Cell growth and carotenoid content of cultures of 13 chlorophycean microalgae are shown in Table 1. In all species lutein was the most abundant carotenoid. The highest lutein levels (24 mg L⁻¹) were found in C. sorokiniana, Monoraphidium braunii, Scenedesmus armatus and Scenedesmus vacuolatus (Table 1). Other carotenoids like antheraxanthin, α-carotene, β-carotene, violaxanthin and zeaxanthin were also produced in all the tested species albeit at concentrations lesser than lutein. Astaxanthin and canthaxanthin were present only in Chlorella fusca, Chlorella zofingiensis, Chlorococcum sp. and S. vacuolatus. With regard to growth, C. sorokiniana and S. armatus exhibited the highest specific growth rates of 0.11 and 0.09 h⁻¹, respectively, and for most species the maximum biomass value attained in the culture ranged between 7.5 to 8.5 g L⁻¹, except for Chlamydomonas reinhardtii, Chlorella luteoviridis and Chlorella stigmatofora, which showed lower biomass values.

Some of the species considered in this work, such as C. sorokiniana, M. braunii, S. armatus and S. vacuolatus are of potential practical interest on the basis of their high lutein level. Among these, C. sorokiniana has been selected for further work focused on the production of lutein, since this microalga also shows the highest specific growth rate (0.11 h⁻¹) and a high biomass value (8.0 g L⁻¹).

To obtain high lutein producing mutants of C. sorokiniana, cells were subjected to random chemical mutagenesis with MNNG and mutants were screened on the basis of their resistance either norflurazon or nicotine, growth rate and lutein content. First, the survival curve for mutagenesis with MNNG was performed to determine the MNNG concentration which resulted in around 5–10% of cells viability (Figure 2), and afterwards, a wide range of concentrations of the herbicides were tested to find out the minimal concentration which inhibited the grown of the wild strain, resulting in 400 μM for nicotine and 4 μM for norflurazon (data not shown).

According to the first step of selection, 745 herbicide-resistant colonies were obtained, from which only 222 (210 resistant to norflurazon and 12 resistant to nicotine) were selected according to the second step criterion, which consisted of discarding mutants exhibiting either low growth or unstable resistance to the herbicides. The selected mutants were grown in shaken liquid cultures under photoautotrophic standard conditions to determine lutein content and growth rate. Increases in volumetric and cellular lutein contents of the best mutants as compared to the wild type are shown in Table 7. The mutant MR-16, resistant to nicotine, exhibited volumetric and cellular lutein contents 2.0- and 1.4-fold higher, respectively, than those of the wild strain. This mutant also showed a 3.0-fold higher volumetric α-carotene and antheraxanthin levels and 2.0-fold higher β-carotene and zeaxanthin contents as compared with those of the wild type, being the total volumetric carotenoid content 2.0-fold higher than in the wild strain (data not shown). The mutants DMR-5 and DMR-8, both resistant to norflurazon, stood out in terms of cellular content in lutein, showing 53–55% increase relative to the wild type, reaching values of 7.0 mg lutein g⁻¹ DW.

Moreover, the growth of the selected high lutein producing mutants under photoautotrophic conditions was similar or even higher than that of the parental strain, as shown in the Figure 3 for MR-16.

All mutants of C. sorokiniana obtained in the processes of mutagenesis were analyzed for their viability and stability by consecutive sub-cultures in nonselective and selective media with herbicides to check their resistance.

MR-16 mutant exhibited both a growth and lutein content higher than the wild strain under photoautotrophic conditions (Figure 3, Table 7). For this reason, cells of the wild type and the MR-16 mutant were grown under the best photoautotrophic and mixotrophic conditions found previously in the wild strain of C. sorokiniana for lutein accumulation in order to know accurately the improvement of the MR-16 mutant as compared to the wild type. As shown in Table 8, the volumetric and cellular lutein contents increased both by 68% in the mutant MR-16 with respect to the wild strain when the mutant was grown photoautotrophically, attaining maximum values of 42.0 mg L⁻¹ and 5.0 mg g⁻¹ DW at the end of the deceleration phase. On the other hand, specific growth rate was either constant or very similar at the different conditions assayed in both strains. However when the mutant was grown mixotrophically, biomass decreased 40% as compared to the wild strain. The addition of acetate and glucose to cells of the MR-16 mutant had a negative effect decreasing a 48% the maximum lutein content in the culture, and a 33% the cellular content with regard to the parental strain. A difference between the wild and the mutant C. sorokiniana strains was that the wild type showed the maximum values of cellular lutein during the early deceleration phase, whereas MR-16 mutant attained these contents at the end of the deceleration phase, due to a continuous and higher lutein accumulation.

