Biotechnology for Biofuels

Biotechnology for Biofuels

Enhancement of docosahexaenoic acid production by overexpression of ATP-citratelyase and acetyl-CoA carboxylase in Schizochytrium sp.

--Manuscript Draft--

Manuscript Number: BBIO-D-19-00571R1

Full Title: Enhancement of docosahexaenoic acid production by overexpression of ATP-citratelyase and acetyl-CoA carboxylase in Schizochytrium sp.

Article Type: Research

Section/Category: Bacterial and fungal/yeast genetics, physiology, and metabolic engineering

Funding Information: National Natural Science Foundation ofChina(No. 31470190)

Dr. Zhi Chen

Abstract: Background: Docosahexaenoic acid (DHA) is an important omega-3 long-chainpolyunsaturated fatty acid that has a variety of physiological functions for infantdevelopment and human health. Although metabolic engineering was previouslydemonstrated to be a highly efficient way to rapidly increase lipid production, metabolicengineering has seldom been previously used to increase DHA accumulation inSchizochytrium spp.Results: Here, a sensitive β-galactosidase reporter system was established to screenfor strong promoters in Schizochytrium sp. Four constitutive promoters ( EF-1α p ,TEF-1 p , ccg1 p , and ubiquitin p ) and one methanol-induced AOX1 promoterwere characterized by the reporter system with the promoter activity ccg1 p > TEF-1p > AOX1 p (induced) > EF-1α p > ubiquitin p . With the strong constitutivepromoter ccg1 p , Schizochytrium ATP-citrate lyase (ACL) and acetyl-CoAcarboxylase (ACC) were overexpressed in Schizochytrium sp. ATCC 20888. The cellswere cultivated at 28°C and 250 rpm for 120 h with glucose as the carbon source.Shake flask fermentation results showed that the overexpression strains exhibitedgrowth curves and biomass similar to those of the wild-type strain. The lipid contents ofthe wild-type strain and of the OACL, OACC, and OACL-ACC strains were 53.8, 68.8,69.8, and 73.0%, respectively, and the lipid yields of the overexpression strains wereincreased by 21.9, 30.5, and 38.3%, respectively. DHA yields of the wild-type strainand of the corresponding overexpression strains were 4.3, 5.3, 6.1, and 6.4 g/L, i.e.,DHA yields of the overexpression strains were increased by 23.3, 41.9, and 48.8%,respectively.Conclusions: Acetyl-CoA and malonyl-CoA are precursors for fatty acid synthesis.ACL catalyzes the conversion of citrate in the cytoplasm into acetyl-CoA, and ACCcatalyzes the synthesis of malonyl-CoA from acetyl-CoA. The results demonstrate thatoverexpression of ACL and ACC enhances lipid accumulation and DHA production inSchizochytrium sp.Keywords: Schizochytrium sp., Docosahexaenoic acid, ATP-citrate lyase, Acetyl-CoAcarboxylase, β-galactosidase reporter system, Constitutive promoter

Corresponding Author: Zhi Chen, Ph.D.China Agricultural UniversityCHINA

Corresponding Author E-Mail: [email protected]

Corresponding Author SecondaryInformation:

Corresponding Author's Institution: China Agricultural University

Corresponding Author's SecondaryInstitution:

First Author: Xiao Han

First Author Secondary Information:

Order of Authors: Xiao Han

Powered by Editorial Manager® and ProduXion Manager® from Aries Systems Corporation

Zhunan Zhao

Ying Wen

Zhi Chen, Ph.D.

Order of Authors Secondary Information:

Response to Reviewers: Response to ReviewersBiotechnology for BiofuelsBBIO-D-19-00571Enhancement of docosahexaenoic acid production by overexpression of ATP-citratelyase and acetyl-CoA carboxylase in Schizochytrium sp.

Reviewer #1: It's a good work with significant improvement of DHA production as wellas a tool box for further application in biofuels production in this field. There are somecomments hope authors to take into consideration.1.The work focused on the availability of carbon supply to FA synthesis, however, thedesaturation is also very important to DHA production. The whole profile dynamicinformation of FA should be provided to have a overall view on FA synthesis processother than Fig 5.Response: Thank you very much for your valuable comments and suggestions. Weagree with you that the investigation of the whole dynamic profile will provide moreinformation. Though we can determine DHA yields through gas chromatography, ourlab can not analyze the fatty acid composition due to the lack of some fatty acidstandards. Fatty acid composition analysis in the study was carried out by the detectionplatform of our university which is closed now and will not open in months due to theoutbreak of new coronavirus. So we can not do the experiment. Guo et al. (2016) havereported that the changes of fatty acid composition in Schizochytrium sp. HX-308cultured at different aeration rates. The contents of DHA in TFAs from 40-h to 120-hwere increased gradually from 38.01% to 43.59% and from 40.52% to 44.85 ataeration rates of 0.5 vvm and 1.0 vvm, respectively, and decreased gradually from39.91% to 36.61% at aeration rate of 1.5 vvm. According to the report, DHA content ofSchizochytrium sp. was slightly changed during fermentation and was affected by theoxygen supply.Reference:Guo DS, Ji XJ, Ren LJ, Li GL, Yin FW, Huang H. Development of a real-timebioprocess monitoring method for docosahexaenoic acid production by Schizochytriumsp. Bioresour Technol. 2016;216:422-427.

2.In the industrial process, the glycerol is widely used as carbon other than glucose, sothe performance with glycerol is expected.Response: Thank you very much for your suggestion. As a byproduct in the productionof biodiesel, glycerol is a cost effective carbon source for lipid production. Glycerol wasused as the carbon source for the shake flask fermentation as suggested. Although itwas reported that some Schizochytrium spp. could use glycerol efficiently, we foundhere that Schizochytrium sp. grew poorly in the glycerol medium and the production oflipid and DHA were very low (Fig. I). We were quite surprised by the results, and theshake flask fermentation experiments were repeated with new batch of glycerol, andlower amount of glycerol (60 g/L) was also used to rule out the possibility of theinhibition of high glycerol on growth. Schizochytrium sp. still grew poorly in the glycerolfermentation medium (Fig. IIa, b). Glucose fermentation medium was also used as acontrol in which Schizochytrium sp. grew well and accumulated lipid efficiently (Fig. IIc,d). Even though Schizochytrium sp. grew poorly in glycerol medium, the lipid contentsand DHA yields of ACL and ACC overexpression strains were slightly higher thanthose of the WT strain (Fig. I). Therefore, overexpression of ACL and ACC did increasethe production of lipid and DHA in Schizochytrium sp. ATCC 20888. Since cells grewpoorly in glycerol medium, we would rather not include the results of the glycerolmedium in the manuscript. We don’t know why this strain grow poorly with glycerol, butit seems Schizochytrium spp. differ very much from each other, for example, thereported genome sizes of Schizochytrium limacinum SR21 and Schizochytrium sp.CCTCC M209059 were 63 Mb and 39 Mb, respectively.

Fig. I Effect of ACL and ACC overexpression on dry cell weight (DCW), lipidaccumulation, and DHA production by Schizochytrium sp. in the glycerol medium. Cells


were grown in fermentation medium with 100 g/L glycerol as the carbon source for 120h. a Dry cell weight (DCW, g/L). b Total lipids (% DCW). c DHA yield (g/L)

Fig. II Effect of ACL and ACC overexpression on dry cell weight (DCW) and lipidaccumulation by Schizochytrium sp. Cells were grown in fermentation medium for 120h with 60 g/L glycerol (a), 100 g/L glycerol (b), 100 g/L glucose (c, d) as the carbonsource. a, b, c Dry cell weight (DCW, g/L). d Total lipids (% DCW)

3.The expression of digital should be improved, and comma is expected to be used.Such as 1,269-bp fragment and 6,000 rpm, etc.Response: Thank you for the suggestion. We have corrected the expression of digitalin the text.

Reviewer #2I have now reviewed the manuscript, which was well-written in many aspects includingabstract, backrounds, m&m and conslusion. However, the discussion section must bere-written since it was written as a the repeatition of the results section. References,Figures and Tables are enough. Some methods are also missing. Therefore, I suggestmajor revision. After revision, it should be reconsidered. My spesific comments can befound on the text attached.Response: Thank you very much for your valuable comments and suggestions. Assuggested, we have carefully revised the manuscript, added some methods andreferences, and reorganized the discussion section. Most active tense sentences wererevised to passive tense except some (Line 120 and Line 175) in which the change willlead to inconsistent subject. Some specific concerns are as follows.

1.Line 25: Is the host or wild type? Please specify it. Line 28: What are they, pleasespecify them.Response: The specification of the strains was listed in Table 2.

2.Line 31: What is the difference between the Lipid yields and DHA yields?Response: Thank you for the question. Lipid included triacylglycerol, free fatty acids,and phospholipids, etc, which was extracted by acid-heating extraction. The fatty acidcomponents of lipids were DHA, palmitic acid (C16:0), myristic acid (C14:0), and DPA,etc (Fig. 6). For DHA yields determination, fatty acid methyl esters were firstly preparedfrom lipid samples and DHA was measured by gas chromatography.

3.Line 133, “β-galactosidase with very low activity”: please give some critical activity inthe text.Response: Only qualitative colorimetric detection was carried out on GPY plates. Theβ-galactosidase activities of the transformants in the liquid medium (seed medium)were shown in Fig. 1d.

4.Line 176, “ACL and ACC”: Italic or non-italic. Please be coherent.Response: The italic form indicates the encoding gene. And non-italic form indicatesprotein or enzyme. It was changed to be “ACL and ACC genes”.

5.Line 191: “Thus, overexpression of ACL and/or ACC did not affect cell growth ofSchizochytrium”. Any reason.Response: The results indicated that overexpression of ACL and ACC did notsignificantly affect the consumption of carbon and nitrogen sources of thetransformants, that is probably the reason why cell growth (dry cell weights) was notaffected.

6.Line 206: “Overexpression of ACL in Schizochytrium sp. WT did not significantlyaffect the percentage of TFAs represented by DHAs……”, Why? Any reason?Line 596 (Table 1): Although DHA yields, Lipid yields, and Lipid contents of therecombinant strains were higher than that of WT, DHA contents were similar to eachother. Any reason.Response: Overexpression of ACL and ACC enhanced the supply of precursors(acetyl-CoA and malonyl-CoA) for the synthesis of DHA and other fatty acids.Therefore, the increased precursors supply in the recombinant strains enhanced theproduction of DHA and other fatty acids at the same time. That is the reason of thesimilar DHA contents between WT and the recombinant strains. The related discussion


can be found in Line 251-260.

7.The discussion section was similar to the results section or was written as arepeatition of the results. Please compare your results with the existing results in theliterature. What were the similarities and differencies of your results with the data in theliterature? Please discuss them. Also, what is the industrial significance of this study?Please also elucidate this situation. To sum up, please re-write the discussion section.Response: Thank you very much for your suggestion. In the discussion section, wesummarized the main findings in the study which is the foundation for the follow-updiscussion. These sentences have been rephrased to avoid repetition. And thediscussion section has been carefully revised as suggested.

8.Line 270 (Microorganisms and culture conditions section): There are four differentmedium in this section, please use references for each medium and explain why themedium was different?Response: The references have been added to the text. LB medium is used forcultivation of E. coli in which plasmids were constructed. The media and growthconditions for Schizochytrium sp. were according to Ling et al. [5] with modifications. Infact, we did quite a lot of work to optimize the media (especially the fermentationmedium). GPY solid medium which contained less glucose were used for rapidcultivation of Schizochytrium sp. and selection of single colony of the transformants(supplemented with 40 μg/mL zeocin). Fermentation medium which contained highconcentration of glucose was used for lipid and DHA production (high C/N helps lipidaccumulation). And seed medium which had moderate concentration of glucoseguaranteed rapid growth of Schizochytrium sp. and ample cells for fermentationmedium. Cells would grow slowly in fermentation medium if they were inoculateddirectly from GPY solid medium.

