Physicochemical Changes of Air-Dried and Salt-Processed Ulva rigida over Storage Time

The impact of air-drying at 25 °C, brining at 25%, and dry-salting (at 28% and 40%) on the quality and nutritional parameters of Ulva rigida were evaluated over six months of storage. Overall, the main changes occurred in physical aspects during storage time, with U. rigida intensifying its yellow/browning tones, which were more evident in salt-treated samples. The force necessary to fracture the seaweed also increased under all the preservative conditions in the first month. Conversely, the nutritional parameters of U. rigida remained stable during the 180 days of storage. All processed samples showed a high content of insoluble and soluble fibers, overall accounting for 55%–57% dw, and of proteins (17.5%–19.2% dw), together with significant amounts of Fe (86–92 mg/kg dw). The total fatty acids pool only accounted for 3.9%–4.3% dw, but it was rich in unsaturated fatty acids (44%–49% total fatty acids), namely palmitoleic (C16:1), oleic (C18:1), linoleic (C18:2), linolenic (C18:3), and stearidonic (18:4) acids, with an overall omega 6/omega 3 ratio below 0.6, a fact that highlights their potential health-promoting properties.


Introduction
Seaweeds, i.e., marine macroalgae (including Chlorophyta, Rhodophyta, and Ochrophyta/ Phaeophyceae), are considered one of the non-animal foods of the future due to their ability to grow without using arable land or freshwater resources, combined with their recognized richness in valuable nutrients and phytochemicals, including proteins with high nutritional value, bioactive peptides, insoluble and soluble fibers, polyunsaturated fatty acids, minerals, vitamins, and polyphenols [1][2][3][4][5].
The direct consumption of macroalgae as food is still incipient in Western countries when compared to the Asiatic countries, but this trend is changing over the past years, mostly based on health claims associated with their regular consumption [6]. Indeed, food and nutraceutical industries have grown interest in introducing macroalgae as an ingredient in functional foods, and the number of products containing this "new ingredient" launched on the market is growing fast, particularly in Europe [7]. The global functional food market, evaluated at about $168 billion in 2013 and estimated to reach $305.4 billion by 2020 [8], is one of the market opportunities for the direct application of seaweeds, of purified extracts, or purified fractions.
However, the use of characteristic European macroalgae as a food ingredient faces huge challenges, that go from the sustainable production of biomass to hold the market development without disrupting marine resources and many others, directly or indirectly linked to it. Among direct implications to high biomass production, fast and controlled preservation methods will be required. In fact, t0, t30, t60, t120, t180 correspond to day 0 (i.e., just after the application of the processing treatment), and after storage for 30, 60, 120, and 180 days, respectively. Storage was done at room temperature (air-dried) or 4 • C (salt-processed).
Values are presented as mean ± standard deviation, n = 3. Different letters in the line indicate significant differences (p < 0.05) according to Tukey's test.
The gathered data also allow us to conclude that the mean moisture value of dehydrated U. rigida did not vary significantly over 180 days, indicating that the storage conditions were adequate to maintain this parameter stable in the air-dried samples. A similar trend was also observed for the brined and dried-salted samples at 40%, suggesting that under these conditions, the osmotic equilibrium was reached in the first 4 h (time interval between collection and analysis of the sample after arrival at the laboratory), though they were maintained at low temperatures. In turn, osmotic balance in dry-salting at 28% samples was achieved later on, since values at t0 were higher than those measured at t30 (70.9% and 61.9%, respectively).

