Breeding of New Strains of Gracilariopsis lemaneiformis with High Agar Content by ARTP Mutagenesis and High Osmotic Pressure Screening

ARTP (atmospheric and room temperature plasma mutagenesis) mutagenesis was tried on G. lemaneiformis, and mutagenesis conditions were confirmed. An osmotic pressure screening program was established. Mutants were identified and characterized of relevant physiological traits. The aim of the study is to try to use ARTP mutagenesis and osmotic pressure screening for the breeding of high-agar G. lemaneiformis. Treatment time of 46 s was found to be an optimal mutagenesis time. The mutagenized spores were initially screened with 58‰ salinity artificial seawater, and then, the surviving spores were screened twice with 60‰ salinity artificial seawater in their vertical growth phase and branch growth phase, respectively. Four fast-growing and hypertonic resistance gametophytes were selected. The actual photosynthetic efficiency [Y(PSII)], photochemical quenching (qL), and non-photochemical quenching (NPQ) of four mutants were measured. The values of Y(PSII) and qL of HAGL-X3 and HAGL-X5 were higher than those of the control in the early stage of salt stress. NPQs of HAGL-X3 and HAGL-X5 were higher than control in most of the times. The growth rates of the four mutants were higher than that of the control. HAGL-X4 was the highest. The agar content was measured; HAGL-X5 displayed the highest agar content among the tested strains. HAGL-X5 was more in line with expectations, because of its high agar content and good hypertonic resistance. In this study, the mutant of G. lemaneiformis with high agar content was obtained by the procedure, which provided a certain reference for the selection of G. lemaneiformis strains with high agar content.


Introduction
Gracilariopsis lemaneiformis is a kind of marine red algae, which has very important economic value (Qi et al. 2010;Li et al. 2008;Kang et al. 2016). It is a source of agar, an indispensable additive in medical and food industries (Selby and Whistler 1993). The main purpose of G. lemaneiformis breeding in the past was to obtain varieties of high-temperature tolerance, fast growth, and high biomass accumulation, such as the variety "981" and "07-2" (Zhang et al. 2012). However, there is no precedent for the selection and breeding of high agar varieties. This may be partly because of lacking of certain technical means for breeding of the G. lemaneiformis with high agar.
There were cases of microalgae of breeding of highyield secondary metabolite strains via hyperosmotic screening. For example, the breeding of diatom Nitzschia sp. with high lipid content. Due to the fact that osmotic pressure can affect the lipid content of diatom principle, mutagenized diatoms were screened for high lipid strains using high osmotic pressure, and target mutants were obtained (Cheng et al. 2014). Similarly, it was also reported that the agar content of G. lemaneiformis was increased from both hypertonic and hypotonic cultures, which suggested that the content of agar is affected by the physical quantity of osmotic pressure (Bird 1988;Hurtado-Ponce and Pondevida 1997;Hu et al. 2019) Therefore, it is reasonable to exploit osmotic pressure screening for breeding of G. lemaneiformis strains with high agar contents.
With appropriate equipment, physical methods achieve mutagenesis conveniently, efficiently, and safely. Based on its penetrating power, there are no restrictions on the material, including which atmospheric and room temperature plasma (ARTP) is one of the widely used for recent years. The ARTP developed in recent years has gradually become the preference choice of physical mutagenesis mean due to its safety, efficiency, and non-polluting advantages (Li et al. 2007(Li et al. , 2011. ARTP acted on the DNA of the organism and caused DNA damage, which led to SOS repair and mutagenesis. In recent years, ARTP mutagenesis has been widely used in microorganisms and microalgae, such as breeding of Bacillus cereus producing high-yield chitosanase (Zhang et al. 2021), breeding of high acarbose-producing Actinomyces (Ren et al. 2017), and improving of lipid productivity of the oleaginous microalgae Chlorella pyrenoidosa (Cao et al. 2017).
In the present study, ARTP mutagenesis was tried on G. lemaneiformis. Mutagenesis conditions of ARTP were confirmed. An osmotic pressure screening program was established. Mutants were identified and characterized of relevant physiological traits. The aim of the study is to try to use ARTP mutagenesis and osmotic pressure screening for the breeding of high agar G. lemaneiformis. The research laid a foundation for the technical approach to ARTP mutagenesis and further cultivar development of G. lemaneiformis.