9.Line 289: Why the promoters and terminators were amplified from different sources?Please elucidate it.Response: We intended to test the promoter strengths of some commonly-usedeukaryotic promoters (ubiquitinp, EF-1αp, ccg1p, TEF-1p, and AOX1p) and screen forstrong promoters in Schizochytrium (Line 145-162). Therefore, some promoters andterminators were amplified directly from fungi and yeast expression plasmids, andsome were amplified from Schizochytrium sp. itself (endogenous promoters can berecognized by RNA polymerase more efficiently). The results indicated that ccg1p,TEF-1p, and AOX1p from the commonly-used expression plasmids displayed strongerpromoter activities, which was not unexpected because expression plasmids usuallyutilize very strong promoters to ensure the efficient expression of the target genes.Identification of promoters with different promoter strengths can provide suitablepromoters for a balanced expression of the relevant pathways in Schizochytrium.

10.The fermentations in Fig. 1d and Fig. 3a-d can be modeled kinetically using logistic,Luedeking piret and modified luedeking piret models, but not this study. It can beinteresting.Response: Thank you so much for your kind suggestion. We will try to employ thesemodels for growth and product modelling in the future study. It will be very interesting.

Reviewer #3The manuscript by Han et al. tested several constitutive promoters fromSchizochytrium sp. and used the strong ccg1p promoter to overproduce ACL and ACC,which increased the yields of DHA and total lipids.Comments:1.How many independent strains of each transformant were generated and used forthe study? It seems like there is only one strain for each construct, which is notacceptable.Response: Thank you very much for your valuable comments and the question. Aftertransformation, we have selected eight independent strains for each kind oftransformants. The transformants were cultured in fermentation broth for 5 days,stained with nile red dye for neutral lipid staining, and the fluorescence intensities weremeasured by a multifunctional plate reader. All transformants displayed higherfluorescence intensities than the wild-type strain, indicating that the transformantsaccumulated more lipids than the WT strain (Fig. IIIa, b, c). After preliminary screening,two independent strains with the highest fluorescence for each kind of transformants


were cultured for determination of lipid and DHA yields. The same kind oftransformants displayed similar results (Fig. IIId). All fermentation experiments were atleast repeated twice. For clarity, results from one strain of each kind of transformantswere provided in the text.

Fig.III Effect of ACL and ACC overexpression on lipid accumulation by Schizochytriumsp.. a, b, c Nile red staining of WT, OACL, OACC, and OACL-ACC. d Lipid contents ofWT, OACL, OACC, and OACL-ACC. The strains were cultured in fermentation mediumfor 120 h

2.RNA or Protein levels of ACL and ACC should be examined in the overexpressingstrains.Response: Thank you very much for your suggestion. The transcriptional levels of ACLand ACC were determined by qRT-PCR in WT, OACL, OACC, and OACL-ACC whichwere cultivated in fermentation media for 2 and 4 d. The transcriptional levels of ACLand ACC were greatly increased in the corresponding transformants. The results wereadded to the text (Line 188-194) and the new Fig. 3.

3.Lipid content was increased from 54% of the WT to ~70% of the overexpressingstrains. Figure 4d, why the nile red staining shows such low level of signals in the WT?Did the expression of ACL and ACC changed the morphology of lipid bodies? i.e. sizeand numbers.Response: Thank you for the question. The nile red staining is a very efficient means toindicate the intracellular lipid accumulation. During our nile red staining preliminaryscreening, we found that the fluorescence signals had a positive correlation with thelipid contents, but the correlation was not strictly linear. The increase rates offluorescence signals were usually higher than the increase rates of lipid contents(please see Fig. III, response to comment 1). Besides, fluorescence attenuates duringmicroscopic observations. WT was the first one observed and its analysis might takelonger time than the others’. Microscopic analysis was repeated with 48-h culturedcells (it will be easier to observe the morphology of lipid bodies with “younger” cells)and all photos were taken with the same observation time. Fig. 4d was replaced withthe new figure (now as Fig. 5d), and the relative fluorescence intensities of cells wereanalyzed by ImageJ (Fig. IV), which had a good correlation with the lipid contents. Wethank you very much for your kind reminder. Since Schizochytrium cells aggregatedand produced zoospores during cultivation, it was difficult to distinguish lipid bodies.From what we saw, Schizochytrium formed a large lipid body per cell and no significantdifference of morphology wad noticed between WT and the transformants (Fig. 5d).

Fig.IV The relative fluorescence intensity of WT, OACL, OACC, and OACL-ACC

4.Explain or discuss the changes of fatty acid composition especially on the increase ofDHA in ACC and ACL-ACC.Response: Thank you for your suggestion. Please see answer to comment 6 ofReviewer 2. The discussion about the changes of fatty acid composition on theincrease of DHA in OACC and OACL-ACC strains can be found in Line 251-260.

Reviewer #4Some grammatical problems are seen in the whole manuscript which should berevised. Also, the organization of the manuscript especially in the introduction part isnot good. One more attention is needed in the discussion part. Some related paperswith for lipid extraction (with modifications) are introduced here for improvement of thelipid extraction part, determination of biomass dry weight and some other part in thematerial and method, cite these references in this part. Paper titles are in theparentheses. Add a table in discussion part and compare related studies.Best wishes1-Process Safety and Environmental Protection III (2017) 757-765http://dx.doi.org/10.1016/j.psep.2017.08.029 (Optimizing of lipid production inCryptococcus heimaeyensis through M32 array of Taguchi design)2-Process Safety and Environmental Protection III (2017) 747-756http://dx.doi.org/10.1016/j.psep.2017.08.027 (Single cell oil and its application forbiodieselproduction)


3-Int. J. Environ. Sci. Technol. DOI 10.1007/s13762-014-0687-8 (Recycling oflignocellulosic waste materials to produce high-value products: single cell oil andxylitol)4-Int. J. Environ. Sci. Technol. (2014) 11:597-604 DOI 10.1007/s13762-013-0373-2(Improving microbial oil production with standard and native oleaginous yeasts byusing Taguchi design)5-(Medium optimization for biotechnological production of single cell oil using Yarrowialipolytica M7 and Candida sp.)6-(Selection and optimization of single cell oil production from Rodotorula 110 usingenvironmental waste as substrate)Response: Thank you very much for your kind suggestions and the recommendedrelated references. We have carefully revised the manuscript and reorganized thediscussion section. We have earnestly read the recommended references with greatinterest. Three related references have been cited in the methods (determination ofglucose and lipid extraction). We will consider the methods in the related references inthe future study.

Additional Information:

Question Response


1

Enhancement of docosahexaenoic acid production by 1

overexpression of ATP-citrate lyase and acetyl-CoA 2

carboxylase in Schizochytrium sp. 3

4

Xiao Han, Zhunan Zhao, Ying Wen, Zhi Chen* 5

6

State Key Laboratory of Agrobiotechnology and Key Laboratory of Soil Microbiology, 7

Ministry of Agriculture, College of Biological Sciences, China Agricultural University, 8

Beijing 100193, China. 9

10

*Correspondence: [email protected] 11

12

Marked Up Manuscript Click here to access/download;Manuscript;Marked UpManuscript.docx

Click here to view linked References

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

https://www.editorialmanager.com/bbio/download.aspx?id=158450&guid=294739dc-5dd9-42cd-a3fc-a8540e3110ea&scheme=1

https://www.editorialmanager.com/bbio/download.aspx?id=158450&guid=294739dc-5dd9-42cd-a3fc-a8540e3110ea&scheme=1

https://www.editorialmanager.com/bbio/viewRCResults.aspx?pdf=1&docID=5132&rev=1&fileID=158450&msid=2bc6910d-e3d8-48ba-928e-d87084fee8ca

2

Abstract 13

Background: Docosahexaenoic acid (DHA) is an important omega-3 long-chain 14

polyunsaturated fatty acid that has a variety of physiological functions for infant 15

development and human health. Although metabolic engineering was previously 16

demonstrated to be a highly efficient way to rapidly increase lipid production, 17

metabolic engineering has seldom been previously used to increase DHA 18

accumulation in Schizochytrium spp. 19

Results: Here, a sensitive β-galactosidase reporter system was established to screen 20

for strong promoters in Schizochytrium sp. Four constitutive promoters (EF-1αp, 21

TEF-1p, ccg1p, and ubiquitinp) and one methanol-induced AOX1 promoter were 22

characterized by the reporter system with the promoter activity ccg1p > TEF-1p > 23

AOX1p (induced) > EF-1αp > ubiquitinp. With the strong constitutive promoter ccg1p, 24

Schizochytrium ATP-citrate lyase (ACL) and acetyl-CoA carboxylase (ACC) were 25

overexpressed in Schizochytrium sp. ATCC 20888. The cells were cultivated at 28°C 26

and 250 rpm for 120 h with glucose as the carbon source. Shake flask fermentation 27

results showed that the overexpression strains exhibited growth curves and biomass 28

similar to those of the wild-type strain. The lipid contents of the wild-type strain and 29

of the OACL, OACC, and OACL-ACC strains were 53.8, 68.8, 69.8, and 73.0%, 30

respectively, and the lipid yields of the overexpression strains were increased by 21.9, 31

30.5, and 38.3%, respectively. DHA yields of the wild-type strain and of the 32

corresponding overexpression strains were 4.3, 5.3, 6.1, and 6.4 g/L, i.e., DHA yields 33

of the overexpression strains were increased by 23.3, 41.9, and 48.8%, respectively. 34

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

3

Conclusions: Acetyl-CoA and malonyl-CoA are precursors for fatty acid synthesis. 35

ACL catalyzes the conversion of citrate in the cytoplasm into acetyl-CoA, and ACC 36

catalyzes the synthesis of malonyl-CoA from acetyl-CoA. The results demonstrate 37

that overexpression of ACL and ACC enhances lipid accumulation and DHA 38

production in Schizochytrium sp. 39

Keywords: Schizochytrium sp., Docosahexaenoic acid, ATP-citrate lyase, Acetyl-CoA 40

carboxylase, β-galactosidase reporter system, Constitutive promoter 41

42

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

4

Background 43

Docosahexaenoic acid (DHA, C22:6-∆4,7,10,13,16,19) is an omega-3 long-chain 44

polyunsaturated fatty acid (LC-PUFA). As the principal omega-3 fatty acid in brain 45

gray matter, DHA has neurotrophic and neuroprotective properties that are required 46

for normal perinatal cortical maturation [1]. In addition, DHA supplementation 47

improves human health by increasing cardioprotective, anti-inflammatory, and 48

anticancer activities [2, 3]. DHA is therefore widely used as a nutritional supplement, 49

often as a nutraceutical. 50

The conventional source of DHA is fish oil obtained from cold-water marine fish. 51

Seasonal variation, overharvest, and population decline, however, prevent the steady 52

supply of DHA that is required to meet the increasing market demands. Other 53

commercial sources of DHA production are thraustochytrids, which are marine 54

microorganisms [4]. Schizochytrium spp., as well as other thraustochytrids (such as 55

species of Thraustochytrium and Ulkenia), are excellent DHA producers [5, 6]. 56

Schizochytrium spp. can produce total fatty acids (TFAs) that represent up to 70% of 57

the cell weight, with DHA representing 25–45% of TFAs [7, 8]. Owing to the 58

increasing demand for DHA, many researchers have attempted to increase DHA 59

production by Schizochytrium spp. [5, 6, 9]. To date, most studies of DHA production 60

by Schizochytrium spp. have focused on the adaptive evolution of the strains [9]; on 61

the optimization of medium composition including sources of carbon and nitrogen 62

and the addition of inorganic salts and antioxidants [5, 6, 10, 11]; and on cultivation 63

conditions and cultivation styles [12, 13]. Only a few studies have employed 64

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

5

metabolic engineering to increase DHA accumulation in Schizochytrium. Yan et al. 65

[14], for example, introduced the Escherichia coli acetyl-CoA synthase gene into 66