Superficial Color
Surface color is a quality attribute of food that is commonly affected by processing and storage [16]. Thermal processing, in particular, can severely alter surface color due to chemical and enzymatic degradation of pigments [16]. In turn, some of these degradations may be minimized in salt processing samples if kept under low temperatures, but conversely, in these treatments, flow of pigments from the food matrix along with the water might occur [9].
The color coordinates (CIE L*a*b*) of air-dried and salt-processed U. rigida over six months of storage is summarized in Table 2. Note that, in this system, the results are expressed as negative or positive values in relation to a particular color coordinate: a* represents greenish and reddish colors in case of negative and positive values, respectively, while negative and positive b* values are bluish and yellowish tonalities, respectively. The L* coordinate measures the luminosity as an approximation to a greyscale, ranging between black (0) and white (100) [17] and the browning index (BI, estimated by considering a*, b*, and L* parameters) is defined as brown color purity [18]. BI-browning index; ∆E-total color difference; t0, t30, t60, t120, t180 correspond to day 0 (i.e., just after the application of the processing treatment), and after storage for 30, 60, 120, and 180 days, respectively. Storage was done at room temperature (air-dried) or 4 • C (salt-processed). Values are presented as mean ± standard deviation, n = 3 (except for ∆E, which corresponds to the mean value). Different letters in the line indicate significant differences (p < 0.05) according to Tukey´s test.
The greenish tonality of U. rigida, just after the application of the treatments and rehydration, was not significantly different among the samples (as reflected by a* values close to −15), regardless, it tended to be less intense in the air-dried ones. In turn, its fade during storage was more pronounced in salt-processed samples. While the mean a* value in dried Ulva was slightly lowered in the first 30 days, from −14.78 ± 0.35 to −13.08 ± 0.95, and kept constant for up to six months of storage, those of salt-processed algae continued to decrease until 120 days (brine and salted 28%) or 180 days (salted 40%), reaching values of −10 to −11. The superior impact observed in salt-treated samples is probably partly due to some flow of chlorophylls along with the water and to their degradation, which results in the gray-brown compounds pheophytin or pheophorbide. Degradation of chlorophylls in salt-processed Ulva might be favored by their high water content (57%-71%) in comparison to that of air-dried samples (14%-15%), regardless if they were kept under lower temperatures (4 • C).
As for a*, the b* coordinate in air-dried U. rigida was slightly distinct from those in salted-processed samples (values of 42 vs. 39, respectively), suggesting that air-drying could intensify the yellow tonalities of this macroalgae. As well, changes in the b* coordinate during storage were evident in all samples, with ∆ (i.e., variation between t0 and t180) of 5.7, 6.5, and 8-9 in air-dried, salted at 28%, brined and salted at 40%, respectively, overall indicating a clear intensification of yellow tonalities of U. rigida in this period. As for other vegetables, this color change is expected to be associated with some changes/degradation of carotenoids, as also a reflection of those occurring in chlorophylls [16]. In fact, the green color of U. rigida is mainly due to the presence of high quantities of chlorophylls that mitigate the yellow color of carotenoids, whereas chlorophyll degradation intensifies their yellow coloration [19].
In general, browning coloration results from both enzymatic and non-enzymatic oxidation of phenolic compounds or non-enzymatic Maillard reactions between reducing sugars and amino acids, the latter being particularly promoted at high temperature [16]. Once cell walls and cellular membranes lose their integrity (a fact that might occur because of the water loss during the treatments), enzymatic oxidation proceeds much more rapidly. As can be concluded from Table 2, our results indicated that at t0, the brown tone of air-dried samples was more intense in those treated with salt, a fact that probably results from non-and enzymatic reactions occurring during the drying processing, as previously mentioned. Notably, after air-drying, the browning tone of Ulva was maintained constant for at least 30 days. In fact, ∆BI in air-dried samples occurred between t30 and t60 and later from t120 to t180. Conversely, significant changes in BI of salt-processed samples (kept at 4 • C) were visible in the two first months and these seemed to be delayed at a lower water content (salted at 40% in comparison to brine/salted at 28%). This also supports the hypothesis that BI changes during algae storage are dictated by enzymatic browning phenomena.
