequently destroyed their function.Primers used in this study are listed in Supplementary Table 1.Transformation of DNA fragments was carried out with LiOAc method [38].

3.Results and Discussion

3.1.Construction and characterization of the Pseudomonas-Saccharomyces consortia

Xylose utilization,cell dry weight (CDW),lactic acid residual,mcl-PHAs production and mcl-PHAs content(wt%)were measured at 48 and 72 h for all of the consortia and the single P.putida strain.P.putida did not consume xylose well,and as a result it grew poorly.In the consortia,as S.cerevisiae concentration increased,more xylose was consumed and growth of the consortia was also improved,accompanying with more lactic acid residual was detected (Fig.1).Indeed,the S.cerevisiae strain has the ability to utilize xylose efficaciously,and its main product is lactic acid,a preferable substrate of P.putida (data not shown),contributing to cell growth better than xylose dose,thus leading to higher xylose utilization efficiency of the microbial consortia.

However,neither xylose nor lactic acid could be converted to precursors of mcl-PHAs without special domestication or engineering[4,39].In fact,sugars,including the easily used glucose for vast species,cannot be converted into mcl-PHAs’ precursors unlessthere is 3-hydroxyacyl-CoA-acyl carrier protein transferase in strains [40,41].This means the above substrates could only serve as unrelated carbon source and only be responsible for biomass growing in this study,so they have nothing to do with mcl-PHAs accumulation.As a result,octoate was added as a precursor of mcl-PHAs.While octoate was a related carbon source for P.putida,there was sufficient xylose act as substrate for growth,so octoate was mainly used for synthesizing mcl-PHAs [4].In Fig.1,mcl-PHAs accumulation and content of the consortia (152.3 mg·L-1,11.4%) were greater than those in the P.putida pure culture(12.69 mg·L-1,2.66%).However,from the point of mcl-PHAs content,the 1:1 consortium behaved the same tendency with the single P.putida,their mcl-PHAs percentage both declined from 48(Fig.1a) to 72 h (Fig.1b).By contrast,consortia with more S.cerevisiae kept mcl-PHAs from being degraded,and even increased slightly between 48 and 72 h.The continued production of mcl-PHAs over 72 h indicated that S.cerevisiae contributed to store mcl-PHAs steadily.It also demonstrated that P.putida was not killed by S.cerevisiae even though the inoculation ratio was reached 1:10 v/v,and this was confirmed by observation under microscope.It is worth mentioning that according to Table 1,mcl-PHAs production in Pseudomonas genus is between 100 and 600 mg·L-1,so their accumulation in our study,even if low,is not too little to be of value.Besides,reasons for low mcl-PHAs production were that neither the consortium was cultivated under favorable PHAs accumulation conditions such as nutritional limitation [6],nor the P.putida was genetically modified like expression of phaG[42].So the yield has potential to be increased by adopting any optimization.Moreover,actual values of mcl-PHAs content in the consortia should be higher than the calculated ones,because the CDW was weighted as a mixed mass of the two strains.

Table 2 Plasmids and strains used in this study

Fig.1.Mcl-PHAs production,xylose utilization,PHAs content,CDW and lactic acid of Pseudomonas-Saccharomyces consortia at 48 h (a) and 72 h (b).

Fig.2.Colony counting of P.putida and S.cerevisiae by serial dilution method,bacterial suspension used to spread on the plates was 100 μl.(a)P.putida counting in LB plates at 48 h.(b) S.cerevisiae counting in YPX plates at 48 h.(c) P.putida counting in LB plates at 72 h.(d) S.cerevisiae counting in YPX plates at 72 h.

In order to determine the approximate proportion of the two microbes,we counted colonies by serial dilution method (Fig.2).At the beginning of co-cultivation,P.putida was about 106cfu·ml-1,equivalent to the number of S.cerevisiae in the 1:10 consortium.However,the ratio changed dramatically after cultivation.At 48 h,P.putida had reached 108cfu·ml-1(Fig.2a),but S.cerevisiae decreased to 105cfu·ml-1(Fig.2b),only 0.1% of the concentration of P.putida.It seemed that S.cerevisiae was gradually perishing,unexpectedly,at 72 h,S.cerevisiae rebounded to 106cfu·ml-1(Fig.2d),and P.putida also increased tenfold to 109cfu·ml-1(Fig.2c).It was then understandable that the continuous accumulating of mcl-PHAs in the 1:10 consortium was due to the growing number of P.putida.Besides,the changing of S.cerevisiae colony from 48 to 72 h implied that P.putida did favor to S.cerevisiae resuscitation.It is interesting to note that,despite the limited concentration of S.cerevisiae,this consortium displayed better sedimentation performance than the P.putida pure culture(Supplementary Fig.S1).In order to investigate differences between the single P.putida strain and the 1:10 consortium,experiments below were performed.

