The growing consumption of foods with nutritional claims has led the industry to develop added value products with health-promoting properties. Today there is a wide availability of functional foods meeting a broad range of consumer needs. In addition, the demand for lactose- or salt-free products is increasing. To this latter regard, the development of products with a reduced amount of sodium (Na) is an important issue and a public health concern, given that an excess of Na intake has been associated with higher cardio- and cerebrovascular risks and related diseases, such as myocardial infarction and stroke (Ha, Reference Ha2014). The World Health Organization (WHO) recommends a dietary Na reduction of 30%, corresponding to a maximum of 5 g/d of salt, by 2025 (Lacruz et al., Reference Lacruz, Kluttig, Hartwig, Löer, Tiller, Greiser, Werdan and Haerting2015).
Considering that cheese consumption is increasing worldwide, the production of low-NaCl cheeses is becoming an important task for the dairy industry. Removing or lowering NaCl in a cheese is challenging as NaCl modulates cheese ripening by influencing microbial and enzymatic activities, which in turn affect lactose metabolism, cheese pH, and the degradation of fats and casein with the consequent release of free fatty acids, peptides, and free amino acids (AAs) (Møller et al., Reference Møller, Rattray, Høier and Ardö2012). NaCl also exerts an antagonistic action against spoilage and pathogenic microorganisms (McSweeney, Reference McSweeney, Kilcast and Angus2007). Some attempts have already been made to lower the total NaCl content in cheese or to replace it, partially or totally, using alternative salts, with serious consequences on product flavor and texture. To overcome this problem a number of strategies, including the use of salt replacers and flavor enhancers, have been suggested to restore the flavor imbalance in products with reduced NaCl levels (Cruz et al., Reference Cruz, Faria, Pollonio, Bolini, Celeghini, Granato and Shah2011; Binia et al., Reference Binia, Jaeger, Hu, Singh and Zimmermann2015; van Buren et al., Reference van Buren, Dötsch-Klerk, Seewi and Newson2016). Studies aimed at salt replacement based on the reduction of Na ions (Na) by other salt-tasting cations such as potassium (K), ammonium (NH4+), calcium (Ca2+), and lithium (Li+) as well as by anions such as phosphate and glutamates, did not always give products of acceptable quality (Cruz et al., Reference Cruz, Faria, Pollonio, Bolini, Celeghini, Granato and Shah2011). No literature is available on NaCl reduction of Grana and similar cheeses, which usually contain a high (between 1.4–1.8%) NaCl concentration (Gobbetti et al., Reference Gobbetti, Neviani and Fox2018). In this study, cheeses with reduced amount of NaCl were produced according to a generic Grana cheese technology. NaCl reduction was achieved by salting cheeses in experimental brines containing different concentrations of KCl and proportionally reduced amounts of NaCl. The migration of Na and K into the cheese during ripening was investigated. The effect of NaCl replacement on ripening and sensory properties of cheeses was also evaluated.
Materials and methods
Cheese manufacture and experimental design
Bulk milk collected from different farms of northern Italy was standardized by Granarolo S.p.a. (Bologna, Italy) to 2.4% of fat and 3.7% of proteins (approx. 3.0% casein), equivalent to a fat/casein ratio of ~0.80 to minimize the composition variability between the different cheesemaking trials. Before cheese manufacture, milk was subjected to a mild thermization (63 ± 1 °C, 30 s) using a plate heat exchanger (Milkproject, Merone, CO, Italy) and cooled to 32 °C. Cheeses were manufactured following a generic Gran-type cheesemaking process at the pilot plant belonging to CREA-ZA (Lodi, Italy), using for each trial 550 l of milk. Whey starter culture was kindly provided by a dairy plant producing Grana located in Lodi (Italy). The amount of whey starter to be added was calculated to achieve an increment of milk acidity of 1.60 °SH/100 ml. Twenty-five mg/l of lysozyme (Food.com, Cremona, Italy) and 0.3 ml/l of liquid calf rennet (strength 100.000 International Milk Clotting Units, IMCU/kg, Santamaria Srl, Burago di Molgora, Italy) were added. At coagulation, the renneting time was determined and after cutting, the curd was mechanically stirred and scalded using a steam generator until the cooking temperature (53 °C in approx. 