Overall, it is clear that just after the air-drying process at 25 • C, if rehydrated, U. rigida is visually more yellow/brown than those treated with salt. Moreover, as expected, the total color difference parameter (calculated on the basis of a*, b*, and L* coordinates) confirmed that changes in colors during the storage time were particularly evident in salt-treated samples, probably due to superior changes in the cells' structures, which might contribute to additional losses/changes in pigments. In fact, this is partially supported by the results of Figure 1, which show higher levels of extracted chlorophylls and total carotenoids (presumably lutein as shown in our previous work [15]) in air-dried U. rigida at t180, when compared to salt-treated samples.
Molecules 2019, 24, x FOR PEER REVIEW 4 of 12 partly due to some flow of chlorophylls along with the water and to their degradation, which results in the gray-brown compounds pheophytin or pheophorbide. Degradation of chlorophylls in saltprocessed Ulva might be favored by their high water content (57%-71%) in comparison to that of airdried samples (14%-15%), regardless if they were kept under lower temperatures (4 °C).
As for a*, the b* coordinate in air-dried U. rigida was slightly distinct from those in saltedprocessed samples (values of 42 vs. 39, respectively), suggesting that air-drying could intensify the yellow tonalities of this macroalgae. As well, changes in the b* coordinate during storage were evident in all samples, with ∆ (i.e., variation between t0 and t180) of 5.7, 6.5, and 8-9 in air-dried, salted at 28%, brined and salted at 40%, respectively, overall indicating a clear intensification of yellow tonalities of U. rigida in this period. As for other vegetables, this color change is expected to be associated with some changes/degradation of carotenoids, as also a reflection of those occurring in chlorophylls [16]. In fact, the green color of U. rigida is mainly due to the presence of high quantities of chlorophylls that mitigate the yellow color of carotenoids, whereas chlorophyll degradation intensifies their yellow coloration [19].
In general, browning coloration results from both enzymatic and non-enzymatic oxidation of phenolic compounds or non-enzymatic Maillard reactions between reducing sugars and amino acids, the latter being particularly promoted at high temperature [16]. Once cell walls and cellular membranes lose their integrity (a fact that might occur because of the water loss during the treatments), enzymatic oxidation proceeds much more rapidly. As can be concluded from Table 2, our results indicated that at t0, the brown tone of air-dried samples was more intense in those treated with salt, a fact that probably results from non-and enzymatic reactions occurring during the drying processing, as previously mentioned. Notably, after air-drying, the browning tone of Ulva was maintained constant for at least 30 days. In fact, ΔBI in air-dried samples occurred between t30 and t60 and later from t120 to t180. Conversely, significant changes in BI of salt-processed samples (kept at 4 °C) were visible in the two first months and these seemed to be delayed at a lower water content (salted at 40% in comparison to brine/salted at 28%). This also supports the hypothesis that BI changes during algae storage are dictated by enzymatic browning phenomena.
Overall, it is clear that just after the air-drying process at 25 °C, if rehydrated, U. rigida is visually more yellow/brown than those treated with salt. Moreover, as expected, the total color difference parameter (calculated on the basis of a*, b*, and L* coordinates) confirmed that changes in colors during the storage time were particularly evident in salt-treated samples, probably due to superior changes in the cells' structures, which might contribute to additional losses/changes in pigments. In fact, this is partially supported by the results of Figure 1, which show higher levels of extracted chlorophylls and total carotenoids (presumably lutein as shown in our previous work [15]) in airdried U. rigida at t180, when compared to salt-treated samples. Figure 1. Levels of chlorophyll a, b and total carotenoids extracted from Ulva rigida submitted to different preservative processes (air-drying, brining, salting at 28% and 40%) after six months of storage. The macroalgae were stored at room temperature (air-dried) or at 4 °C (salt-processed). The results correspond to mean ± standard deviation (n = 3).  Figure 1. Levels of chlorophyll a, b and total carotenoids extracted from Ulva rigida submitted to different preservative processes (air-drying, brining, salting at 28% and 40%) after six months of storage. The macroalgae were stored at room temperature (air-dried) or at 4 • C (salt-processed). The results correspond to mean ± standard deviation (n = 3).