Experimental Materials
G. lemaneiformis used in the experiment was WLP-1 which was tetrasporophyte and provided by the Key Laboratory of Marine Genetics and Breeding Ministry of Education (Ocean University of China). Culture conditions were under the temperature of 20 ℃, with light intensity of 15 μmol m -2 s -1 , and the light-dark cycle of 12 L:12 D.

Cultivation of Experimental Materials and Collection of Tetraspores of G. lemaneiformis
The young branches were cut from WLP-1 and cultured with seawater supplemented with Provasoli medium (Provasoli 1968) at a temperature of 20 °C and a light intensity of 30 μmol m -2 s -1 . The light-dark cycle is 12 L:12 D, and the medium is changed every 3 days. After the shoots mature, they were placed under the conditions of 15 μmol m -2 s -1 light intensities to be induced to release of spores (Zhou et al. 2013). The released spores were centrifuged at 500 G for 15 min, and then 200 G for 10 min for concentration. The precipitated spores were resuspended in certain volume of sea water for quantitation using a hemocytometer.

ARTP Treatment
The mutagenic power of ARTP (Wuxi Yuanqing Tianmu Biological Technology Co., Ltd, China) was 120 W. The highpurity helium gas volume was set to 10 SLM [SLM (selenomethionine) is a unit of ventilation, expressed in standard liters per minute], the air pressure was controlled between 0.1 and 0.2 MPa, and processing distance was 2 mm. The temperature of the instrument was set to 20 °C.

Determination of Survival Rate and Death Rate
The time-lethal rate curve was drawn using collected spores with a concentration of approximately 1.85 × 10 5 /ml. Six time gradients of 0 s, 10 s, 20 s, 30 s, 40 s, and 50 s (s stands for seconds) were adopted, and three parallels were set for each mutagenesis time. After mutagenesis, the tetraspores were put into a six-well plate, and then, 10 ml of sterilized seawater was added. Spores were cultured statically under 20 °C, a light intensity of 15 μmol m -2 s -1 , and a light-dark cycle of 12 L:12 D. The sterilized seawater was changed every 2 days. After 1 week of cultivating, death rate was calculated under a microscope by the following formula: F (%) = (1-S/T) × 100%, where F (%) represents the mortality rate, S represents the number of surviving spores, and T represents the total number of spores.

Determination of Screening Salinity
Artificial seawater with high salinity was used for hypertonic screening after mutagenesis. The artificial seawater formula followed those in the literature with slightly modified (Berges et al. 2001), and the salinity was 50‰, 52‰, 54‰, 56‰, 58‰, and 60‰, respectively. Various salinities were achieved by changing the content of NaCl, and the contents of other elements were the same under each salinity. Three parallels were operated on each salinity gradient. About 6000 spores were used for one salinity gradient and was added to six-well plates in artificial seawater with different salinities. The sterilized seawater was changed every 2 days. After 1 week of cultivating, they were placed under a microscope and counted for survival and mortality using the following formula: S (%) = (1-F/T) × 100%, where S (%) represents the survival rate, F represents the number of dead spores, and T represents the total number of spores.

Confirmation of the High Osmotic Pressure Tolerance of Screen-Out Strains
The tested strains were cut, each with 30 algal tips, each 1 cm and divided into 3 parallels, each with 10 tips for parallel. Individual grown-up from releasing of WLP-1 was used 1 3 as control and treated in the same way. The algal tips were put in artificial seawater with salinity of 60‰ and cultivated statically under 20 °C, a light intensity of 15 μmol m -2 s -1 , and a light-dark cycle of 12 L:12 D for 24 days. Culture medium was changed every 2 days. The actual photosynthetic efficiency [Y(PSII)], photochemical quenching (qL), and non-photochemically quenched (NPQ) of the above algal tips were measured every 2 days.

Determination of Chlorophyll Fluorescence Parameters
Algal tips with the length of 1 cm were cut and treated with a salinity of 30‰ of artificial seawater for 1 day and then transferred to 60‰ of artificial seawater. The FluorCam fluorescence imaging system FC 800-C (Photon Systems Instruments, Czech Republic) was then used. Determination of chlorophyll fluorescence parameters of mutants and controls was performed every 2 days, generally at 8 am, start time point of light phase. Three parallels were set for each strain, and each parallel included ten algal tips. Parameters for determining photosynthetic efficiency [Y(PSII)], photochemical quenching (qL), non-photochemically quenched (NPQ) (Genty et al. 1989;Kramer et al. 2004;Bilger and Björkman 1990;Van Kooten and Snel 1990) were set as 10% culture light intensity, 35% saturated light intensity, 1 exposure, and 5% sensitivity.