5 min) was reached (online Supplementary Table S1). The curd was left under whey for hardening for about 40 min, then cut into two equivalent portions of about 24 kg/each, and molded. Cheeses were left at 20 °C for 48 h to complete acidification and whey drainage, and then were immersed into the brines for 11 d. Brines were prepared as supersaturated solutions by dissolving 30 kg of salt mixture in 70 kg of tap water at 65 °C. Using food grade KCl (Jiangusu Kolod Food Ingredients Co. Ltd, Lianyungang City, China) and NaCl marine salt (Salinen Italia Srl, Milan, Italy), four brines were prepared: the first brine (K1) was made with a 1 : 0.89 (w/w) mixture of NaCl/KCl, the second (K2) was made with a 1 : 0.67 (w/w) mixture of NaCl/KCl whereas the third (K3) was made with a 1 : 0.32 (w/w) mixture of NaCl/KCl. The fourth brine (Na) was composed of only marine NaCl and used as control. After salting, corresponding K-cheeses (named as K1, K2, and K3 cheeses) and controls were ripened for 9 months. Of the duplicated wheels generated from each vat, one was used to take samples for the analysis and the second one was kept intact for sensory evaluation at the end of ripening. Cheese productions were replicated three times for each brine formulation, carried out on different days.
Compositional analysis and cheese yield
Fat and protein content of milk were determined using a FT-IR Milkoscan FT2 (Foss, Padova, Italy). The pH of both milk and cheeses was measured by a portable pH-meter (Portavo-907, Knick, Germany). Acidity of milk and whey starters was evaluated by titration with 0.25 N NaOH and expressed as °SH/50 ml. Weight loss during brining was calculated by weighing cheeses before and immediately after the immersion in the brines (T 0). Cheese composition and yield were determined at T 0 and after 9 months of ripening (T 9). Actual yield of cheese was normalized to reference level of milk fat (2.4%, w/w), milk protein (3.7%, w/w) and to cheese moisture content (40%, w/w) and calculated as reported by Francolino et al. Reference Francolino, Locci, Ghiglietti, Iezzi and Mucchetti2010. Dry matter was determined by oven drying at 102 °C (IDF 4, 2004). Total ash was obtained gravimetrically after dry ashing at 550 °C. The fat and protein contents were determined using the Van Gulik and Kjeldahl methods, respectively (FIL-IDF ISO, 3433, 2008 IDF 222:2008; FIL-IDF ISO, 8968-1 IDF 020-1, 2014). Sugars, organic acids and pyroglutamic acid were determined by HPLC according to Bouzas et al. (Reference Bouzas, Kantt, Bodyfelt and Torres1991). The content of the free amino acids (AAs) was determined according to Hogenboom et al. (Reference Hogenboom, D'Incecco, Fuselli and Pellegrino2017).
Sampling and analysis of cations
Salt distribution along the cheese was evaluated by dividing the cheese wheel into portions corresponding to three different layers, i.e. the outer (0–3 cm), the intermediate (3–6 cm) and the innermost (6–9 cm) layer. At three (T 3), six (T 6), and 9 months (T 9) of ripening, three pieces of cheese were extracted by means of a coring device of equal depth and the sample of each layer resulted from mixing and shredding the three pieces of the same layer. The distribution of Na and K were expressed as relative percentage of each layer respect to the sum of the three layers. The content of Na and K in the ripened cheeses (T 9) was carried out on a slice representing the whole cheese and results are the average of three replicates. To check the contribution of the experimental brines in the final product, the net concentration of cations coming from the brine was determined by subtracting the native content in Na and K of the unsalted curd from the total content found in the cheese. The native content of Na and K was calculated as the average of 8 unsalted curds, normalized to the average dry matter reduction of the ripened (T 9) cheeses. Na and K content in cheeses and brines was determined by atomic absorbance spectrometer (Varian AA240FS) according to ISO 8070 (IDF 119, 2007).