Since irradiance, temperature, NaCl and nitrate concentration are known to affect the levels of different carotenoids in microalgae, the effect of these factors on cell growth and lutein accumulation was analyzed in C. sorokiniana which has been selected among thirteen chlorophycean microalgae, since it showed the highest growth rate and lutein accumulation. In this selected microalga lutein content in the culture was optimal at moderate irradiances, since both biomass and cellular content in lutein decreased at both high and low irradiance (Table 2). A similar trend with regard to the cellular content has also been observed for lutein and β-carotene in different species of C. zofingiensis and Muriellopsis sp. [26,16,18]. In contrast, in H. pluvialis and C. zofingiensis the cellular content of secondary carotenoids, such as astaxanthin and canthaxanthin, followed the opposite trend to lutein, being enhanced at high irradiance [18,27]. C. sorokiniana exhibited maximal volumetric and cellular lutein content in the range of 28 to 32 °C decreasing at lower and higher temperatures (Table 3); these results being in agreement with those obtained in the lutein-producing microalga C. zofingiensis [16] and Scenedesmus almeriensis [20,28]. Contrastingly, extreme temperatures triggered carotenogenesis in H. pluvialis. It has been suggested that endogenously generated active oxygen is responsible for the stimulation of astaxanthin synthesis at high temperature in this microalga [24]. Nitrogen limitation enhances the synthesis of secondary carotenoids such as astaxanthin, but reduces biomass yield. However, nitrogen at none limiting concentrations is required for primary carotenoids accumulation due possibly to the need of a continued synthesis of light-harvesting proteins and structural xanthophylls under optimal growth conditions [3,16,18]. This agrees with the results obtained in C. sorokiniana since both volumetric and cellular lutein content increased when nitrate concentration in the culture medium was raised from 10 to 40 mM, decreasing slightly at higher nitrate concentrations (Table 4). On the other hand, although in heterotrophic cultures of C. protothecoides maximal lutein productivities were achieved using urea as the nitrogen source [29], in C. sorokiniana grown mixotrophically either a slight decrease or no effect in growth and lutein content was registered when sodium nitrate was replaced by either urea (Table 6) or ammonium nitrate (data not shown). NaCl stress, which has been described to induce the biosynthesis of secondary carotenoids [30], seems not to trigger the biosynthesis of the primary ones. Thus in C. sorokiniana, both growth and lutein content decreased drastically when NaCl concentrations were increased (data not shown), and in the cases of other chlorophyta as Muriellopsis sp., C. zofingiensis and S. almeriensis, the lutein levels per cell remained practically constant at the different NaCl concentrations assayed [16,18,20]. Therefore, although stress factors, such as high irradiance, extreme temperatures, high NaCl concentration or nutrients limitation enhance the cellular accumulation of secondary carotenoids, such as astaxanthin [1,18,31], these factors do not increase the cellular levels of lutein in C. sorokiniana, since lutein is a primary carotenoid, being required for the structure and function of the light-harvesting complexes in photosynthesis [2], and accordingly, conditions that increase photoautotrophic growth of this microalga are also those enhancing lutein accumulation.