Determination of the Growth Rate
The methods for determination of growth rate followed those in the literature with slightly modified (Araño et al. 2000). Thirty algal tips were cut from each sample, each with a length of 1 cm. The 30 algal tips were divided into three parallels, each with 10 tips, and a control group was set to do the same treatment. Culture conditions of the material were under the temperature of 20 ℃, with the light intensity of 15 μmol m -2 s -1 , and the light-dark cycle of 12 L:12 D. The seawater was changed every three day. The length of each algae tip was measured once a week, and the algal tip was cut back to 1 cm again. The material was measured for 4 weeks, and then, the growth rate of the mutant and the control group was calculated using the formula V L (cm/week) = (L-L 0 )/T. In the formula, L is the length of the algae at the end of the experiment (cm); L 0 is the length of the algae at the beginning of the experiment (cm); T is the culture time (weeks).

Determine of the Agar Content
The agar extraction method followed those in the literature with slightly modified (Villanueva et al. 2010). Fresh algae of 2.0 g were taken and placed in an oven at 60 °C to a constant weight (dry weight is about 0.2 g). NaOH of 2.5 mol/l was added at the rate of 400 μl per 0.01 g of algae (dry weight), and then, the material was treated in a water bath at 85 °C for 2 h. The processed material was filtered through 4 layers of gauze to remove the lye and washed three times with distilled water. The algae was put into a beaker containing double-distilled water and 0.1 mol/l HCl was added. After adjusting the pH of the algae to 6.5, the algae was washed three times with double-distilled water. The processed algae body was added with distilled water (1 g to 60 ml double distilled water) and placed in a pressure cooker (Model: CT88A8818052, Chitong Instrument Co., Ltd, China) at 120 °C for 2 h. The material was centrifuged quickly at 32,300 g for 1.5 min. The collected supernatant was placed in a container like a petri dish (about 9 cm in diameter) with a volume of about 70 cm 3 folded with foil paper which was weighed in advance. The foil paper with the samples was put into a clean petri dish and frozen overnight at −20 °C. The solidified agar was thawed at room temperature and rinsed several times with distilled water. The above-mentioned materials were then placed in an oven at 60 °C to be dried to a constant weight. The extract together with the foil paper was weighed and recorded, and then, the dry weight of the agar was obtained by subtracting the weight of the foil paper. The calculation formula of agar content can be expressed as AG (%) = [(W Fp+Ag -W Fp )/ W GL ] × 100%, where AG (%) represents the content of agar, and W Fp+Ag represents the total weight of foil paper and agar (dry weight), W Fp represents the weight of the foil paper, and W GL represents the dry weight of the sample taken.

Statistical Processing
Excel 2016 was used to organize the original data. SPSS 26 (International Business Machines Corporation America) statistical software was used for data processing and difference significance analysis, and the significance level was set as p < 0.05.

The Determination of the Optimal Mutagenesis Time and Screening Salinity
The time-lethal curve (a) of the ARTP tetraspores of G. lemaneiformis is shown in Fig. 1. It was shown that with the increase of the mutagenesis time, the mortality of the spores gradually increased. Formulas Y = 1.8227 X-3.3156 and R 2 = 0.9778 were obtained. The half-lethal action time of the ARTP mutagenesis was 29 s, and when the death rate of tetraspores was 80%, the action time of the ARTP mutagen was 46 s. In this experiment, in order to improve the efficiency of mutagenesis and to obtain more mutants, the lethality rate of 80% was adopted. Therefore, treatment time of 46 s was used in subsequent mutagenesis in the experiment.
The survival rate (b) of tetraspores decreased successively with the increasing of salinity. When the salinity was 60‰, the survival rate of tetraspores was about 4.10%. When the salinity was 58‰, the survival rate of tetraspores was 22.67%. In subsequent experiments, salinity of 58‰ as well as 60‰ was chosen as the osmotic pressure screening condition for the mutants.