Microbiological analysis
Milk and cheese samples were collected aseptically and stored in sterile bottles (or bags), transferred to the laboratory under refrigeration and immediately analyzed. Cheese samples (10 g) were homogenized in sterile 2% trisodium citrate solution (1 : 10 w:v). Ten-fold dilutions of all samples were performed in sterile quarter-strength Ringer solution (Thermo Scientific Oxoid, Basingstoke, UK). Microbiological quality of the vat milk was evaluated by determining the total bacterial count (TBC), total coliforms, Escherichia coli, propionic acid bacteria (PAB), and butyric acid clostridia (BAC) spores. TBC was carried out on Plate Count Agar (PCA, Oxoid) after incubation at 30 °C for 72 h. Total coliforms and E. coli were counted on ChromID Coli agar (BioMérieux, Bagno a Ripoli, Italy) at 37 °C for 48 h. PAB were counted on Pal Propiobac agar medium (Laboratoires Standa, Caen, France) at 30 °C for 6 d under anaerobic conditions, according to Thierry and Madec (Reference Thierry and Madec1995). The number of BAC spores was determined with the Most Probable Number (MPN) method using the Bryant Burkey broth containing resazurin and lactate (5% final concentration) (BBL, Biolife, Milan, Italy). Three 10-fold dilutions (1 ml) with five tubes at each dilution were inoculated in BBL tubes (10 ml). All tubes were sealed with 2 ml of melted paraffin/vaseline mixture (Sacco S.r.l, Cadorago, Italy) and heated at 80 °C for 15 min. Incubation was carried out at 37 °C for 7 d. MPN counts were expressed as spores/l. Mesophilic and thermophilic lactic acid bacteria (LAB) were determined at T 0, T 3 and T 9. Thermophilic LAB were counted on whey agar medium (WAM) after anaerobic incubation at 42 °C for 48 h (Gatti et al., Reference Gatti, Lazzi, Rossetti, Mucchetti and Neviani2003). Mesophilic non-starter lactic acid bacteria (NSLAB) were enumerated by spread-plating on maltose-MRS-vancomycin agar according to Di Lena et al. (Reference Di Lena, Quero, Santovito, Verran, De Angelis and Fusco2015) after incubation at 30 °C for 48 h under anaerobic conditions. On cheeses at T 3 and T 9, total coliforms, E. coli, PAB and BAC spores were determined as previously described for milk samples. Coagulase positive staphylococci were counted on Baird Parker agar base supplemented with rabbit plasma fibrinogen (BP-RPF, BioMerieux), incubated at 37 °C for 24 h. The presence of Listeria monocytogenes was evaluated in 25 g of sample in accordance with the ISO standard method 11290-1 (ISO, 2017).
Sensory analysis
Using randomized three-digit codes, a discrimination triangle test between the K-cheeses (K1, K2, K3) and controls was performed with the support of a panel (n = 20–24) of skilled, non-trained members. Tests were carried out in different days on cheeses at T 9. Cheese slices were removed from the crust and chopped into small cubes (2–3 cm). The panel was asked to detect the odd cheese sample.
Statistical analysis
One-way analysis of Variance (ANOVA; α = 0.05) was performed to assess statistical differences on technological, microbiological and chemical data based on the different salting options. Pearson correlation coefficient was calculated with the software package Statistica for Windows version 6.1 (StatSoft Inc., Tulsa, Oklahoma, USA). For the sensory analysis, a binomial test was carried out to assess if the number of correct responses of the panel gave a higher probability level than a random classification process.
Results
Milk characteristics and cheesemaking process
Fat and protein content of milk was 2.40 ± 0.08 and 3.71 ± 0.06%, respectively, in all the trials. The mean pH value was 6.70 ± 0.04 and the mean titratable acidity was 3.75 ± 0.09 °SH/50 ml. After the mild heat treatment, TBC of milk was on average 4.55 ± 0.74 log CFU/ml, while coliforms and E. coli were at levels of 0.57 ± 0.95 log CFU/ml and 0.57 ± 0.63 log CFU/ml, respectively. Concerning the main potential spoilage bacteria for this type of cheese, PAB and BAC were present at very low levels. PAB showed an average count of 1.61 ± 0.52 log CFU/ml. The spores of BAC ranged from 2.26 to 3.36 log MPN/l (in 1 of 12 samples), with an average of 2.42 ± 0.34 log MPN/l.