Mechanical Properties
Texture is another physical parameter that is frequently affected by industrial processing, and in the case of U. rigida, a change in blade fracturability may occur [20,21]. As depicted in Figure 2, the force required to fracture U. rigida blades varied among the samples, but changes were also relevant during storage. At t0, air-dried U. rigida presented the highest resistance to fracturing, in comparison to brined and salted at 40% (force values of 2.5 N vs. 1.3-1.6 N) or even those salted at 28% (although differences with the latter were less evident). Overall, the results suggest that for short storage periods, salt-processed U. rigida had less fracturability as compared to air-dried samples. Besides, these are also in line with those reported by Prinzivalli et al. [22], who showed that osmotic dehydration or salting decreases the amount of force necessary to perforate vegetable samples.

Mechanical Properties
Texture is another physical parameter that is frequently affected by industrial processing, and in the case of U. rigida, a change in blade fracturability may occur [20,21]. As depicted in Figure 2, the force required to fracture U. rigida blades varied among the samples, but changes were also relevant during storage. At t0, air-dried U. rigida presented the highest resistance to fracturing, in comparison to brined and salted at 40% (force values of 2.5 N vs. 1.3-1.6 N) or even those salted at 28% (although differences with the latter were less evident). Overall, the results suggest that for short storage periods, salt-processed U. rigida had less fracturability as compared to air-dried samples. Besides, these are also in line with those reported by Prinzivalli et al. [22], who showed that osmotic dehydration or salting decreases the amount of force necessary to perforate vegetable samples. Figure 2. Force (N) required to fracture Ulva rigida submitted to different preservative processes (airdrying, brining, salting at 28% and 40%), over six months of storage; t0, t30, t60, t120, t180 correspond to day 0 (i.e., just after the application of the processing treatment), and after storage for 30, 60, 120, and 180 days, respectively. Storage was done at room temperature (air-dried) or at 4 °C (saltprocessed). The results correspond to mean ± standard deviation (n > 5). Different letters in a treatment condition indicate significant differences (p < 0.05) according to Tukey´s test.
Over the 180 days of storage, all U. rigida showed a tendency to increase their resistance to fracture, which on average reached values close to 3-4 N. Curiously, this increment occurred in a shorter period for air-dried and brined macroalgae when compared to salted samples (1 vs. 2 months). Also note that it is possible that the increment of fracturability of U. rigida is partially related to their conservation at low temperatures, since cold is known to contribute to firmness of foods [23].

Nutritional Parameters
Ulva sp. are recognized for their richness in fibers (mostly ulvans) and minerals, among which Fe is of most importance, as accumulation is assumed to be much superior than in Rhodophyta and Phaeophyta, reaching values of 6 g/kg dw [24,25]. Alike other green species, they are also a good source of proteins (10%-25% dw), containing considerable levels of essential amino acids [26]. Moreover, despite lipids may in general only represent up to about 5% of the whole algal dry weight, they display an important nutritional value, with emphasis in n-3 polyunsaturated fatty acids (PUFAs) like α-linolenic acid, eicosapentaenoic acid, and docosahexaenoic acid [27,28]. However, as for other natural products, the overall nutritional properties of Ulva sp. are dependent on factors which, among others, include specific species, seasonality, conditions, processing, and storage conditions [29,30].
In order to evaluate possible differences on the nutritional value of U. rigida submitted to distinct processing over storage time, levels of protein, fiber, iron, and of fatty acids (FA) were evaluated just after application of the treatment (day 0, t0) and at the end of six months of storage (t180). In this context, please note that regardless of the nitrogen-to-protein conversion factor of 6.25 being the most  Figure 2. Force (N) required to fracture Ulva rigida submitted to different preservative processes (air-drying, brining, salting at 28% and 40%), over six months of storage; t0, t30, t60, t120, t180 correspond to day 0 (i.e., just after the application of the processing treatment), and after storage for 30, 60, 120, and 180 days, respectively. Storage was done at room temperature (air-dried) or at 4 • C (salt-processed). The results correspond to mean ± standard deviation (n > 5). Different letters in a treatment condition indicate significant differences (p < 0.05) according to Tukey´s test.
Over the 180 days of storage, all U. rigida showed a tendency to increase their resistance to fracture, which on average reached values close to 3-4 N. Curiously, this increment occurred in a shorter period for air-dried and brined macroalgae when compared to salted samples (1 vs. 2 months). Also note that it is possible that the increment of fracturability of U. rigida is partially related to their conservation at low temperatures, since cold is known to contribute to firmness of foods [23].