3.2.P.putida size

Size of P.putida at 72 h was observed with optical and scanning electron microscopy.When co-cultivated with S.cerevisiae,it was obvious that the length of P.putida was 4-5 μm (Fig.3a,b),which was 2-3 times of the length when cultivated alone (Fig.3c,d).

One speculation of the different finding may derive from that S.cerevisiae supplied some additional nutrition to P.putida and thus contributed to its better growth,but this speculation was disproved as P.putida did not achieve an equivalent length when grown in the highly nutritious LB medium (Fig.3e).

Another reason could be the effect of mcl-PHAs percentage,which was 2.66% in the P.putida pure culture and 11.4% in the 1:10 consortium.It seems that there is a positive correlation between PHAs content and cell size:the more PHAs granules,the bigger cell volume.This was confirmed by Wang et al.[43],who enlarged E.coli by overexpressing the sulA gene to inhibit cell division FtsZ ring assembly,leading to the formation of filamentary E.coli that had elongated cell size which was visible under SEM.The increasing cell volumes contributed to higher PHAs accumulation and better precipitation when compared to rod shape E.coli.However,as the complete genome sequence of P.putida strain NBRC14164 was only recently published [44],and we are using it here for the first time as a mcl-PHAs-producing strain,little information is available about its mcl-PHAs related genes or division controlling genes,further studies of genes in P.putida would be very valuable.In any case,enhanced sedimentation ability due to enlarged cell size allowed for easy biomass recovery [45].

3.3.Effects of polyacrylamides and S.cerevisiae on P.putida sedimentation

During the process of centrifugation,despite speed was high,the rotating time was not sufficient for all of microbial suspension to settle.Hence,sedimentary efficiency was calculated according to the changes of OD600before and after centrifugation.In consortia,as the concentration of S.cerevisiae increased,the sedimentary efficiency increased a lot.As in Table 3,this improvement was similar to the effect of adding 0.01% -0.05% anionic polyacrylamide.The more anionic polyacrylamide there was,the more anions in the solution,however,excessive flocculant (0.1%) made the solution so viscous as to suppress the settlement.The 0.01%-0.05%range showed positive correlation to the sedimentary efficiency,according to which we assumed that P.putida acted like cations.If this is true,then it would be comprehensible that why the cationic polyacrylamide was ineffective or even counterproductive based on the principle that same charges repelling each other.To determine the electrical property of single strains and the consortium,their zeta potentials were measured and were presented in Table 4.There was one positive number+2.03 mV,indicating that only the surface of single P.putida was positively charged.This finding is consistent with the isoelectric point theory:when environmental pH is lower than pI,a given material possesses a positive charge.At the end of cultivation,the pH of the medium was about 3.7 (Supplementary Table S2),which was lower than the pI of P.putida (4-5,according to the gram staining method).Zeta potential of single S.cerevisiae was -27.4,these cells around P.putida neutralized the positive charge,resulting in electrically neutral surfaces that did not repulse each other,and eventually tended to subside.Zeta potential of the mixture consisting of separately cultivated P.putida and S.cerevisiae as well as zeta potential of the 1:10 consortium were both electronegative,which suggested that S.cerevisiae influenced the surface charge of this two systems dramatically.However,it is interesting to note that the two systems differ in the potential value,even though their concentration of the two microbes are the same.This means that microbial consortium is not equal to simply blending two strains together,and that there are some amazing interactions which can change the external physical properties of the consortium.We therefore conclude that,in terms of surface charge,S.cerevisiae plays a similar role to an anionic polyacrylamide.

Fig.3.Different length of P.putida.(a),(b)P.putida size in the 1:10 consortium under optical microscope and scanning electron microscope.(c),(d)P.putida size of the pure culture under optical microscope and scanning electron microscope.(e) P.putida size of the pure culture in the LB medium.

Table 3 Sedimentary efficiency of cationic/anionic polyacrylamide on P.putida

Table 4 Zeta potential of single strains and their mixture and co-culture

From the change of the Zeta potential of single P.putida and the microbial consortium,it was obvious that the surface charge of P.putida was strongly affected by S.cerevisiae,namely there was physical interaction between members in the microbial consortium.This provide another perspective for the study of communication between microorganisms,which is different from interactions based on the chemicals,such as metabolic exchanges[46],co-factor communication [47],quorum sensing [48],etc.