All the technological parameters measured during the cheesemaking showed that the process was extremely reproducible (online Supplementary Table S1). All the cheeses reached the desired target pH (5.10 ± 0.03) at the end of the second turning of molded cheeses, before brining. The evolution of pH during ripening showed that the acidification process was not influenced by the different salting options (online Supplementary Fig. S1). As ripening proceeded, pH increased, reaching similar values (5.53 ± 0.07) at T 9 in all the samples. During brining, with respect to controls (3.19 ± 0.11% w/w), a lower weight loss (2.81 ± 0.30, 1.92 ± 0.12, and 2.71 ± 0.45% in K1, K2, and K3 cheeses, respectively) linked to a more limited whey release in K-brines was observed in the K-cheeses.
Cation determination and distribution along the cheese
The content of Na and K in the freshly prepared brines is shown in online Supplementary Table S2. With respect to the control brine, K1, K2, and K3 brines showed a reduced content of Na of approx. 50, 40, and 30%, respectively. Figure 1 reports the average % distribution of Na and K through the three layers of the cheeses. After 3 months, a centripetal gradient of the absorbed Na was still evident whereas K seemed to be homogeneously distributed, independently of its concentration in the brine. K cations seemed to diffuse into cheese more rapidly than Na although Bona et al. (Reference Bona, dos Santos Ferreira da Silva, Borsato, Monken e Silva and de Souza Fidelis2010) calculated a higher diffusion coefficient of NaCl compared to KCl in Prato cheese, which has a structure, size and ripening time quite different from Grana cheese. At T 6, both cations showed an equal average distribution along the three layers, thus confirming previous findings (Morris, Reference Morris1985; Srbinovska et al. Reference Srbinovska, Čizbanovski, Džabirski, Andonov and Palasevski2001).
Fig. 1. Percent distribution of Na (a, b, c) and K (d, e, f) cations in three different depth layers of cheeses at T 3, T 6 and T 9 months of ripening in the cheeses derived from the K-brines.
At T 9, the content of Na and K was also measured on a representative cheese slice and the results are summarized in Table 1. The native concentrations of Na and K in the curds, which were 25.22 ± 4.17 and 116.18 ± 9.51 mg/100 g, respectively, were subtracted from total Na and K to calculate the exogenous sodium (Naex) and potassium (Kex) absorbed from the brines. In cheeses immersed in the brine with the highest concentration of potassium (K1), Kex resulted higher than Naex (0.37 ± 0.04 vs. 0.27 ± 0.02 g/100 g, respectively), which allowed a reduction in the total Na content of 51.3% with respect to control cheeses. In cheese salted with K2 and K3 brines, the Kex was lower than Naex and the percentages of Na reduction were lower (27.1 and 21.5%, respectively; Table 1). A linear correlation (R 2 = 0.99) between the ratio of exogenous amounts of cations found in the cheese (Kex/Naex) and the ratio of the cations (K/Na) in the relative brine was observed (online Supplementary Fig. S2). This could be useful to predict the contribution of the exogenous cations absorbed from brines with a given mixture of NaCl and KCl.
Table 1. Total content of Na and K and the exogenous content (Naex and Kex) in the cheeses analyzed after 9 months of ripening. The ratio between the two cations and the reduction of Na obtained in comparison with the control are reported
K1 = cheeses salted in brine with NaCl/KCl ratio 1/0.89 (w/w).
K2 = cheeses salted in brine with NaCl/KCl with ratio 1/0.67 (w/w).
K3 = cheeses salted in brine with NaCl/KCl ratio 1/0.32 (w/w).
Na = cheeses salted in brine with only NaCl.
n.d. = not detectable.