Schizochytrium sp. TIO1101, which increased the biomass and TFA production of the 67

resulting transformant by 29.9% and 11.3%, respectively. Introduction of an 68

exogenous ω-3 desaturase gene into Schizochytrium sp. converted 3% 69

docosapentaenoic acid (DPA) into DHA [15]. By increasing the number of active ACP 70

domains of PUFA synthase, DHA productivity was increased by 1.8-fold in a 71

recombinant E. coli expressing Schizochytrium PUFA biosynthetic genes [16]. These 72

studies demonstrate that metabolic engineering can increase DHA production by 73

Schizochytrium spp. 74

Metabolic engineering has also been used with the oleaginous yeast Yarrowia 75

lipolytica, i.e., metabolic engineering efficiently increased the yeast’s production of 76

total lipids and ω-3 PUFAs [17-20]. By rewiring the metabolic pathways of Y. 77

lipolytica, researchers increased lipid accumulation > 60-fold, and caused lipid 78

content to approach 90% of cell mass [18]. Compared to the 10–15% lipid content in 79

wild-type (WT) Y. lipolytica [21], Schizochytrium spp. produces much higher lipid 80

levels, and ω-3 PUFA DHA represents up to 45% of TFAs. Because the genome 81

sequences of several strains of thraustochytrids (Schizochytrium, Thraustochytrium, 82

and Aurantiochytrium) are now available [22-24], metabolic engineering should be an 83

efficient way to rapidly increase their production of DHA and lipids. 84

Acetyl-CoA is precursor for fatty acid synthesis. ATP-citrate lyase (ACL) 85

catalyzes the conversion of citrate and CoA into acetyl-CoA and oxaloacetate, along 86

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

6

with the hydrolysis of ATP [25]. ACL is present in all eukaryotes except 87

non-oleaginous yeasts. In animals and oleaginous basidiomycete yeasts, ACL is 88

encoded by a single gene [26, 27]; in plants and some filamentous fungi, ACL usually 89

consists of two subunits (ACL1 and ACL2) with homology to the N- and C-terminals 90

of the animal ATP-citrate lyase polypeptide [28]. Fatty acid content was increased in Y. 91

lipolytica by overexpression of ACL1 and ACL2 on a non-lipogenic medium in an 92

obese strain [29] or overexpression of ACL from Mus musculus [30]. Acetyl-CoA 93

carboxylase (ACC) catalyzes the synthesis of malonyl-CoA from acetyl-CoA, which 94

is the rate-limiting step in fatty acid synthesis [21]. There are two types of ACCs in 95

nature: in most bacteria and plant chloroplasts, ACC usually consists of multiple 96

subunits, including the biotin carboxylase (BC), the biotin carboxyl carrier protein 97

(BCCP), the α-carboxyltransferase (α-CT) and the β-carboxyltransferase (β-CT); but 98

in mammals, fungi, and the cytoplasm of most plants, ACC is a single multifunctional 99

polypeptide [31]. Overexpression of acetyl-CoA carboxylase in the presence of 100

thioesterase in E. coli led to a 6-fold increase in the rate of fatty acid synthesis [32]. 101

ACC overexpression increased lipid content in Y. lipolytica and free fatty acid 102

production in S. cerevisiae [21, 33]. 103

To date, very few studies have attempted to enhance DHA production in 104

Schizochytrium spp. through metabolic engineering. In this study, a sensitive 105

β-galactosidase reporter system was established in Schizochytrium sp. to screen for 106

strong promoters. Because a sufficient supply of acetyl-CoA and malonyl-CoA is a 107

prerequisite for efficient lipid accumulation, Schizochytrium ACL and ACC were 108

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

7

overexpressed under the strong constitutive promoter ccg1p in Schizochytrium sp. 109

ATCC 20888 to enhance lipid accumulation and DHA production. 110

111

Results 112

Developing a β-galactosidase reporter system in Schizochytrium 113

Metabolic engineering involves the rewiring various metabolic pathways to redirect 114

metabolic flux towards the synthesis of target compounds. As a consequence, the 115

expression of relevant pathways must be strictly coordinated to achieve a balanced 116

expression and to avoid metabolic bottlenecks [34]. It is therefore crucial that a 117

reliable reporter system to monitor gene expression levels is established in 118

Schizochytrium. The E. coli β-galactosidase structural gene lacZ has been widely used 119

as a candidate reporter gene, providing convenient methods for qualitative 120

colorimetric detection on agars with 5-bromo-4-chloro-3-indolyl 121

β-D-galactopyranoside (X-gal) and quantitative β-galactosidase activity assays with 122

O-nitrophenyl-β-D-galactopyranoside (ONPG) [35]. To determine whether the 123

β-galactosidase reporter works in Schizochytrium, we constructed the reporter plasmid 124

pPICZαA-ubiquitinp-lacZ , in which the E. coli lacZ gene was driven by a ubiquitin 125

promoter-terminator system (Fig. 1a). pPICZαA containing lacZ without a ubiquitin 126

promoter (termed pPICZαA-AOX1p-lacZ) was also constructed as a control plasmid. 127

The corresponding transformants of Schizochytrium sp. ATCC 20888 were selected on 128

glucose-peptone-yeast extract (GPY) plates with zeocin. DNA fragments of 5.0 and 129

3.3 kb were amplified from the genomic DNAs of the ubip-lacZ and AOX1p-lacZ 130

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

8

transformants, respectively; the sizes of the fragments corresponded with the sizes of 131

the lacZ expression cassettes (Fig. 1b), indicating that both plasmids were integrated 132

into chromosomes. 133

The transformants of ubip-lacZ and AOX1p-lacZ grew normally on GPY agar. 134

When the substrate X-gal was added to GPY plates, Schizochytrium sp. WT produced 135

orange colonies with a very slight blue color, indicating that the WT possessed 136

endogenous β-galactosidase with very low activity. AOX1p-lacZ produced colonies 137

that were similar to those of the WT, while ubip-lacZ produced blue colonies (Fig. 1c). 138

β-galactosidase enzymatic activity was much higher in the lysate of the ubip-lacZ 139

transformant than in the AOX1p-lacZ transformant (Fig. 1d). These results show that, 140

although Schizochytrium WT possesses endogenous β-galactosidase activity, 141

β-galactosidase is able to serve as a sensitive reporter system in Schizochytrium. 142

143

Selection of strong promoters in Schizochytrium 144

The use of four commonly-used eukaryotic promoters (EF-1αp, ccg1p, TEF-1p, and 145

AOX1p) in addition to ubiquitinp was characterized in the β-galactosidase reporter 146

system. In pPICZαA-AOX1p-lacZ, a methanol-induced AOX1p [36] is present 147

upstream of the lacZ gene. On X-gal plates, AOX1p-lacZ produced orange colonies 148

like those of the WT, but the transformants with other promoters produced blue 149

colonies, and the transformants with ccg1p or TEF-1p produced the bluest colonies. 150

β-galactosidase activity in AOX1p-lacZ without methanol induction treatment was 151

similar to that in the WT (Fig.1c), indicating that AOX1p is inactive without induction 152

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

9

by methanol. The enzymatic activities driven by ccg1p, TEF-1p, EF-1αp, or ubiquitinp 153

were much higher than that of WT, with promoter activity of ccg1p > TEF-1p > 154

EF-1αp > ubiquitinp (Fig. 1d). These findings indicate that the four promoters are 155

strong, constitutively-expressed promoters. When methanol was added to a 12-h 156

culture of AOX1p-lacZ to a final concentration of 1% (vol/vol), the β-galactosidase 157

activity increased substantially and remained at a high level, with the induction 158

strength between TEF-1p and EF-1αp, indicating that AOX1p can be recognized by 159

Schizochytrium RNA polymerase and induced by methanol. Therefore, AOX1 160

promoter can serve as a methanol-induced promoter in Schizochytrium, although the 161

induction time and strength require optimization. 162

163

Construction of ACL and ACC-overexpression strains in Schizochytrium sp. 164

In cytoplasm, ATP-citrate lyase converts intracellular citrate to acetyl-CoA in an 165

ATP-dependent manner, and acetyl-CoA carboxylase catalyzes the synthesis of 166

malonyl-CoA from acetyl-CoA [25]. Acetyl-CoA and malonyl-CoA are the precursors 167

for fatty acids synthesis [21, 29]. A BLAST search of the Schizochytrium sp. CCTCC 168

M209059 genome [22] revealed one putative ACL-encoding gene and one putative 169

ACC-encoding gene (Additional file 2: Table S1). The putative ACL (422 aa) contains 170

an ATP citrate (pro-S)-lyase domain at the N-terminus, which is homologous to 171

ATP-citrate lyase subunit 1, and a citrate-binding domain at the C-terminus, which is 172

homologous to ATP-citrate lyase subunit 2 (Fig. 2a). Therefore, Schizochytrium 173

ATP-citrate lyase functions as a single-subunit ACL. Like ACC in most fungi, ACC 174

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

10

(2,352-aa) in Schizochytrium is a single multifunctional polypeptide, containing a 175

biotin carboxylase (BC) domain, a biotin carboxyl carrier protein (BCCP) domain, 176

and a carboxyl transferase (CT) domain (Fig. 2a). 177

To promote lipid biosynthesis and DHA accumulation in Schizochytrium sp. 178

ATCC 20888, we developed transformants that overexpressed Schizochytrium 179

ATP-citrate lyase and acetyl-CoA carboxylase. ACL and ACC genes were amplified 180

from the cDNA of Schizochytrium sp. ATCC 20888 and were cloned separately or 181

together into pPICZαA, in which the cloned gene was driven by the ccg1 promoter 182

and terminator (Additional file 1: Figure S1). After transformation, 2.8-, 8.7-, and 183

12.0-kb PCR fragments containing the corresponding expression cassettes 184

(ccg1p-ACL-ccg1t, ccg1p-ACC-ccg1t, and ccg1p-ACL-ccg1t-ccg1p-ACC-ccg1t) were 185

amplified from the genomic DNAs of pPICZαA-ACL, pPICZαA-ACC, and 186

pPICZαA-ACL-ACC transformants (termed OACL, OACC, and OACL-ACC) (Fig. 187

2b), indicating that the plasmids were integrated into the chromosomes. The 188

transcription levels of ACL and ACC were examined by qRT-PCR in WT, OACL, 189

OACC, and OACL-ACC cultivated in fermentation broth for 2 and 4 d. Compared to 190

WT, the expression of ACL were increased in OACL and OACL-ACC, and the 191

expression of ACC were increased in OACC and OACL-ACC at both time points, 192

indicating that transcription levels of ACL and ACC were increased in the 193

corresponding overexpression strains (Fig. 3). 194

195

ACL and ACC overexpression enhanced lipid accumulation 196

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

11

Shake-flask fermentation results showed that the WT strain and the overexpression 197

transformants did not significantly differ in their consumption of carbon and nitrogen 198

sources, pH values, or dry cell weights (DCW) (Fig. 4). DCW reached their maximum 199

values on day 5, which were 23.7, 22.7, 23.9, and 24.3 g/L for WT, OACL, OACC, 200

and OACL-ACC, and then decreased slightly with further cultivation (Fig. 4d, 5a). 201

Thus, overexpression of ACL and/or ACC did not affect cell growth of 202

Schizochytrium. 203

After 5 days of cultivation, lipid yields were lowest for the WT (12.8 g/L), 204

highest for OACL-ACC (17.7 g/L), and intermediate for OACL (15.6 g/L), OACC 205

(16.7 g/L) (Table 1). A similar pattern was evident for lipid content, i.e., lipid content 206

was substantially higher in the overexpression strains than in the WT (Fig. 5b). 207

Compared to the WT strain, the lipid yields were increased by 21.9% in OACL, by 208

30.5% in OACC, and by 38.3% in OACL-ACC. Microscopic observation revealed an 209

increased intensity of lipid droplet staining in the overexpression strains (Fig. 5d). The 210

findings indicated that overexpression of ACL and ACC increased lipid production in 211