In photoautotrophic mass cultures of microalgae for the production of biomass and valuable compounds, the main limiting factor is usually light availability, which in many cases limits cell density and productivity of the cultures, making it unprofitable for industry. An alternative to overcome this problem is to use mixotrophic cultures. In C. sorokiniana, the addition of acetate to the cultures enhanced both growth and volumetric and cellular lutein contents (Table 5). Although it is known that acetate enhances growth and synthesis of astaxanthin in H. pluvialis [32], the effect of this carbon source on the biosynthesis of lutein has not been studied in other microalgae. Even though the addition of glucose decreased the cellular content in lutein, it increased biomass considerably; therefore, the volumetric lutein content was enhanced and the highest volumetric lutein levels (35 mg L⁻¹) were achieved in cultures supplemented with this carbon source (Table 6). Therefore the supply of an extra carbon source to cultures supported higher growth and productivity, overcoming the limitation by light. Recently, it has been shown that in heterotrophically grown C. zofingiensis, glucose increased the cellular accumulation of astaxanthin and zeaxanthin by increasing the transcription levels of both β-carotene ketolase and β-carotene hydroxylase involved in the synthesis of these carotenoids, but decreased the cellular content of lutein and chlorophyll [33,34] and in C. protothecoides lutein productivity was much higher in heterotrophic cultures supplemented with glucose than in photoautotrophic cultures, since in the former very high cell densities were achieved [35]. Therefore, our results are in agreement with these findings, which indicate that although glucose decreases the celular content of lutein, it increases considerably the biomass, therefore resulting in a higher lutein production in mixotrophic cultures as compared to those performed photoautotrophically.

There are few reports concerning the isolation of high carotenoid yielding mutants of microalgae by random mutagenesis. A mutant of Chlorella regularis showing a high cellular content of lutein has been isolated by Ishikawa et al. (2004) [36], although it exhibited lower growth than the wild strain and no volumetric lutein content data have been reported. Some mutants with enhanced accumulation of astaxanthin have also been described, however their growth rates are usually lower than that of the wild type [37,23,38]. In our experiments with C. sorokiniana, mutants were selected not only on the basis of high lutein content, but also according to a high growth rate. Thus, the mutant MR-16 exhibited a growth and cellular lutein content higher than those of the wild type and, as a consequence, lutein yields 2.0-fold higher than those measured in the wild strain were achieved (Table 7 and Figure 3). In addition, this mutant shows the highest lutein content described in the literature under photoautotrophic conditions of growth at laboratory scale [14,16,18–20], attaining a value of 42.0 mg L⁻¹ at the end of the deceleration phase (Table 8). The mutant MR-16 was resistant to nicotine, a specific inhibitor for the enzyme lycopene β-cyclase, involved in lutein biosynthesis, which could possibly have an altered specific activity in this mutant, for instance a modified structure for the herbicide-binding site, and/or a different expression of this enzyme, which means a higher enzyme activity, and would result into an improved lutein production under photoautotrophic growth. In addition, the higher growth rate of this mutant with respect to the wild type strain under photoautotrophic conditions could be due to the higher cellular lutein content, since this carotenoid plays important roles in the function, structure and photo-protection of the photosynthetic apparatus, leading to a higher photosynthetic efficiency. It is relevant to mention that the addition of acetate and/or glucose to the culture medium decreased the biomass in the cultures of MR-16. The mechanism of this effect is not understood yet. It is possible that the mutation has provoked a metabolic alteration causing a weak repression of growth by glucose or acetate.

Although microalgae are not used yet as a lutein source at industrial scale, first steps have been made at laboratory and pilot scales in the recent years [16,14,28,21]. The data here reported point to C. sorokiniana as an attractive candidate for the production of lutein, since it shows a high growth rate (0.12 h⁻¹), a volumetric lutein content of 35 mg L⁻¹ and cellular lutein contents of 5.2 mg g⁻¹ DW, which are enhanced by random mutagenesis up to 42.0 mg L⁻¹ and 7.0 mg g⁻¹ DW. These values are similar or higher than those reported for other lutein producing microalgae grown at laboratory scale under photoautotrophic conditions. Thus, C. zofingiensis shows a specific growth rate of 0.04 h⁻¹ and a volumetric and cellular lutein content of 20.0 mg L⁻¹ and 3.0 mg g⁻¹ DW, respectively [18,14]; Scenedesmus almeriensis has been reported to accumulate 5.5 mg g⁻¹ DW and a specific growth rate of 0.07 h⁻¹ [20]; Muriellopsis exhibited a growth rate of 0.17 h⁻¹ and a volumetric and cellular lutein content of 29.0 mg L⁻¹ and 5.5 mg g⁻¹ DW, respectively [16]; and Coccomyxa acidophila has been reported to accumulate 6.1 mg g⁻¹ DW under extreme culture conditions [19]. From all this information, we can propose C. sorokiniana as a promising microalgal species for the production of lutein for commercial applications.