Nutritional Parameters
Ulva sp. are recognized for their richness in fibers (mostly ulvans) and minerals, among which Fe is of most importance, as accumulation is assumed to be much superior than in Rhodophyta and Phaeophyta, reaching values of 6 g/kg dw [24,25]. Alike other green species, they are also a good source of proteins (10%-25% dw), containing considerable levels of essential amino acids [26]. Moreover, despite lipids may in general only represent up to about 5% of the whole algal dry weight, they display an important nutritional value, with emphasis in n-3 polyunsaturated fatty acids (PUFAs) like α-linolenic acid, eicosapentaenoic acid, and docosahexaenoic acid [27,28]. However, as for other natural products, the overall nutritional properties of Ulva sp. are dependent on factors which, among others, include specific species, seasonality, conditions, processing, and storage conditions [29,30].
In order to evaluate possible differences on the nutritional value of U. rigida submitted to distinct processing over storage time, levels of protein, fiber, iron, and of fatty acids (FA) were evaluated just after application of the treatment (day 0, t0) and at the end of six months of storage (t180). In this context, please note that regardless of the nitrogen-to-protein conversion factor of 6.25 being the most commonly used, it overestimates the protein content in seaweeds [31] and because of that, a conversion factor of 5 is more accurate [32].
At t0, all processed samples showed a high content of insoluble and soluble fibers, overall accounting for 55%-57% dw and of proteins (17.5%-19.2% dw), as shown in Figure 3. As well, the amounts of Fe and total FA were close among the distinct samples (86-92 mg/100 g dw and 3.9%-4.3% dw, respectively). Moreover, the gathered results regarding these parameters, at t180, remained close to those at day 0, thus suggesting that nutritional value of air-dried and salt-processed U. rigida are not significantly changed over six months (if kept under the herein applied conditions), yet one must note that levels might not reflect specific changes in the nutrients. In this regard, among fibers, proteins, and fatty acids, the latter are the most prone to oxidation processes and were further analyzed as indicators of possible changes due to processing and storage.
commonly used, it overestimates the protein content in seaweeds [31] and because of that, a conversion factor of 5 is more accurate [32].
At t0, all processed samples showed a high content of insoluble and soluble fibers, overall accounting for 55%-57% dw and of proteins (17.5%-19.2% dw), as shown in Figure 3. As well, the amounts of Fe and total FA were close among the distinct samples (86-92 mg/100g dw and 3.9%-4.3% dw, respectively). Moreover, the gathered results regarding these parameters, at t180, remained close to those at day 0, thus suggesting that nutritional value of air-dried and salt-processed U. rigida are not significantly changed over six months (if kept under the herein applied conditions), yet one must note that levels might not reflect specific changes in the nutrients. In this regard, among fibers, proteins, and fatty acids, the latter are the most prone to oxidation processes and were further analyzed as indicators of possible changes due to processing and storage. Figure 3. Content of fiber (A), iron (B), protein, as determined by N x correction factor of 5; (C) and total fatty acids (D) of Ulva rigida submitted to different preservative processes (air-drying, brining, salting at 28% and 40%), just after the application of treatment (t0, full representations) and after six months of storage (t180, line representations) at room temperature (air-dried) or 4 °C (salt-processed). The results correspond to mean ± standard deviation (n = 3).
Despite the occurrence of slight variations in specific fatty acids amongst the processed samples, they showed an FA composition mainly rich in palmitic (C16:0), palmitoleic (C16:1 n-7), oleic (C18:1 n-9), linoleic (C18:2), linolenic (C18:3), stearidonic (18:4), and behenic (C22:0) acids, with trace amounts of the omega-3 fatty acids eicosapentaenoic (C20:5) and eicosatetraenoic (C20:4), and an overall low omega-6/omega-3 ratio (0.38-0.56) and unsaturated fatty acid (UFA)/saturated fatty acid (SFA) ratio (0.79-0.94), as shown in Table 3. In general, this profile is in line with reported data for U. rigida, despite some differences on the total amount of lipids and/or fatty acids, as well as on the relative abundance of specific FA, a fact that is attributed to the impact of multiple factors (growth conditions, seasonal effects, processing, and others) on the chemical composition of algae, as well as to the distinct analytical methods applied [31][32][33]. Notably, in general, the relative abundance of U. rigida FA at t180 was not significantly different from that at t0, thus indicating that FA were kept stable during the storage period of six months both in dried and salt-processed samples. The results correspond to mean ± standard deviation (n = 3).