Schizochytrium sp., probably by increasing the supply of acetyl-CoA and 212

malonyl-CoA. 213

214

ACL and ACC overexpression promoted DHA production 215

Gas chromatography analysis showed that the main fatty acid components of 216

Schizochytrium sp. ATCC 20888 were DHA, palmitic acid (C16:0), myristic acid 217

(C14:0), and docosapentaenoic acid (DPA, C22:5) (Fig. 6). Overexpression of ACL in 218

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

12

Schizochytrium sp. WT did not significantly affect the percentage of TFAs represented 219

by DHAs (36.4% for the WT, and 36.2% for OACL) (Fig. 6), but slightly decreased 220

the percentage represented by palmitic acid and slightly increased the percentage 221

represented by C16:1. In ACC-overexpression strains, the percentage of TFAs 222

represented by DHAs increased slightly for OACC (37.6%) and OACL-ACC (37.9%), 223

and the percentage represented by oleic acid and DPA decreased slightly (Fig. 6). 224

Compared to the DHA yield of the WT (4.3 g/L), the DHA yields of OACL, OACC, 225

and OACL-ACC strains were 5.3, 6.1, and 6.4 g/L, increased by 23.3, 41.9, and 226

48.8%, respectively (Fig. 5c). These results indicated that overexpression of ACL and 227

ACC greatly increased DHA production in Schizochytrium sp. ATCC 20888. 228

229

Discussion 230

In this study, a sensitive β-galactosidase reporter system in Schizochytrium was 231

developed and used to compare the strengths of some commonly-used eukaryotic 232

promoters. Although endogenous β-galactosidase is present in Schizochytrium, the 233

reporter system was able to detect different levels of β-galactosidase activity with the 234

LacZ-reporter driven by different promoters. ccg1p, TEF-1p, EF-1αp, and ubiquitinp 235

promoters are constitutive promoters and the AOX1p promoter is inducible by 236

methanol in Schizochytrium. More work can be carried out for subsequent research, 237

such as screening for more endogenous constitutive promoters with different 238

expression intensities and optimization of the induction conditions for the 239

methanol-induced AOX1p promoter. The β-galactosidase reporter system will facilitate 240

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

13

characterization of novel genetic elements and will help identify promoters for 241

fine-tuning gene expression in Schizochytrium. 242

Acetyl-CoA and malonyl-CoA are precursors for fatty acid synthesis [37]. 243

Previous studies have shown that increasing the substrates supply significantly 244

enhanced the synthesis of fatty acids and lipids in bacterium, yeast and fungi [30, 32, 245

38]. In the study, lipid and DHA production were greatly increased by overexpression 246

of ACL and ACC. ACL converts intracellular citrate to acetyl-CoA [29], and ACC 247

catalyzes the synthesis of malonyl-CoA from acetyl-CoA [21]. It follows that 248

overexpression of ACL and ACC evidently promoted production of intracellular 249

acetyl-CoA and malonyl-CoA, resulting in enhanced biosynthesis of fatty acids, 250

which in turn promoted lipid accumulation. The percentage of TFAs represented by 251

DHAs in ACL-overexpression strain was similar to that in the WT, indicating that 252

overexpression of the ACL gene led to increased intracellular acetyl-CoA pools, which 253

promoted the production of both saturated fatty acids and polyunsaturated fatty acids. 254

In ACC-overexpression strains, the percentage of TFAs represented by DHAs 255

improved slightly, while the percentage of TFAs represented by oleic acid and DPA 256

decreased slightly. Thus, the increased malonyl-CoA pool in ACC-overexpression 257

strains increased DHA production more than saturated fatty acid production. The 258

findings suggested that the supply of malonyl-CoA is the limiting factor of DHA 259

overproduction in Schizochytrium. In oleaginous microorganisms, efficient fatty acid 260

synthesis requires not only an abundant supply of acetyl-CoA and malonyl-CoA, but 261

also an ample NADPH supply [38, 39]. Therefore, combining this strategy with other 262

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

14

strategies rewiring the metabolic flux of Schizochytrium towards precursors and 263

NADPH accumulation might significantly improve its lipogenesis capability and 264

DHA productivity. 265

Schizochytrium spp. are excellent producers of ω-3 PUFA: they can synthesize 266

DHA de novo and DHA represents up to 45% of TFAs [7, 8]. To date, most studies of 267

Schizochytrium spp. have focused on the optimization of fermentation media and 268

cultivation conditions and the adaptive evolution of the strains [5, 6, 9-12]. Very few 269

studies have employed metabolic engineering to improve DHA production of 270

Schizochytrium spp. [14-16], mainly because the genetic background of 271

Schizochytrium remains poorly understood. Metabolic engineering has been proven to 272

be a highly efficient way to increase total lipids and ω-3 PUFA accumulation in the 273

yeast Y. lipolytica [17-20]. In the current investigation, we demonstrated that 274

metabolic engineering is an efficient way for increasing lipid accumulation and DHA 275

production in Schizochytrium. The study provided a sensitive reporter system to 276

monitor gene expression levels in Schizochytrium and a genetically engineered 277

Schizochytrium sp. for industrial production of DHA. 278

279

Conclusions 280

A strain of Schizochytrium sp. ATCC 20888 that overexpressed ACL and ACC under 281

the strong constitutive promoter ccg1p was constructed and thereby produced high 282

quantities of DHA. Under shake flask culture conditions, OACL, OACC, and 283

OACL-ACC strains attained a dry cell weight of 22.7, 23.9, and 24.3 g/L, respectively. 284

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

15

Compared to the WT, total lipid content of OACL, OACC, and OACL-ACC strains 285

reached 68.8, 69.8, and 73.0%, respectively, and the lipid yields of the overexpression 286

strains were increased by 21.9, 30.5, and 38.3%, respectively. A final DHA yield of 287

6.4 g/L in OACL-ACC was achieved, which was 48.8% higher than that of the WT. 288

Next, fermentation control of OACL-ACC in fermentors will be optimized to make it 289

more suitable for industrial application. 290

291

Methods 292

Microorganisms and culture conditions 293

Strains and plasmids used in the study are listed in Table 2. Media and growth 294

conditions of Schizochytrium sp. were according to Ling et al. [5] with modifications. 295

Schizochytrium sp. was cultured at 28°C on solid GPY medium containing per liter 20 296

g of glucose, 10 g of peptone, 5 g of yeast extract, 20 g of sea crystal, and 20 g of agar. 297

Transformants were selected and cultured on GPY supplemented with 40 μg/mL 298

zeocin. For lipid and DHA production, 250-mL flasks containing 50 mL of seed 299

medium (containing per liter 30 g of glucose, 10 g of peptone, 5 g of yeast extract, 300

and 20 g of sea crystal) were inoculated with Schizochytrium sp. cells and incubated 301

for 24 h at 28°C on a rotary shaker (230 rpm). The seed culture was inoculated at 5% 302

(vol/vol) into 50 mL of fermentation medium (containing per liter 100 g of glucose, 5 303

g of yeast extract, 3.94 g of NaCl, 0.264 g of KCl, 0.5 g of (NH4)2SO4, 1 g of KH2PO4, 304

1.43 g of MgSO4, 0.04 g of CaCl2, 10 g of sodium glutamate, 0.001 g of vitamin B1, 305

and 0.001 g of vitamin B12) and was then incubated at 28°C on a rotary shaker (250 306

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

16

rpm) for 168 h. For qualitative colorimetric and quantitative detection of 307

β-galactosidase, cells were cultured on GPY plates with 40 μg/mL X-gal and in seed 308

medium, respectively. E. coli was grown in LB medium at 37°C [35]. 309

310

Construction of β-galactosidase reporter plasmids and of ACL-, and 311

ACC-overexpression plasmids 312

To construct β-galactosidase reporter plasmids, a 3,075-bp fragment containing the 313

coding sequence of the lacZ gene was amplified from pMC1403 [35] by PCR using 314

primer pair lac-Fw and lac-Rev (Additional file 2: Table S2). The promoters and 315

terminators of EF-1α [40] and ubiquitin [15] were amplified from Schizochytrium sp. 316

ATCC 20888; ccg1 promoter and terminator were amplified from the Neurospora 317

expression vector pCCG.N-3xMyc [41]; and the TEF-1 promoter and CYC-1 318

terminator were amplified from the yeast expression vector pPICZαA [42] using the 319

primer pairs listed in Additional file 2: Table S2. After purification, the lacZ gene was 320

digested with KpnI/NotI, the promoters were digested with EcoRI/KpnI, and the 321

terminators were digested with NotI/XbaI; the digested gene, promoters, and 322

terminators were then simultaneously ligated into EcoRI/XbaI-digested pPICZαA to 323

generate reporter plasmids. Ligation reactions were performed overnight at 16°C 324

using T4 DNA Ligase (TaKaRa, Japan). 325

To construct the ACL overexpression plasmid, a 1,269-bp fragment of the ACL 326

gene was amplified from Schizochytrium sp. ATCC 20888 cDNA. After purification, 327

the KpnI/NotI-digested ACL gene, EcoRI/KpnI-digested ccg1p, and 328

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

17

NotI/XbaI-digested ccg1t were simultaneously ligated into EcoRI/XbaI-digested 329

pPICZαA to generate the overexpression plasmid pPICZαA-ACL. For construction of 330

the ACC overexpression plasmid, a 7,059-bp ACC fragment was amplified from the 331

same cDNA and purified. The ACC gene and the ccg1 promoter and terminator were 332

inserted into EcoRI-digested pPICZαA to generate pPICZαA-ACC using the 333

Seamless assembly cloning kit (Clone Smarter, USA) following the manufacture’s 334

protocol. To construct the ACL and ACC co-overexpression plasmid, the 2,611-bp 335

ccg1p-ACL-ccg1t expression cassette was amplified from plasmid pPICZαA-ACL and 336

was inserted into EcoRI-digested pPICZαA-ACC to generate pPICZαA-ACL-ACC by 337

Seamless assembly cloning. 338

339

Transformation of Schizochytrium sp. 340

Transformation of Schizochytrium was performed as described previously with 341

modification [15]. Schizochytrium sp. cells were cultured in seed medium for 24 h to 342

the logarithmic growth phase, and were harvested by centrifugation (5,900 g, 4°C, 10 343

min) (HITACHI CF16RXⅡ, Japan), washed with ice-cold sterile water, washed with 344

1 M sorbitol, and then suspended in 1 M sorbitol. The plasmids were linearized with 345

restriction enzyme BamHI before transformation. The competent cells and 5 μg of 346

linearized plasmid DNA were placed in a 0.1-cm-gap cuvette. The parameters of 347

electroporation were 0.75 kV, 200 Ω, 50 μF, twice. After electroporation, 1 mL of seed 348

medium was added to the mixture, which was incubated at 28°C for 4 h. The 349

transformants were spread on GPY plates with 40 μg/mL zeocin and grown at 28°C. 350

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

18

351

Genomic PCR analysis of transformants 352

Genomic DNAs of putative transformants were extracted according to Lippmeier et al. 353

[43]. To confirm Schizochytrium sp. transformants, the incorporation of the expression 354

cassette into the genome was verified by PCR using primers AOX-Fw and AOX-Rev 355

(Additional file 2: Table S2). PCR reactions were set up using Taq DNA polymerase 356

(TaKaRa, Japan) following the manufacture’s protocol. PCR amplification parameters 357

were as follows: 5 min at 95°C; followed by 30 cycles of 50 s at 95°C, 50 s at 55°C, 358

and 5 min at 72°C; and a final extension for 10 min at 72°C. 359

360

β-galactosidase activity assay 361

Schizochytrium sp. cells cultured in seed medium were collected by centrifugation at 362

the indicated time, washed with phosphate buffer saline solution (PBS, 38.7 mM 363

Na2HPO4•12H2O, 11.3 mM NaH2PO4•2H2O, and 150 mM NaCl), resuspended in 1 364

mL PBS, and disrupted with Mini Bead Beater (Biospec Mini-Bead-Beater-16 Model 365