The species of microalgae used in this work were: Chlorella fusca 211-8b, Chlorella sorokiniana 211-32, Chlorella zofingiensis 211-14, Monoraphidium braunii 202-7d, Scenedesmus vacuolatus 211-15 and Chlorella stigmatofora 9–86 from SAG, Culture Collection of Göttingen University (Germany); Chlorella luteoviridis 258, Chlorococcum sp. 2438, Scenedesmus armatus 2533, Scenedesmus quadricauda 76 and Scenedesmus obliquus 393 from UTEX, Culture Collection of Algae of the University of Texas (USA); Chlorella vulgaris 101 from UAM, Culture Collection of Microalgae of Universidad Autónoma de Madrid (Spain); and Chlamydomonas reinhardtii CC621(−) from Culture Collection of Institut für Biologie III at the University of Freigburg (Germany).

Cells of C. sorokiniana in the exponential phase of growth (10⁶ cells mL⁻¹) were harvested by centrifugation (2700× g, 10 min), washed with sterile water and treated with 0.1 mg mL⁻¹ of 1-methyl-3-nitro-1-nitrosoguanidine (MNNG) (survival rate 5–10%) for 1 h. This mutagen is known to induce nucleotides substitutions at high frequencies and little lethality, and the inhibitors nicotine and norflurazon were used as the selection method. These herbicides inhibit the carotenogenic enzymes lycopene β-cyclase and phytoene desaturase, respectively. Therefore, the screening of the mutants was performed by their resistance to the carotenoid biosynthesis inhibitors nicotine and norflurazon.

The treated cells were washed with sterile water, resuspended in Arnon modified medium and incubated under dim light during 24 h. After the incubation, cells were spread on solid modified Arnon medium containing, either 4 μM norflurazon (Supelco, Bellefonte, PA, USA) or 400 μM nicotine (Sigma-Aldrich, Steinheim, Germany) and incubated at 25 °C and 50 μmol photons m⁻² s⁻¹ for 3–4 weeks. Then, the herbicide-resistant colonies were sub-cultivated several times in solid medium containing either norflurazon or nicotine to check their resistance to the herbicide and growth in solid medium. Herbicide-resistant mutants that showed good growth were grown in 100 mL-capacity erlenmeyers under photoautotrophic conditions, shaken at 100 rpm and illuminated from the top at 50 μmol photons m⁻² s⁻¹, in order to analyze carotenoids content and growth.

For dry weight (DW) determinations, 5 mL aliquots of the cell culture were filtered through pre-dried Whatman GF/C paper (Whatman International Ltd., Maidstone, England), washed three times with distilled water, and the filters containing the algae were dried at 80 °C for 24 h.

Specific growth rate (μ) was calculated from the measured DW during the exponential phase of growth, using the equation: μ = (ln x₂ − ln x₁)/(t₂ − t₁), where x₂ and x₁ represent DW values in terms of g L⁻¹ at times t₂ and t₁, respectively.

For carotenoid analysis, pigments were extracted with methanol at 70 °C, centrifuged, the supernatant evaporated under N₂ and the pellet dissolved in methanol. Then the samples were centrifuged and analyzed by HPLC using a Waters Spherisorb ODS2 column (4.6 × 250 mm, 5 μm particle size) (Waters, Mildford, MA, USA). The chromatographic method described by Cordero et al. [40] was used. The pigments were eluted at a flow rate of 1.2 mL min⁻¹ and detected at 440 nm using a Waters 2996 photodiode-array detector. Identification of carotenoids was achieved by comparison of the individual characteristic absorption spectrum and the retention time with known standards. Quantification was performed using a calibration curve generated with commercially available carotenoids standards from Sigma-Aldrich (St. Louis, MO, USA) and DHI (Holsholm, Germany).

Samples analyzed were withdrawn along the curve of growth. Independent triplicate analyses were carried out for each sample, the results representing the mean values. The standard deviation (SD) is omitted since it was lower than 10% of the mean values.