Mutagenesis and Series Salinity Screening
Mutagenesis time of 46 s was used to treat spores with a scale of 2.7 × 10 5 , and then, artificial seawater with salinity of 58‰ was exploited for screening for 1 week. Obtained and transferred to normal seawater for culture were 8680 survived spores. The spores entered the vertical stage after 2 weeks in normal culture.
In order to further reduce the number of tetraspores, 60‰ salinity artificial seawater was used for the second hypertonic screening, and the screening time was 3 weeks. Seventeen dominant individuals were selected from the surviving individuals, and these dominant mutants were subsequently cultured in normal seawater.
The third hypertonic screening of the spores started after they had branched out with an artificial seawater treatment of 60‰ salinities for 3 weeks. A total of 9 hypertonic tolerant mutants were screened. After 2 months of cultivation, four mutants were selected which displayed better growth trends.

Confirmation of the High Osmotic Pressure Tolerance of Screen-Out Strains
As shown in Table 1, 4 screen-out strains were further validated for their tolerance to the high salinity treatment for 60‰ salinities within a treatment time of 21 days. In the control group, only about 3 algal tips remained in good condition showing bright red color and a complete body. About 15 algal tips of HAGL-X5 remained bright red. HAGL-X3 maintained a bright red state for 9 algal tips, followed by HAGL-X4 with 5, and finally HAGL-X2 with 3. Compared with the control group, the HAGL-X5 and HAGL-X3 displayed superior osmotic pressure resistance properties.

The Actual Photosynthetic Efficiency [Y(PSII)], Photochemical Quenching qL, and Non-photochemically quenched NPQ
The actual photosynthetic efficiency [Y(PSII)] and photochemical quenching qL of control as well as the mutants treated in artificial seawater with a salinity of 60‰ for 24 days are shown in Fig. 2a, b). On day zero, the Y(PSII) and qL values of all mutants were significantly higher than those of the control (p < 0.05), and those for HAGL-X3 and HAGL-X4 were significantly higher than those of HAGL-X2 and HAGL-X5 (p < 0.05). When entering the high salinity treatment (after the second day), the Y(PSII) and qL values of all strains decreased sharply to the lowest values and gradually recovered between the second and tenth days. Among them, the Y(PSII) of HAGL-X5 on the 2nd, 6th, 8th, and 10th days was significantly higher than that of control (p < 0.05), and the Y(PSII) of HAGL-X3 on the 2nd and 8th days were significantly higher than that of control (p < 0.05). The qL values of HAGL-X5 on the 2nd, 4th, 8th, and 10th days were significantly higher than control (p < 0.05), and the qL values of HAGL-X2, HAGL-X3, and  HAGL-X4 on the second day were all significantly higher than control (p < 0.05). After the tenth day, between the 12th and the 16th day, the Y(PSII) of all the strains fluctuated around 0.20, and they were all very close, with no significant difference (p > 0.05). From day 18 to day 24, Y(PSII) of all strains started to decline slowly. At this stage, only the control was significantly higher than that of HAGL-X4 on the 24th day (p < 0.05); there was no significant difference in Y(PSII) between control and other mutants at this stage (p > 0.05). From the 12th to the 22nd day, the qL of all the strains fluctuated around 0.65, and they were also very close, with no significant difference (p > 0.05). On day 24, the qL values of all strain began to decline. There was no significant difference in qL between control and HAGL-X2, HAGL-X3, and HAGL-X5 on the 24th day (p > 0.05).
The NPQ (Fig. 2c) of HAGL-X5 and HAGL-X3 were higher than those of the control before 16 days. The NPQ of HAGL-X3 at the 2nd and 8th days was significantly higher than that of the control. The NPQ of HAGL-X5 at the 16th day was significantly higher than that of the control (p < 0.05). After 16 days, the NPQ of the control and HAGL-X5 was close to each other (p > 0.05). The NPQ of HAGL-X2 was not significantly different from that of the control, while the NPQ of HAGL-X4 was even lower than that of the control as a whole (p < 0.05). From a trend point of view, the NPQ values of HAGL-X5 and HAGL-X3 were first reduced to the lowest value after being stressed, and then, the NPQ value quickly returned to the maximum value (on the sixth day). NPQ of control was also decreased. Then it rose again, returning to a higher level on the sixth day. Both HAGL-X5 and the control decreased slightly and rose to a higher point again at 12 days, and then slowly decreased. The overall trend was similar, but the value of HAGL-X5 was higher than the control overall. After HAGL-X3 rose to its highest point in the sixth day, it declined to a faster rate, but overall it was also higher than the control group. The Growth Rate Figure 3 shows the results of the growth rate measurement of each strain. The growth rate of all mutants was higher than that of the control group, which suggested that the selected mutants had obvious growth advantages. The maximum growth rate of mutant HAGL-X4 was 0.45 cm per week, followed by HAGL-X3 at 0.37 cm per week, followed by HAGL-X5 at 0.32 cm per week, and HAGL-X2 at 0.30 cm per week. The growth rate of HAGL-X3 and HAGL-X4 was significantly higher than that of the control group (p < 0.05), while those of HAGL-X2 and HAGL-X5 had no significant difference with control. Figure 4 shows the measurement results from the agar content of each strain. The highest agar content was evidenced in HAGL-X5 with 7.74%, followed by HAGL-X4 which was 6.01%. The agar content of HAGL-X5 and HAGL-X4 was about 1.5 and 1.2 times that of the control. The agar content of HAGL-X5 was also significantly higher than that of other mutants (p < 0.05), while those of HAGL-X2 and HAGL-X3 was even lower than that of the control. Therefore, HAGL-X5 is a mutant that meets the expectations of the screen.