607EUR, USA) for 50 s each for 5 times. After centrifugation, 500 μL of supernatant 366

was transferred to a new tube. A 500-μL volume of buffer Z (60.0 mM 367

Na2HPO4•12H2O, 39.7 mM NaH2PO4•2H2O, 10.0 mM KCl, 1.0 mM MgSO4•7H2O, 368

and 2.7 mL/L β-mercaptoethanol) and 200 μL of 13.3 mM ONPG were added to the 369

supernatant, and the mixture was incubated at 37°C for 15 min; the reaction was 370

stopped by adding 500 μL of 1 M Na2CO3. The OD420 was recorded with an 371

ultraviolet spectrophotometer for the determination of β-galactosidase activity. The 372

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

19

amount of enzyme that releases 1 µmol of ONP per minute is defined as one unit of 373

enzyme activity. 374

375

Determination of dry cell weight, pH and glucose and nitrogen concentrations 376

Determination of dry cell weight was performed as described previously with 377

modification [10]. A 40-mL volume of fermentation broth was centrifuged at 5,900 g 378

for 10 min, and dry cell weight was determined after freeze-drying for 24 to 48 h to a 379

constant weight. For measurement of pH, glucose and nitrogen concentrations, 1 mL 380

of broth was centrifuged (Heraeus BIOFUGE pico, Germany) at 13,523 g for 10 min, 381

and the supernatant was used for determination. The pH was measured by a laboratory 382

pH meter (METTLER TOLEDO FiveEasy, Switzerland). The concentration of 383

glucose was determined by the 3, 5-dinitrosalicylic acid (DNS) method [44, 45]. NH4+ 384

concentration was measured by the indophenol blue spectrophotometric method [46]. 385

386

Microscopic analysis 387

Nile red staining of cells was conducted as described previously with modification 388

[18]. A 1-mL volume of a culture grown in fermentation medium for 48 h was 389

collected by centrifugation, washed twice with PBS solution, and resuspended in 1 390

mL of PBS. Cells were stained with nile red dye (0.5 mg/L) and were incubated for 5 391

min in the dark. Fluorescence images were captured with a LEICA TCS SP8 392

microscope equipped with an oil immersion objective (×1,000 magnification). 393

394

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

20

Lipid extraction and fatty acid composition analysis 395

Lipids were extracted as described previously with some modification [47-49]. About 396

0.3 g of a freeze-dried Schizochytrium sp. pellet was mixed with 6 mL of 4 M HCl for 397

30 min and then incubated in boiling water for 8 min before 16 mL of 398

methanol/chloroform (1:1, vol/vol) was added. The preparation was mixed vigorously, 399

and then centrifuged at 129 g for 10 min. The lower phase was transferred to a 400

pre-weighed glass tube and evaporated under a stream of nitrogen. 401

Fatty acid methyl esters (FAMEs) were prepared according to Ren et al. [50] with 402

some modifications. About 30 mg of lipid sample was transferred to a glass tube 403

before 1 mL of internal standard (methyl nonadecanoate, C19:0, 1 mg/mL) and 1 mL 404

of 0.5 M KOH in methanol were added; the mixture was incubated at 65°C in a water 405

bath for 15 min. After the mixture had cooled to room temperature, 2.1 mL of 406

methanol and 0.9 mL of 45% BF3-ether were added to the tube, which was incubated 407

at 65°C for 5 min. Then 1 mL of hexane and 2 mL of saturated sodium chloride 408

solution were added; the preparation was mixed vigorously and allowed to stand for 409

10 min. The upper layer of the solution was transferred to a new tube and used for 410

analysis of fatty acid composition. FAMEs were separated by gas chromatography 411

(WUFENG GC522) with an Agilent J & W DB23 capillary column (30 m×0.25 mm 412

i.d.). Nitrogen was used as the carrier gas at a flow rate of 2 mL/min. The injector was 413

at 250°C. The column temperature was increased from 150°C to 200°C at the rate of 414

5°C per min, was kept at 200°C for 1 min, was then raised to 230°C at the rate of 4°C 415

per min, and was maintained at 230°C an additional 9 min. 416

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

21

417

RNA preparation and quantitative real-time PCR analysis (qRT-PCR) 418

Schizochytrium sp. cells cultured in fermentation medium were collected at 2 and 4 d, 419

frozen in liquid nitrogen, and ground to fine powder. Total RNA was extracted with 420

TRIzol reagent (Tiangen, China) according to the manufacturer’s protocol. cDNA was 421

synthesized by M-MLV (RNase H-; TaKaRa) with oligo-dT18 from 4 µg of total RNA. 422

qRT-PCR analysis was performed using FastStart Universal SYBR Green Master 423

(ROX) with primers listed in Table S2. PCR included a 10 min preincubation at 95°C, 424

followed by 40 cycles of denaturation at 95°C for 10 s, and annealing and extension at 425

60°C for 30 s. The relative expression levels were determined according to the 426

comparative Ct method, using actin as the internal control. 427

428

Statistical analysis 429

All experiments were performed with three biological replicates. Statistical analyses 430

were performed using one-way ANOVAs and Duncan’s multiple range tests or 431

two-tailed Student’s t-tests. And it was considered indicative of statistical significance 432

at p < 0.05. 433

434

Abbreviations 435

ACC: acetyl-CoA carboxylase; ACL: ATP-citrate lyase; DCW: dry cell weight; DHA: 436

docosahexaenoic acid; DNS: 3,5-dinitrosalicylic acid; DPA: docosapentaenoic acid; 437

FAMEs: Fatty acid methyl esters; LC-PUFA: long chain polyunsaturated fatty acid; 438

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

22

ONPG: O-nitrophenyl-β-D-galactopyranoside; PBS: phosphate buffer saline; TFAs: 439

total fatty acids; WT: wild type; X-gal: 5-bromo-4-chloro-3-indolyl 440

β-D-galactopyranoside. 441

442

Declarations 443

Authors’ contributions 444

ZC and XH designed the study. XH and ZZ performed the experiments. YW helped 445

with analysis and discussion of results. XH and ZC wrote the manuscript. All authors 446

read and approved the finalized manuscript. 447

Acknowledgements 448

The authors are grateful to Prof. B. Jaffee for English editing of the manuscript. 449

Competing interests 450

The authors declare that they have no competing interests. 451

Availability of supporting data 452

All data supporting the conclusions of this article are included in the manuscript and 453

in the additional information. 454

Consent for publication 455

All authors have read and approved the final manuscript. 456

Ethics approval and consent to participate 457

Not applicable. 458

Funding 459

This work was financially supported by the National Natural Science Foundation of 460

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

23

China (No. 31470190). 461

462

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

24

References 463

1. McNamara RK. DHA deficiency and prefrontal cortex neuropathology in 464

recurrent affective disorders. J Nutr. 2010;140:864-8. 465

2. Hashimoto M, Hossain S, Al Mamun A, Matsuzaki K, Arai H. 466

Docosahexaenoic acid: one molecule diverse functions. Crit Rev Biotechnol. 467

2017;37:579-97. 468

3. Zarate R, El Jaber-Vazdekis N, Tejera N, Perez JA, Rodriguez C. Significance 469

of long chain polyunsaturated fatty acids in human health. Clin Transl Med. 470

2017;6:25. 471

4. Fan KW, Chen F, Jones EB, Vrijmoed LL. Eicosapentaenoic and 472

docosahexaenoic acids production by and okara-utilizing potential of 473

thraustochytrids. J Ind Microbiol Biotechnol. 2001;27:199-202. 474

5. Ling X, Guo J, Liu X, Zhang X, Wang N, Lu Y, et al. Impact of carbon and 475

nitrogen feeding strategy on high production of biomass and docosahexaenoic 476

acid (DHA) by Schizochytrium sp. LU310. Bioresour Technol. 477

2015;184:139-47. 478

6. Ren LJ, Sun XM, Ji XJ, Chen SL, Guo DS, Huang H. Enhancement of 479

docosahexaenoic acid synthesis by manipulation of antioxidant capacity and 480

prevention of oxidative damage in Schizochytrium sp. Bioresour Technol. 481

2017;223:141-8. 482

7. Ashford A, Barclay WR, Weaver CA, Giddings TH, Zeller S. Electron 483

microscopy may reveal structure of docosahexaenoic acid-rich oil within 484

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

25

Schizochytrium sp. Lipids. 2000;35:1377-86. 485

8. Raghukumar S. Thraustochytrid marine protists: production of PUFAs and 486

other emerging technologies. Mar Biotechnol. 2008;10:631-40. 487

9. Sun XM, Ren LJ, Ji XJ, Chen SL, Guo DS, Huang H. Adaptive evolution of 488

Schizochytrium sp. by continuous high oxygen stimulations to enhance 489

docosahexaenoic acid synthesis. Bioresour Technol. 2016;211:374-81. 490

10. Ganuza E, Anderson AJ, Ratledge C. High-cell-density cultivation of 491

Schizochytrium sp. in an ammonium/pH-auxostat fed-batch system. 492

Biotechnol Lett. 2008;30:1559-64. 493

11. Hu XC, Ren LJ, Chen SL, Zhang L, Ji XJ, Huang H. The roles of different 494

salts and a novel osmotic pressure control strategy for improvement of DHA 495

production by Schizochytrium sp. Bioprocess Biosyst Eng. 2015;38:2129-36. 496

12. Zeng Y, Ji XJ, Lian M, Ren LJ, Jin LJ, Ouyang PK, et al. Development of a 497

temperature shift strategy for efficient docosahexaenoic acid production by a 498

marine fungoid protist, Schizochytrium sp. HX-308. Appl Biochem Biotechnol. 499

2011;164:249-55. 500

13. Chi ZY, Liu Y, Frear C, Chen SL. Study of a two-stage growth of 501

DHA-producing marine algae Schizochytrium limacinum SR21 with shifting 502

dissolved oxygen level. Appl Microbiol Biotechnol. 2009;81:1141-8. 503

14. Yan J, Cheng R, Lin X, You S, Li K, Rong H, et al. Overexpression of 504

acetyl-CoA synthetase increased the biomass and fatty acid proportion in 505

microalga Schizochytrium. Appl Microbiol Biotechnol. 2013;97:1933-9. 506

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

26

15. Ren LJ, Zhuang XY, Chen SL, Ji XJ, Huang H. Introduction of omega-3 507

desaturase obviously changed the fatty acid profile and sterol content of 508

Schizochytrium sp. J Agric Food Chem. 2015;63:9770-6. 509

16. Hayashi S, Satoh Y, Ujihara T, Takata Y, Dairi T. Enhanced production of 510

polyunsaturated fatty acids by enzyme engineering of tandem acyl carrier 511

proteins. Sci Rep. 2016;6:35441. 512

17. Xue Z, Sharpe PL, Hong SP, Yadav NS, Xie D, Short DR, et al. Production of 513

omega-3 eicosapentaenoic acid by metabolic engineering of Yarrowia 514

lipolytica. Nat Biotechnol. 2013;31:734-40. 515

18. Blazeck J, Hill A, Liu L, Knight R, Miller J, Pan A, et al. Harnessing Yarrowia 516

lipolytica lipogenesis to create a platform for lipid and biofuel production. Nat 517