3.4.Inactivation of flocculation related genes in S.cerevisiae

Results of gene inactivation are presented in Supplementary Fig.2.After inactivation of flo1,flo8 or mss11 in S.cerevisiae,sedimentation of P.putida was reduced (Table 5,Supplementary Fig.S3).Among the three genes,flo1 is the one coding adhesion proteins on the cell surface,while the other two genes are transcriptional regulation genes,which works in activating expression of flo1 and another flocculation functional gene flo11.Therefore,flo8 Δ and mss11 Δ had stronger impact on the flocculation than flo1 Δ.With impaired flocculability of S.cerevisiae,sedimentation ability of the 1:10 consortium was also decreased.

This illustrates the important role that surface adhesion proteins encoded or regulated by flocculation genes played on the consortium.A widely accepted mechanism of flo protein cell-cell adhesion is the lectin hypothesis [49].That is,in the presence of Ca+,adhesion proteins are able to bind highly branched mannose polymers located in the cell walls of adjacent cells [50].Except the degraded sedimentation of S.cerevisiae itself,we speculate that there are receptors or binding proteins on the cell walls of P.putida which act like mannose polymers.Although the published genome of P.putida strain NBRC14164 possesses dozens of receptors and hundreds of binding proteins (https://www.ncbi.nlm.nih.gov),few studies have focused on cell wall proteins,even fewer have investigated cell wall function of P.putida as a member of synthetic microbial consortia.Hence,it deserves further study in this field.

3.5.Comparison of the mixture of individually growing strains and co-culture

In order to figure out the enhanced sedimentation was a coculture effect or just originated when P.putida and S.cerevisiae were put together,these two strains were cultivated separatelyand then were mixed according to their final concentration as in the 1:10 microbial consortium.The consequent sedimentary efficiency was 73%(Supplementary Fig.S4),though higher than single P.putida,was much inferior to the consortium in which the two strains grew together.This manifested that S.cerevisiae,although functioned to some degree,could not improve the sedimentation instantly,which indicated that it affected P.putida in different ways:the one requested a long time to evolve their interaction,such as the size change,while the other was similar to contacttriggered effect like the charge neutralization.However,except physical interactions above,biochemistry influence (flo proteins)should not be ignored,too.Therefore,the better sedimentation of the 1:10 consortium was a real combination result,rather than a phenomenon originated by simply blending the cells together.

Table 5 Sedimentary efficiency of 1:10 consortia after S.cerevisiae genes inactivation

4.Conclusions

Here,we constructed a mcl-PHAs-producing Pseudomonas-Sac charomyces consortium with xylose as the main substrate for the first time.Strains in this consortium showed the reciprocal symbiotic relationship.The consortium accumulated more mcl-PHAs than P.putida pure culture,although the yield was still low,it had the potential to be improved because strains were cultivated without any optimization.Except mcl-PHAs accumulation,the consortium also possessed excellent sedimentation efficiency.Reasons for this increased settleability were explored.First,it was observed that settlement of S.cerevisiae was good and its size was bigger than P.putida.Although we suspected that S.cerevisiae acted like magnets encircled by P.putida,and formed flocks in the consortium,microscopic examination revealed an increase in the length of P.putida instead of any flock.As the increase in P.putida length could not entirely explain the improvement in sedimentation,an assumption was theorized that S.cerevisiae was acting like flocculant.This idea was supported by the difference in zeta potential (as a proxy for surface charge) among the single strain of P.putida,S.cerevisiae,their mixture and co-culture.In addition,it was found that adhesion proteins on the cell walls of S.cerevisiae had direct relationship with sedimentary performance:sedimentary efficiency decreased with the inactivation of the flocculation-related genes.We thus inferred that there were receptor proteins on the surface of P.putida combining with the adhesion proteins on S.cerevisiae.

Causes of better sedimentation of the consortium are multifactorial,including physical interactions as well as biochemistry influence.It is possible that there are other explanations for this enhanced sedimentation,such as the presence of particular fermented liquid ingredients,and other P.putida surface characteristics such as hydrophobicity-hydrophilicity.Nevertheless,our analysis has highlighted new aspects of the interactions of members in consortia.Benefits of adopting this consortium include not only enhanced mcl-PHAs accumulation and cells settleability,but also reduced energy consumption without adoption of any other pretreatment methods.More advantages of the Pseudomonas-Saccharomyces consortium need to be explored,and more comprehensive information systems also need to be built.We plan to expand our work in future to investigate metabolic network construction,cofactor flow distribution and so on.

Acknowledgements

The authors are very grateful for the kind donation of xyloseutilizing S.cerevisiae from Dr.Yingjin Yuan of Tianjin University(China),and for the kind donation of plasmid Pxp320 and Prs405 as well as strain S.cerevisiae W303-1A from Dr.Wenyu Lu of Tianjin University (China).

Supplementary Material

Supplementary material to this article can be found online at https://doi.org/10.1016/j.cjche.2018.11.013.