Discussion
This experiment successfully screened out HAGL-X5 strains with high agar content via hypertonic screening. Researchers have conducted a lot of research on the factors that affect the content of G. lemaneiformis or Gracilaria agar. Cultivation salinity (Teo et al. 2009;Siow et al. 2012), dark treatment (Ekman et al. 1991;Freile-Pelegrín et al. 2002), nitrogen and phosphorus contents (Lewis and Hanisak 1996), cultivation season (Lee et al. 2016), etc. will have a relatively large impact on the agar content of G. lemaneiformis or Gracilaria. Among these many influencing factors, seawater salinity is easy to control in the laboratory, and the mutants can be reduced by hyperosmolarity screened during the breeding process to reduce the workload. Therefore, it is most feasible to use high osmotic pressure for breeding. It was reported that plants (Kumar et al. 2017;Munns et al. 2019), fungi, yeast (Mager and Marco 2002;Hohmann et al. 2007), cyanobacteria, and other microalgae (Bisson and Kirst 1995) will respond to a series of physiological and biochemical changes when they are subjected to high osmotic stress. When cells are under hypertonic stress, they will lose water quickly, which will cause cell shrinkage and cell membrane invagination (Park et al. 2016). The force's balance of cell wall will be changed due to cell contraction, which led to structure shift of the cell wall to enhance its mechanical strength to deal with hypertonic stress (Monniaux and Hay 2016). Secondly, in order to avoid further loss of water in the cells, which will affect the normal physiological and biochemical reactions, the cells will form a large number of osmotic regulators to adjust the osmotic pressure. These osmotic regulators mainly include mannitol, proline, betaine, trehalose, florideside, etc. Some of these osmotic regulators not only regulate cell osmotic pressure, but also play an important role in maintaining the normal conformation of proteins in cells (Singh et al. 2015). Through the above analysis, it can be seen that when cells are subjected to high osmotic pressure treatment, on the one hand, cells maintain the basic shape of cells by producing osmotic Fig. 3 The measurement results of the growth Fig.4 Measurement results of agar content in each strain pressure regulators, and provide a good ionic environment for normal metabolism. On the other hand, the cell walls of cells thicken in response to high osmotic pressure (Karandashova and Elanskaya 2005;Wang et al. 2019;Lv et al. 2021;Li et al. 2019). Therefore, when we use highsalinity artificial seawater for screening, it is highly possible to obtain strains of G. lemaneiformis physiological changes such as thickening of the cell wall and increase in cell content. At the same time, we carried out high osmotic pressure screening for different developmental stages of G. lemaneiformis to ensure the stability of the mutational traits.
Compared with the about 6.10% agar content of cultivar MT-18 (Chang et al. 2014), the agar content of HAGL-X5 was 7.74% which is applaudable. This experiment also provided a valuable reference to the subsequent breeding of G. lemaneiformis with high polysaccharide contains.
At present, ARTP mutagenesis has been successfully applied to the breeding of more than 40 kinds of microorganisms, including bacteria, actinomycetes, fungi, yeasts, and microalgae (Ottenheim et al. 2018). In algae breeding, it was mainly applied for microalga, such as Crypthecodinium cohnii and Chlorella pyrenoidosa where high polysaccharide and lipid contains were mainly targeted (Liu et al. 2015;Cao et al. 2017). In terms of power and optimal mutagenesis time, the former were 150 W, 70 s, and the latter were 120 W, 60 s, while those of G. lemaneiformis spores were 120 W, 46 s, which was not very different from those of microalgae. In terms of mutation rate, a total of 50,000 microalgae cells were mutagenized in the polysaccharide mutagenesis experiment, and 12 mutants that met the requirements were obtained, and the mutation rate was 0.024%. The mutation