Commun. 2014;5:3131. 518

19. Yuzbasheva EY, Mostova EB, Andreeva NI, Yuzbashev TV, Laptev IA, 519

Sobolevskaya TI, et al. Co-expression of glucose-6-phosphate dehydrogenase 520

and acyl-CoA binding protein enhances lipid accumulation in the yeast 521

Yarrowia lipolytica. N Biotechnol. 2017;39:18-21. 522

20. Xu P, Qiao K, Stephanopoulos G. Engineering oxidative stress defense 523

pathways to build a robust lipid production platform in Yarrowia lipolytica. 524

Biotechnol Bioeng. 2017;114:1521-30. 525

21. Tai M, Stephanopoulos G. Engineering the push and pull of lipid biosynthesis 526

in oleaginous yeast Yarrowia lipolytica for biofuel production. Metab Eng. 527

2013;15:1-9. 528

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

27

22. Ji XJ, Mo KQ, Ren LJ, Li GL, Huang JZ, Huang H. Genome sequence of 529

Schizochytrium sp. CCTCC M209059, an effective producer of 530

docosahexaenoic acid-rich lipids. Genome Announc. 2015;3:e00819-15.. 531

23. Zhao X, Dauenpen M, Qu C, Qiu X. Genomic analysis of genes involved in 532

the biosynthesis of very long chain polyunsaturated fatty acids in 533

Thraustochytrium sp. 26185. Lipids. 2016;51:1065-75. 534

24. Liu B, Ertesvag H, Aasen IM, Vadstein O, Brautaset T, Heggeset TM. Draft 535

genome sequence of the docosahexaenoic acid producing thraustochytrid 536

Aurantiochytrium sp. T66. Genom Data. 2016;8:115-6. 537

25. Chen H, He X, Geng H, Liu H. Physiological characterization of ATP-citrate 538

lyase in Aspergillus niger. J Ind Microbiol Biotechnol. 2014;41:721-31. 539

26. Zhao S, Torres A, Henry RA, Trefely S, Wallace M, Lee JV, et al. ATP-citrate 540

lyase controls a glucose-to-acetate metabolic switch. Cell Rep. 541

2016;17:1037-52. 542

27. Chavez-Cabrera C, Marsch R, Bartolo-Aguilar Y, Flores-Bustamante ZR, 543

Hidalgo-Lara ME, Martinez-Cardenas A, et al. Molecular cloning and 544

characterization of the ATP citrate lyase from carotenogenic yeast Phaffia 545

rhodozyma. FEMS Yeast Res. 2015;15. 546

28. Gawryluk RM, Eme L, Roger AJ. Gene fusion, fission, lateral transfer, and 547

loss: not-so-rare events in the evolution of eukaryotic ATP citrate lyase. Mol 548

Phylogenet Evol. 2015;91:12-6. 549

29. Dulermo T, Lazar Z, Dulermo R, Rakicka M, Haddouche R, Nicaud JM. 550

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

28

Analysis of ATP-citrate lyase and malic enzyme mutants of Yarrowia 551

lipolytica points out the importance of mannitol metabolism in fatty acid 552

synthesis. Biochim Biophys Acta. 2015;1851:1107-17. 553

30. Zhang H, Zhang L, Chen H, Chen YQ, Chen W, Song Y, et al. Enhanced lipid 554

accumulation in the yeast Yarrowia lipolytica by over-expression of 555

ATP:citrate lyase from Mus musculus. J Biotechnol. 2014;192 Pt A:78-84. 556

31. Gerhardt EC, Rodrigues TE, Muller-Santos M, Pedrosa FO, Souza EM, 557

Forchhammer K, et al. The bacterial signal transduction protein GlnB 558

regulates the committed step in fatty acid biosynthesis by acting as a 559

dissociable regulatory subunit of acetyl-CoA carboxylase. Mol Microbiol. 560

2015;95:1025-35. 561

32. Davis MS, Solbiati J, Cronan JE, Jr. Overproduction of acetyl-CoA 562

carboxylase activity increases the rate of fatty acid biosynthesis in Escherichia 563

coli. J Biol Chem. 2000;275:28593-8. 564

33. Zhou YJ, Buijs NA, Zhu Z, Qin J, Siewers V, Nielsen J. Production of fatty 565

acid-derived oleochemicals and biofuels by synthetic yeast cell factories. Nat 566

Commun. 2016;7:11709. 567

34. Yu T, Zhou YJ, Huang M, Liu Q, Pereira R, David F, et al. Reprogramming 568

yeast metabolism from alcoholic fermentation to lipogenesis. Cell. 569

2018;174:1549-58. 570

35. Casadaban MJ, Martinezarias A, Shapira SK, Chou J. Beta-galactosidase gene 571

fusions for analyzing gene-expression in Escherichia coli and yeast. Method 572

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

29

Enzymol. 1983;100:293-308. 573

36. Sasagawa T, Matsui M, Kobayashi Y, Otagiri M, Moriya S, Sakamoto Y, et al. 574

High-throughput recombinant gene expression systems in Pichia pastoris 575

using newly developed plasmid vectors. Plasmid. 2011;65:65-9. 576

37. Krivoruchko A, Zhang YM, Siewers V, Chen Y, Nielsen J. Microbial 577

acetyl-CoA metabolism and metabolic engineering. Metab Eng. 578

2015;28:28-42. 579

38. Zhang Y, Adams IP, Ratledge C. Malic enzyme: the controlling activity for 580

lipid production? Overexpression of malic enzyme in Mucor circinelloides 581

leads to a 2.5-fold increase in lipid accumulation. Microbiology. 582

2007;153:2013-25. 583

39. Wasylenko TM, Ahn WS, Stephanopoulos G. The oxidative pentose phosphate 584

pathway is the primary source of NADPH for lipid overproduction from 585

glucose in Yarrowia lipolytica. Metab Eng. 2015;30:27-39. 586

40. Sakaguchi K, Matsuda T, Kobayashi T, Ohara J, Hamaguchi R, Abe E, et al. 587

Versatile transformation system that is applicable to both multiple transgene 588

expression and gene targeting for thraustochytrids. Appl Environ Microbiol. 589

2012;78:3193-202. 590

41. Dong Q, Wang Y, Qi S, Gai K, He Q, Wang Y. Histone variant H2A.Z 591

antagonizes the positive effect of the transcriptional activator CPC1 to regulate 592

catalase-3 expression under normal and oxidative stress conditions. Free Radic 593

Biol Med. 2018;121:136-48. 594

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

30

42. Huang J, Xia J, Yang Z, Guan F, Cui D, Guan G, et al. Improved production of 595

a recombinant Rhizomucor miehei lipase expressed in Pichia pastoris and its 596

application for conversion of microalgae oil to biodiesel. Biotechnol Biofuels. 597

2014;7:111. 598

43. Lippmeier JC, Crawford KS, Owen CB, Rivas AA, Metz JG, Apt KE. 599

Characterization of both polyunsaturated fatty acid biosynthetic pathways in 600

Schizochytrium sp. Lipids. 2009;44:621-30. 601

44. Miller GL. Use of dinitrosalicylic acid reagent for determination of reducing 602

sugar. Anal Chem. 1959;31:426-8. 603

45. Enshaeieh M, Nahvi I, Madani M. Improving microbial oil production with 604

standard and native oleaginous yeasts by using Taguchi design. Int J Environ 605

Sci Te. 2014;11:597-604. 606

46. Gao XF, Yang Y, Liang ZL, Zheng JJ. A comparison of three indophenol blue 607

methods for measuring ammonium in five types of soils in China. Adv Mater 608

Res-Switz. 2013;634-638:59-63. 609

47. Li Z, Yuan H, Yang J, Li B. Optimization of the biomass production of oil 610

algae Chlorella minutissima UTEX2341. Bioresour Technol. 611

2011;102:9128-34. 612

48. Enshaeieh M, Abdoli A, Madani M, Bayat M. Recycling of lignocellulosic 613

waste materials to produce high-value products: single cell oil and xylitol. Int 614

J Environ Sci Te. 2015;12:837-46. 615

49. Madani M, Enshaeieh M, Abdoli A. Single cell oil and its application for 616

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

31

biodiesel production. Process Saf Environ. 2017;111:747-56. 617

50. Ren LJ, Huang H, Xiao AH, Lian M, Jin LJ, Ji XJ. Enhanced docosahexaenoic 618

acid production by reinforcing acetyl-CoA and NADPH supply in 619

Schizochytrium sp. HX-308. Bioprocess Biosyst Eng. 2009;32:837-43. 620

621

622

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

32

Figure Captions 623

624

Fig. 1 Development of a β-galactosidase reporter system and selection of strong 625

promoters in Schizochytrium sp. a Construction of reporter plasmid pPICZαA-lacZ. b 626

PCR verification of transformants. +: positive control, reporter plasmid was used as 627

template; -: negative control, WT genomic DNA was used as template; AOX1p-lacZ, 628

ubip-lacZ, EF-1αp-lacZ, ccg1p-lacZ, and TEF-1p-lacZ: the transformants of 629

pPICZαA-AOX1p-lacZ, pPICZαA-ubiquitinp-lacZ, pPICZαA-EF-1αp-lacZ, 630

pPICZαA-ccg1p-lacZ, and pPICZαA-TEF-1p-lacZ. c Phenotypes of the 631

corresponding transformants. Cells were grown for 96 h on GPY plates with 40 632

μg/mL X-gal. d β-galactosidase enzymatic activities of the transformants. Cells were 633

grown in seed medium for 64 h 634

635

Fig. 2 Overexpression of ATP-citrate lyase and acetyl-CoA carboxylase in 636

Schizochytrium sp. a Schematic diagram of Schizochytrium sp. ACL and ACC. The 637

proposed enzymatic domains are marked in different colors. b PCR verification of the 638

transformants. +: positive control, the plasmid was used as template; -: negative 639

control, WT genomic DNA was used as template; OACL, OACC, and OACL-ACC: 640

genomic DNA from the transformants of pPICZαA-ACL, pPICZαA-ACC, and 641

pPICZαA-ACL-ACC was used as template 642

643

Fig. 3 qRT-PCR analysis of the transcription levels of ACL and ACC in WT, OACL, 644

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

33

OACC, and OACL-ACC. RNAs were isolated from WT, OACL, OACC, and 645

OACL-ACC grown in fermentation media for 2 and 4 days. P values were determined 646

by Student’s t-test. ***, P < 0.001; NS, not significant 647

648

Fig. 4 Time course of fermentation profiles of Schizochytrium sp. WT, OACL, OACC, 649

and OACL-ACC. a Glucose (g/L); b NH4+ (g/L); c pH; d Dry cell weight (DCW, g/L) 650

651

Fig. 5 Effect of ACL and ACC overexpression on lipid accumulation and DHA 652

production by Schizochytrium sp. a Dry cell weight (DCW, g/L). b Total lipids (% 653

DCW). c DHA yield (g/L). Cells were cultured in fermentation medium for 120 h. d 654

Imaging analyses of WT, OACL, OACC, and OACL-ACC. The 48-h cultured cells 655

were stained with nile red dye for neutral lipid staining. Scale bar, 10 µm. Error bars: 656

SD from three independent experiments. Data were analyzed by one-way ANOVAs 657

and Duncan’s multiple range tests in SPSS version 23.0. In a, b, and c, columns with 658

different lowercase letters are significantly different at P < 0.05 659

660

Fig. 6 Effect of ACL and ACC overexpression on fatty acid composition (TFA, %) of 661

Schizochytrium sp. Cells were cultured in fermentation medium for 120 h 662

663

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

34

Tables 664

Table 1 Fermentation characteristics of strains of Schizochytrium sp. 665

Strains DCW (g/L) Lipid yield

(g/L)

Lipid content

(%)

DHA yield (g/L) DHA content

(%)

WT 23.7 ± 0.6a 12.8 ± 0.5d 53.8c 4.3 ± 0.1c 36.4b

OACL 22.7 ± 1.0a 15.6 ± 0.4c 68.8b 5.3 ± 0.2b 36.2b

OACC

OACL-ACC

23.9 ± 1.1a

24.3 ± 1.1a

16.7 ± 0.7b

17.7 ± 0.3a

69.8b

73.0a

6.1 ± 0.1a

6.4 ± 0.2a

37.6a

37.9a

Cells were cultured in fermentation medium for 120 h. Data were analyzed by one-way ANOVAs and Duncan’s 666

multiple range tests in SPSS version 23.0. Values with different lowercase letters are significantly different at P < 667

0.05 668

669

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

35

Table 2 Strains and plasmids used in this study 670

671

672

Strain or plasmid Description Source or reference

Schizochytrium sp.