Sample Collection and Treatments
U. rigida was produced by ALGAplus Lda. (production site located at Ria de Aveiro coastal lagoon, Northern Portugal, 40 • 36 43" N, 8 • 40 43" W), in an open land-based integrated multi-trophic aquaculture (IMTA) system. After hand collection in November 2016, the macroalgae batch was treated according to internal procedures of ALGAplus. Samples were washed with filtered and sterilized (UV and ozone) seawater from Ria de Aveiro, followed by centrifugation to remove excess water. A portion of this batch was dried at 25 • C for 16 h in an industrial convective dryer (customized built by ALGAplus Lda.) and then stored in multilayer paper-plastic bags and kept at the ALGAplus facility in a non-climatized room until analysis, while the remaining parts of the batch were processed by salt-treatments. For brining, 4 kg of U. rigida was submerged in a 25% (w/v) solution of kitchen salt for 5 min, while dry-salting consisted of mixing U. rigida with kitchen salt at 28% (w/w) or at 40% (w/w). All the salt-processed samples were also stored at ALGAplus, in covered Styrofoam boxes at 4 • C until analysis.
The effect of air-drying or salt-processing on moisture content, superficial color, and mechanical properties of U. rigida were assessed at five distinct points during the storage period, namely at day 0, 30, 60, 120, and 180 (t0, t30, t60, t120, and t180, respectively). The nutritional parameters were compared at t0 and t180. After reception at the laboratory, salt was manually removed (if salt-processed). Samples were rehydrated in distilled water for 15 min and then evaluated for color and fracturability. For the nutritional parameters and pigments extraction, rehydrated samples were frozen, freeze-dried, ground (Yellowline A10 mill, 20,000 rpm, IKA, Works Inc., Wilmington, NC, USA) and sieved with <0.25 mm pore sieve.

Texture
The fracturing of the samples was evaluated using a texturometer TA-HDi (Stable Micro Systems), with a 5 kg cell and a 6 mm piercing probe of stainless steel. Ulva rigida was soaked in water for 15 min and placed on the analysis platform. The peak of the strength needed to pierce the samples was registered by the software Texture Expert Exceed 2.64 and expressed in Newtons (N).

Contents of chlorophylls and carotenoids
Chlorophylls and carotenoids were extracted with acetone with 1% of butylated hydroxytoluene (BHT) for 24 h, using powdered samples and a mass/volume ratio of 1:100. The extraction solution was filtered through a nylon filter of 0.45 µm (Whatman™, Buckinghamshire, UK). Absorbance was measured against a blank of acetone with 1% BHT with a wavelength range of 400 to 700 nm. The total amount of chlorophyll-a, chlorophyll-b, and carotenoids were then calculated according to the formulas of Lichtenthaler [35].

Moisture Content
Two grams of U. rigida were placed in previously dried crucibles (2 h, 105 • C). The samples were dried in an oven at 105 • C for 10-12 h and weight was registered after cooling (30 min).

Protein Content
The nitrogen content of samples was determined by elemental analysis using a LECO TruSpec-Micro CHNS 630-200-200 elemental analyzer (St. Joseph, MI, USA) at a combustion furnace temperature of 1075 • C and an afterburner temperature of 850 • C. Nitrogen was detected by thermal conductivity. Protein content was calculated using a nitrogen-protein conversion factor of 6.25.

Dietary Fiber
Macroalgae were analyzed in terms of their insoluble, soluble, and total dietary fiber content, according to the enzymatic gravimetric method AOAC 991.43. This analysis was performed using the Total Dietary Fiber Assay kit (Megazyme, Bray, UK).

Iron (Fe)
A microwave assisted acid digestion procedure was performed for sample mineralization according to Domínguez-González et al. [36], with some modifications. Briefly, dried samples (ca. 200-220 mg) were accurately weighted into acid-washed Teflon vessels and were added with 2 mL HNO 3 69% (w/w). Then, the vessels were closed and placed inside a microwave oven to be digested over 2 cycles of the following extraction program: temperature was first raised to 170 • C (ramp time: 5 min) and held for 10 min. After cooling down, the vessels were carefully opened and 0.25 mL H 2 O 2 30% (w/w) were added, followed by a second microwave digestion cycle. Fe was quantified in a Perkin Elmer (Waltham, MA, USA) Analyst 100 flame atomic absorption spectrometer equipped with single hollow cathode lamps for each element and an air-acetylene burner.