rate of lipid breeding was 57.4% (Liu et al. 2015;Cao et al. 2017). In this breeding, the total number of mutagenized spores was about 2.7 × 10 5 , and then, the artificial seawater of 58‰ was used for 1 week screening. The number of surviving individuals was 8680. Referring to hypertonic resistance, and the mutation rate was 3.21%. After the second hypertonic screening, there were about 6505 remaining individuals, from which 17 individuals with better growth were selected by observation method. Referring to growth trend, the selection rate was 0.26%. It can be found that the mutation/selection rate of ARTP either on macroalgae or on microalgae varies. However, the target mutants can be effectively obtained by ARTP mutagenesis.
In terms of chlorophyll fluorescence, it has been reported that algae photosynthesis is inhibited when they are in high salinity seawater (Masojídek et al. 2000). High osmotic pressure can change the thylakoid structure and block the electron transport chain of the photosynthetically active reaction center PSII, thereby reducing the maximum photosynthetic efficiency. At the same time, due to the ROS generated by hyperosmotic and high-salt stress, lipid peroxidation, DNA damage, and enzyme activity are impaired, and the phytochrome is decomposed (Demidchik 2015;Fatma et al. 2014;Yoshioka-Nishimura 2016). In this experiment, mutants and controls were treated with high salinity for 3 weeks. The actual photosynthetic efficiency [Y(PSII)], photochemically quenched (qL), and non-photochemically quenched (NPQ) of mutants and controls were determined. It can be found that most of the difference of Y(PSII) and qL among strains were observed in the first 10 days. The Y(PSII) and qL values of the mutants were significantly higher than those of the control without salt stress. Further analysis found that from the start point, Y (PSII) and qL of HAGL-X3 and HAGL-X4 were the highest and significantly higher than other strains, which indicated that the actual photosynthetic efficiency and photochemical quenching of HAGL-X3 and HAGL-X4 were higher than those of other strains. Therefore, the light energy conversion rate of these two strains were higher than others, which can explain why HAGL-X3 and HAGL-X4 were higher than other strains in the growth rate. From the second day to the tenth day, the Y(PSII) and qL values of HAGL-X3 and HAGL-X5 were higher than those of control, indicating that HAGL-X3, especially HAGL-X5, can more effectively protect the photosynthetic rental active reaction centers PSII and enhanced photochemical quenching under stress. The nonphotochemical quenching of (NPQ) indicates the strength of the photoprotective mechanism. On the start point, NPQ of HAGL-X3 and HAGL-X4 was significantly lower than other strains, which was just opposite to the status of qL. The NPQ value of HAGL-X5 was higher than control from the second day to the eighteenth day, though with insignificant difference (p > 0.05). The NPQ value of HAGL-X3 was higher than control from the second day to the fourteenth day, and the difference was not significant between the sixteenth day and the twenty-fourth day. The above results indicated that the photoprotection mechanism of the mutants (HAGL-X3 and HAGL-X5) was stronger than that of control in most days except for some day which were lower than control. The better photosynthetic parameters of the mutants (HAGL-X3 and HAGL-X5) than the control may be related to the better physiological state of the mutants.
In short, from the index of agar content, after screening with high osmotic pressure and determination of agar content, HAGL-X5 was considered to be the expected strain. If the effect of growth rate is also considered, HAGL-X4 is the most suitable strain in terms of total agar yield. In this experiment, individuals with high agar content were screened out by hypertonic screening, and a feasible breeding system was reported. The study contributed to breeding work of G. lemaneiformis, especially for the purpose of a strain selection with high agar contents.