ATCC 20888

AOX1p-lacZ

ubip-lacZ

TEF-1p-lacZ

EF-1αp-lacZ

ccg1p-lacZ

wild-type strain (WT)

WT strain carrying pPICZαA-AOX1p-lacZ

WT strain carrying pPICZαA-ubiquitinp-lacZ

WT strain carrying pPICZαA-TEF-1p –lacZ

WT strain carrying pPICZαA-EF-1αp-lacZ

WT strain carrying pPICZαA-ccg1p-lacZ

ACL overexpression strain

American Type Culture

Collection

This study

This study

This study

This study

This study

This study OACL

OACC

OACL-ACC

ACC overexpression strain

ACL and ACC co-overexpression strain

This study

This study

E. coli

JM109 General cloning host for plasmid manipulation Laboratory stock

Plasmids

pPICZαA Yeast expression vector [42]

pPICZαA-AOX1p-lacZ

pPICZαA-ubiquitinp-lacZ

pPICZαA-TEF-1p-lacZ

pPICZαA-EF-1αp-lacZ

pPICZαA-ccg1p-lacZ

lacZ reporter vector using AOX1 promoter and terminator

lacZ reporter vector using ubiquitin promoter and terminator

lacZ reporter vector using TEF-1 promoter and terminator

lacZ reporter vector using EF-1α promoter and terminator

lacZ reporter vector using ccg1 promoter and terminator

This study

This study

This study

This study

This study

pPICZαA-ACL ACL overexpression vector based on pPICZαA This study

pPICZαA-ACC

pPICZαA-ACL-ACC

ACC overexpression vector based on pPICZαA

ACL and ACC co-overexpression vector based on pPICZαA

This study

This study

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

36

Additional files 673

674

Additional file 1: Figure S1. Physical map of overexpression plasmid 675

pPICZαA-ACL. 676

677

Additional file 2: Table S1. The open reading frames of ACL and ACC genes in 678

Schizochytrium sp. Table S2. Primers used in this study. 679

680

681

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65

Fig. 1

BackFront

a

bp

AOX1p ubiquitinp EF-1αp ccg1p TEF-1p

M + - + - + - + - + -

5000 3000 2000

1000

WT

ccg1p-lacZ

b

c

d

0 8 1 6 2 4 3 2 4 0 4 8 5 6 6 4

0

3 0

6 0

9 0

1 2 0

1 5 0 W T

A O X 1 p -la c Z

E F -1 p - la c Z

c c g 1 p - la c Z

T E F -1 p - la c Z

u b ip - la c Z

A O X 1 p -la c Z -M e th a n o l

T im e (h )

-g

ala

cto

sid

as

e a

ctiv

ity

(U

/mL

)

revised Figures Click here to access/download;Figure;revisedFigures.pptx

https://www.editorialmanager.com/bbio/download.aspx?id=158455&guid=a7d39360-5192-4722-a500-9c3f4b69855f&scheme=1

https://www.editorialmanager.com/bbio/download.aspx?id=158455&guid=a7d39360-5192-4722-a500-9c3f4b69855f&scheme=1

Fig. 2

ATP-citrate lyase

ATP citrate-lyase domain Citrate-binding domain

1 422

ACL ACC ACL-ACC

bp

10000

6000

3000

2000

79 356 411

Acetyl-CoA carboxylase

1 2226 2352

332

Biotin carboxylase Biotin carboxyl carrier protein Carboxyl transferase

35 531 670 735 1669

8000

M + - + - + -

a

b

Fig. 3

2 4

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5

A C L

D a y s

Re

lativ

e m

RN

A e

xp

re

ss

ion

W T

O A C L

O A C C

O A C L -A C C

***

N S

****** ***

N S

a b

2 4

0 .0

0 .5

1 .0

1 .5

2 .0

2 .5

A C C

D a y s

Re

lativ

e m

RN

A e

xp

re

ss

ion

W T

O A C L

O A C C

O A C L -A C C

***

N S

*** ***

***

N S

Fig. 4

b a

c d

0 1 2 3 4 5 6 7

0

5

1 0

1 5

2 0

2 5

3 0

D a y s

Dr

y c

ell

we

igh

t (

g/L

)

W T

O A C L

O A C C

O A C L -A C C

0 1 2 3 4 5 6 7

0 .0 0

0 .0 3

0 .0 6

0 .0 9

0 .1 2

0 .1 5

D a y s

NH

4

+ (

g/L

)

W T

O A C L

O A C C

O A C L -A C C

0 1 2 3 4 5 6 7

0

2 0

4 0

6 0

8 0

1 0 0

1 2 0

D a y s

Glu

co

se

(g

/L)

W T

O A C L

O A C C

O A C L -A C C

Fig. 5

d

c ba

WT

OA

CL

OA

CC

OA

CL

-AC

C

0

2

4

6

8

DH

A y

ield

(g

/L)

aa

b

c

WT

OA

CL

OA

CC

OA

CL

-AC

C

0

1 0

2 0

3 0

Dr

y c

ell

we

igh

t (

g/L

) aaaa

WT OACL OACC OACL-ACC

488 nm

Light

Merge

Fig. 6

Additional file 1

Click here to access/downloadSupplementary Material

Additional file 1.pptx

https://www.editorialmanager.com/bbio/download.aspx?id=158469&guid=7102ef9a-6994-4bed-82ca-fdef23cd8995&scheme=1

revised Additional file 2

Click here to access/downloadSupplementary Material

revised Additional file 2.docx

https://www.editorialmanager.com/bbio/download.aspx?id=158461&guid=08376116-5c48-47ae-8643-73ad3eb268b9&scheme=1

Response to Reviewers

Biotechnology for Biofuels

BBIO-D-19-00571

Enhancement of docosahexaenoic acid production by overexpression of ATP-citrate

lyase and acetyl-CoA carboxylase in Schizochytrium sp.

Reviewer #1: It's a good work with significant improvement of DHA production as

well as a tool box for further application in biofuels production in this field. There are

some comments hope authors to take into consideration.

1. The work focused on the availability of carbon supply to FA synthesis, however,

the desaturation is also very important to DHA production. The whole profile

dynamic information of FA should be provided to have a overall view on FA

synthesis process other than Fig 5.

Response: Thank you very much for your valuable comments and suggestions. We

agree with you that the investigation of the whole dynamic profile will provide more

information. Though we can determine DHA yields through gas chromatography, our

lab can not analyze the fatty acid composition due to the lack of some fatty acid

standards. Fatty acid composition analysis in the study was carried out by the

detection platform of our university which is closed now and will not open in months

due to the outbreak of new coronavirus. So we can not do the experiment. Guo et al.

(2016) have reported that the changes of fatty acid composition in Schizochytrium sp.

HX-308 cultured at different aeration rates. The contents of DHA in TFAs from 40-h

to 120-h were increased gradually from 38.01% to 43.59% and from 40.52% to 44.85

at aeration rates of 0.5 vvm and 1.0 vvm, respectively, and decreased gradually from

39.91% to 36.61% at aeration rate of 1.5 vvm. According to the report, DHA content

of Schizochytrium sp. was slightly changed during fermentation and was affected by

the oxygen supply.

Reference:

Guo DS, Ji XJ, Ren LJ, Li GL, Yin FW, Huang H. Development of a real-time

For reviewers Click here to access/download;Personal Cover;Response toReviewers.docx

https://www.editorialmanager.com/bbio/download.aspx?id=158465&guid=851c7aaa-ac2f-4334-9f0a-8c0ad17a69d0&scheme=1

https://www.editorialmanager.com/bbio/download.aspx?id=158465&guid=851c7aaa-ac2f-4334-9f0a-8c0ad17a69d0&scheme=1

bioprocess monitoring method for docosahexaenoic acid production by

Schizochytrium sp. Bioresour Technol. 2016;216:422-427.

2. In the industrial process, the glycerol is widely used as carbon other than glucose,

so the performance with glycerol is expected.

Response: Thank you very much for your suggestion. As a byproduct in the

production of biodiesel, glycerol is a cost effective carbon source for lipid production.

Glycerol was used as the carbon source for the shake flask fermentation as suggested.

Although it was reported that some Schizochytrium spp. could use glycerol efficiently,

we found here that Schizochytrium sp. grew poorly in the glycerol medium and the

production of lipid and DHA were very low (Fig. I). We were quite surprised by the

results, and the shake flask fermentation experiments were repeated with new batch of

glycerol, and lower amount of glycerol (60 g/L) was also used to rule out the

possibility of the inhibition of high glycerol on growth. Schizochytrium sp. still grew

poorly in the glycerol fermentation medium (Fig. IIa, b). Glucose fermentation

medium was also used as a control in which Schizochytrium sp. grew well and

accumulated lipid efficiently (Fig. IIc, d). Even though Schizochytrium sp. grew

poorly in glycerol medium, the lipid contents and DHA yields of ACL and ACC

overexpression strains were slightly higher than those of the WT strain (Fig. I).

Therefore, overexpression of ACL and ACC did increase the production of lipid and

DHA in Schizochytrium sp. ATCC 20888. Since cells grew poorly in glycerol

medium, we would rather not include the results of the glycerol medium in the

manuscript. We don’t know why this strain grow poorly with glycerol, but it seems

Schizochytrium spp. differ very much from each other, for example, the reported

genome sizes of Schizochytrium limacinum SR21 and Schizochytrium sp. CCTCC

M209059 were 63 Mb and 39 Mb, respectively.

Fig. I Effect of ACL and ACC overexpression on dry cell weight (DCW), lipid accumulation, and

DHA production by Schizochytrium sp. in the glycerol medium. Cells were grown in fermentation

medium with 100 g/L glycerol as the carbon source for 120 h. a Dry cell weight (DCW, g/L). b

Total lipids (% DCW). c DHA yield (g/L)

Fig. II Effect of ACL and ACC overexpression on dry cell weight (DCW) and lipid

accumulation by Schizochytrium sp. Cells were grown in fermentation medium for 120 h with

60 g/L glycerol (a), 100 g/L glycerol (b), 100 g/L glucose (c, d) as the carbon source. a, b, c

a b c

WT

OA

CL

OA

CC

OA

CL

-AC

C

0

1 0

2 0

3 0

4 0

Lip

id c

on

te

nt (

%)

b cab

a

c

WT

OA

CL

OA

CC

OA

CL

-AC

C

0 .0

0 .2

0 .4

0 .6

0 .8

1 .0

DH

A y

ield

(g

/L)

ab

cc

0 1 2 3 4 5

0

2

4

6

8

1 0

D a y s

Dr

y c

ell

we

igh

t (

g/L

)

W T

O A C L

O A C C

O A C L -A C C

a b

c d

0 1 2 3 4 5

0

2

4

6

8

1 0

D a y s

Dr

y c

ell

we

igh

t (

g/L

)

W T

O A C L

O A C C

O A C L -A C C

0 1 2 3 4 5

0

2

4

6

8

1 0

D a y s

Dr

y c

ell

we

igh

t (

g/L

)

W T

O A C L

O A C C

O A C L -A C C

0 1 2 3 4 5

0

1 0

2 0

3 0

D a y s

Dr

y c

ell

we

igh

t (

g/L

)

W T

O A C L

O A C C

O A C L -A C C

WT

OA

CL

OA

CC

OA

CL

-AC

C

0

2 0

4 0

6 0

8 0

Lip

id c

on

te

nt (

%)

aab

c

Dry cell weight (DCW, g/L). d Total lipids (% DCW)

3. The expression of digital should be improved, and comma is expected to be used.

Such as 1,269-bp fragment and 6,000 rpm, etc.

Response: Thank you for the suggestion. We have corrected the expression of digital

in the text.

Reviewer #2

I have now reviewed the manuscript, which was well-written in many aspects

including abstract, backrounds, m&m and conslusion. However, the discussion

section must be re-written since it was written as a the repeatition of the results

section. References, Figures and Tables are enough. Some methods are also missing.

Therefore, I suggest major revision. After revision, it should be reconsidered. My

spesific comments can be found on the text attached.

Response: Thank you very much for your valuable comments and suggestions. As

suggested, we have carefully revised the manuscript, added some methods and

references, and reorganized the discussion section. Most active tense sentences were

revised to passive tense except some (Line 120 and Line 175) in which the change

will lead to inconsistent subject. Some specific concerns are as follows.

1. Line 25: Is the host or wild type? Please specify it. Line 28: What are they, please

specify them.

Response: The specification of the strains was listed in Table 2.

2. Line 31: What is the difference between the Lipid yields and DHA yields?

Response: Thank you for the question. Lipid included triacylglycerol, free fatty acids,

and phospholipids, etc, which was extracted by acid-heating extraction. The fatty acid

components of lipids were DHA, palmitic acid (C16:0), myristic acid (C14:0), and

DPA, etc (Fig. 6). For DHA yields determination, fatty acid methyl esters were firstly

prepared from lipid samples and DHA was measured by gas chromatography.

3. Line 133, “β-galactosidase with very low activity”: please give some critical

activity in the text.

Response: Only qualitative colorimetric detection was carried out on GPY plates. The

β-galactosidase activities of the transformants in the liquid medium (seed medium)

were shown in Fig. 1d.

4. Line 176, “ACL and ACC”: Italic or non-italic. Please be coherent.

Response: The italic form indicates the encoding gene. And non-italic form indicates

protein or enzyme. It was changed to be “ACL and ACC genes”.

5. Line 191: “Thus, overexpression of ACL and/or ACC did not affect cell growth of

Schizochytrium”. Any reason.

Response: The results indicated that overexpression of ACL and ACC did not

significantly affect the consumption of carbon and nitrogen sources of the

transformants, that is probably the reason why cell growth (dry cell weights) was not

affected.

6. Line 206: “Overexpression of ACL in Schizochytrium sp. WT did not significantly

affect the percentage of TFAs represented by DHAs……”, Why? Any reason?

Line 596 (Table 1): Although DHA yields, Lipid yields, and Lipid contents of the

recombinant strains were higher than that of WT, DHA contents were similar to

each other. Any reason.

Response: Overexpression of ACL and ACC enhanced the supply of precursors

(acetyl-CoA and malonyl-CoA) for the synthesis of DHA and other fatty acids.

Therefore, the increased precursors supply in the recombinant strains enhanced the

production of DHA and other fatty acids at the same time. That is the reason of the

similar DHA contents between WT and the recombinant strains. The related

discussion can be found in Line 251-260.

7. The discussion section was similar to the results section or was written as a

repeatition of the results. Please compare your results with the existing results in

the literature. What were the similarities and differencies of your results with the

data in the literature? Please discuss them. Also, what is the industrial significance

of this study? Please also elucidate this situation. To sum up, please re-write the

discussion section.

Response: Thank you very much for your suggestion. In the discussion section, we

summarized the main findings in the study which is the foundation for the follow-up

discussion. These sentences have been rephrased to avoid repetition. And the

discussion section has been carefully revised as suggested.

8. Line 270 (Microorganisms and culture conditions section): There are four

different medium in this section, please use references for each medium and

explain why the medium was different?

Response: The references have been added to the text. LB medium is used for

cultivation of E. coli in which plasmids were constructed. The media and growth

conditions for Schizochytrium sp. were according to Ling et al. [5] with modifications.

In fact, we did quite a lot of work to optimize the media (especially the fermentation

medium). GPY solid medium which contained less glucose were used for rapid

cultivation of Schizochytrium sp. and selection of single colony of the transformants

(supplemented with 40 μg/mL zeocin). Fermentation medium which contained high

concentration of glucose was used for lipid and DHA production (high C/N helps lipid

accumulation). And seed medium which had moderate concentration of glucose

guaranteed rapid growth of Schizochytrium sp. and ample cells for fermentation

medium. Cells would grow slowly in fermentation medium if they were inoculated

directly from GPY solid medium.

9. Line 289: Why the promoters and terminators were amplified from different

sources? Please elucidate it.

Response: We intended to test the promoter strengths of some commonly-used

eukaryotic promoters (ubiquitinp, EF-1αp, ccg1p, TEF-1p, and AOX1p) and screen for

strong promoters in Schizochytrium (Line 145-162). Therefore, some promoters and

terminators were amplified directly from fungi and yeast expression plasmids, and

some were amplified from Schizochytrium sp. itself (endogenous promoters can be

recognized by RNA polymerase more efficiently). The results indicated that ccg1p,

TEF-1p, and AOX1p from the commonly-used expression plasmids displayed stronger

promoter activities, which was not unexpected because expression plasmids usually

utilize very strong promoters to ensure the efficient expression of the target genes.

Identification of promoters with different promoter strengths can provide suitable

promoters for a balanced expression of the relevant pathways in Schizochytrium.

10. The fermentations in Fig. 1d and Fig. 3a-d can be modeled kinetically using

logistic, Luedeking piret and modified luedeking piret models, but not this study.

It can be interesting.

Response: Thank you so much for your kind suggestion. We will try to employ these

models for growth and product modelling in the future study. It will be very

interesting.

Reviewer #3

The manuscript by Han et al. tested several constitutive promoters from

Schizochytrium sp. and used the strong ccg1p promoter to overproduce ACL and

ACC, which increased the yields of DHA and total lipids.

Comments:

1. How many independent strains of each transformant were generated and used for

the study? It seems like there is only one strain for each construct, which is not

acceptable.

Response: Thank you very much for your valuable comments and the question. After

transformation, we have selected eight independent strains for each kind of

transformants. The transformants were cultured in fermentation broth for 5 days,

stained with nile red dye for neutral lipid staining, and the fluorescence intensities

were measured by a multifunctional plate reader. All transformants displayed higher

fluorescence intensities than the wild-type strain, indicating that the transformants

accumulated more lipids than the WT strain (Fig. IIIa, b, c). After preliminary

screening, two independent strains with the highest fluorescence for each kind of

transformants were cultured for determination of lipid and DHA yields. The same

kind of transformants displayed similar results (Fig. IIId). All fermentation

experiments were at least repeated twice. For clarity, results from one strain of each

kind of transformants were provided in the text.

b a

d c

Fig.III Effect of ACL and ACC overexpression on lipid accumulation by Schizochytrium sp.. a, b,

c Nile red staining of WT, OACL, OACC, and OACL-ACC. d Lipid contents of WT, OACL,

OACC, and OACL-ACC. The strains were cultured in fermentation medium for 120 h

2. RNA or Protein levels of ACL and ACC should be examined in the

overexpressing strains.

Response: Thank you very much for your suggestion. The transcriptional levels of

ACL and ACC were determined by qRT-PCR in WT, OACL, OACC, and OACL-ACC

which were cultivated in fermentation media for 2 and 4 d. The transcriptional levels

of ACL and ACC were greatly increased in the corresponding transformants. The

results were added to the text (Line 188-194) and the new Fig. 3.

3. Lipid content was increased from 54% of the WT to ~70% of the overexpressing

strains. Figure 4d, why the nile red staining shows such low level of signals in the

WT? Did the expression of ACL and ACC changed the morphology of lipid

bodies? i.e. size and numbers.

Response: Thank you for the question. The nile red staining is a very efficient means

to indicate the intracellular lipid accumulation. During our nile red staining

preliminary screening, we found that the fluorescence signals had a positive

correlation with the lipid contents, but the correlation was not strictly linear. The

increase rates of fluorescence signals were usually higher than the increase rates of

lipid contents (please see Fig. III, response to comment 1). Besides, fluorescence

attenuates during microscopic observations. WT was the first one observed and its

analysis might take longer time than the others’. Microscopic analysis was repeated

with 48-h cultured cells (it will be easier to observe the morphology of lipid bodies

with “younger” cells) and all photos were taken with the same observation time. Fig.

4d was replaced with the new figure (now as Fig. 5d), and the relative fluorescence

intensities of cells were analyzed by ImageJ (Fig. IV), which had a good correlation

with the lipid contents. We thank you very much for your kind reminder. Since

Schizochytrium cells aggregated and produced zoospores during cultivation, it was

difficult to distinguish lipid bodies. From what we saw, Schizochytrium formed a large

lipid body per cell and no significant difference of morphology wad noticed between

WT and the transformants (Fig. 5d).

Fig.IV The relative fluorescence intensity of WT, OACL, OACC, and OACL-ACC

4. Explain or discuss the changes of fatty acid composition especially on the

increase of DHA in ACC and ACL-ACC.

Response: Thank you for your suggestion. Please see answer to comment 6 of

Reviewer 2. The discussion about the changes of fatty acid composition on the

increase of DHA in OACC and OACL-ACC strains can be found in Line 251-260.

Reviewer #4

Some grammatical problems are seen in the whole manuscript which should be

revised. Also, the organization of the manuscript especially in the introduction part is

not good. One more attention is needed in the discussion part. Some related papers

with for lipid extraction (with modifications) are introduced here for improvement of

the lipid extraction part, determination of biomass dry weight and some other part in

the material and method, cite these references in this part. Paper titles are in the

parentheses. Add a table in discussion part and compare related studies.

WT

OA

CL

OA

CC

OA

CL

-AC

C

0

2 0

4 0

6 0

8 0

Flu

or

es

ce

nc

e i

nte

ns

ity

Best wishes

1- Process Safety and Environmental Protection III (2017) 757-765

http://dx.doi.org/10.1016/j.psep.2017.08.029 (Optimizing of lipid production in

Cryptococcus heimaeyensis through M32 array of Taguchi design)

2- Process Safety and Environmental Protection III (2017) 747-756

http://dx.doi.org/10.1016/j.psep.2017.08.027 (Single cell oil and its application for

biodieselproduction)

3- Int. J. Environ. Sci. Technol. DOI 10.1007/s13762-014-0687-8 (Recycling

of lignocellulosic waste materials to produce high-value products: single cell oil and

xylitol)

4- Int. J. Environ. Sci. Technol. (2014) 11:597-604 DOI

10.1007/s13762-013-0373-2 (Improving microbial oil production with standard and

native oleaginous yeasts by using Taguchi design)

5- (Medium optimization for biotechnological production of single cell oil

using Yarrowia lipolytica M7 and Candida sp.)

6- (Selection and optimization of single cell oil production from Rodotorula

110 using environmental waste as substrate)

Response: Thank you very much for your kind suggestions and the recommended

related references. We have carefully revised the manuscript and reorganized the

discussion section. We have earnestly read the recommended references with great

interest. Three related references have been cited in the methods (determination of

glucose and lipid extraction). We will consider the methods in the related references

in the future study.

http://dx.doi.org/10.1016/j.psep.2017.08.029

http://dx.doi.org/10.1016/j.psep.2017.08.027

Dear editor Hallenbeck,

Thank you very much for your decision letter of BBIO-D-19-00571 (Enhancement of

docosahexaenoic acid production by overexpression of ATP-citrate lyase and

acetyl-CoA carboxylase in Schizochytrium sp.). We have carefully revised the

manuscript to address all the comments of the editor and reviewers. New data

including qRT-PCR analysis of ACL and ACC expression (Fig. 3) were added to the

revised manuscript. Some methods and references were added to the text, and the

discussion section was reorganized. The modifications are marked in yellow in the

marked-up manuscript. Details of our revision are included in our point-by-point

“Response to Reviewers”. Some illustrated figures are added to the “Response to

Reviewers”, and the “Response to Reviewers” file has been submitted in the

Biotechnology for Biofuels system.

We would like to thank you and the reviewers for careful reviewing and valuable

comments which helped to improve the manuscript.

Sincerely yours,

Zhi Chen

State Key Laboratory of Agrobiotechnology and Key Laboratory of Soil Microbiology,

Ministry of Agriculture, College of Biological Sciences, China Agricultural University,

Beijing 100193, China

Cover letter for revised Click here to access/download;Personal Cover;Cover letter forrevised.docx

https://www.editorialmanager.com/bbio/download.aspx?id=158466&guid=5e3ea348-75f1-4849-ad64-7ca3aff3da83&scheme=1

https://www.editorialmanager.com/bbio/download.aspx?id=158466&guid=5e3ea348-75f1-4849-ad64-7ca3aff3da83&scheme=1

Documents

Biotechnology for Biofuels