Microbiological characteristics of cheeses
Listeria monocytogenes was absent from the cheese, whereas total coliforms, E. coli and coagulase positive staphylococci were always <1 log CFU/g. Late blowing defects by lactate-fermenting gas-producing anaerobic bacteria never occurred. The visual inspection of the cheeses enabled us to exclude alterations caused by abnormal accumulation of gas. Spores of BAC were under the detection limit of the MPN method (<0.30 log MPN/g) at both ripening times (T 3 and T 9). PAB were found at low levels (between 2 and 3 log CFU/g), with only a significantly higher value in one K2 cheese at T 3 (Table 2). Even at T 9, PAB were still about 2 log CFU/g, independent of the type of salting. The thermophilic LAB, mainly originating from the whey starter, reached levels of 6–7 log CFU/g at T 0 in all the cheeses, and a slight decrease was observed at T 3. At the end of ripening (T 9), thermophilic LAB showed a reduction of more than 3 log (Table 2). Mesophilic non-starter LAB (NSLAB) were initially found (T 0) to be about 3 log CFU/g, suggesting that the moderate heat treatment of milk allowed their survival. During ripening, NSLAB progressively increased, reaching levels of 5–6 log CFU/g until T 9 (Table 2). No significant differences were observed between the four salting options, indicating that the LAB microbiota was not affected, at least in counts, by the partial replacement of NaCl with KCl.
Table 2. Content of mesophilic and thermophilic lactic acid bacteria (LAB) and propionic acid bacteria (PAB) determined in the cheeses at three different stages of ripening (T = 0, 3 and 9 months) expressed in log10 CFU/g ± standard deviation (sd).
*Within a column, values indicated with a superscript differ significantly at P < 0.05 (no superscripts mean that values are not statistically different).
K1 = cheeses salted in brine with NaCl/KCl ratio 1/0.89 (w/w).
K2 = cheeses salted in brine with NaCl/KCl with ratio 1/0.67 (w/w).
K3 = cheeses salted in brine with NaCl/KCl ratio 1/0.32 (w/w).
Na = cheeses salted in brine with only NaCl.
Effect on chemical composition and ripening
None of the measured characteristics differed significantly between the control and K-cheeses (Table 3). K-immersion caused a small and non-significant decrease in dry matter. The ratio between fat, protein and moisture were in the range reported in the literature for Grana cheese (Gobbetti et al., Reference Gobbetti, Neviani and Fox2018). These small or absent differences are in agreement with previous papers regarding the substitution of NaCl with KCl in brine or by dry salting (Katsiari et al., Reference Katsiari, Voutsinas, Alichanidis and Roussis1998; Ayyash et al., Reference Ayyash, Sherkat and Shah2012; Soares et al., Reference Soares, Fernando, Alvarenga and Martins2016; Van Hekken et al., Reference Van Hekken, Tunick, Renye and Tomasula2017). Likewise, no significant differences were observed in cheese yield nor in organic acid content (citric, lactic, acetic acid).
Table 3. Yield, dry matter and chemical composition of the K-salted cheeses (K1, K2 and K3) and controls (Na) determined and the beginning (T 0) and the end (T 9) of ripening. The values are averages of three replicates ± sd
K1 = cheeses salted in brine with NaCl/KCl ratio 1/0.89 (w/w).
K2 = cheeses salted in brine with NaCl/KCl with ratio 1/0.67 (w/w).
K3 = cheeses salted in brine with NaCl/KCl ratio 1/0.32 (w/w).
Na = cheeses salted in brine with only NaCl.
Free pyroglutamic acid was undetectable at the beginning of ripening and increased up to about 0.3 g/100 g at T 9, with the same rate in both K-cheeses and controls (Fig. 2). To better evaluate the maturation process, the free AAs content was investigated. The majority of individual amino acids and the overall amount of free AAs at T 9 did not differ significantly between K-cheeses and controls (online Supplementary Table 3).
Fig. 2. Content of pyroglutamic acid (g/100 g) determined after salting of cheeses (T 0) and after 3, 6 and 9 months of ripening.
Sensory evaluation
In the discrimination triangle test between K-salted cheeses and controls, the number of correctly identified samples/total number of samples (i.e. 8/24 for K1 vs. Na, 10/20 for K2 vs. Na, 6/24 for K3 vs. Na) was not significant. Additionally, a small size consumer test (30 consumers) was carried out, asking to choose the favorite cheese between a control and a K3 salted cheese and if any bitter taste was perceived among the two cheeses. The K3 salted cheese obtained 58% of preference and no bitter aftertaste was noticed by consumers. Glutamic acid is known to be a flavor-enhancing compound providing an umami taste, and the high content of this AA found in all cheeses (online Supplementary Table S3), may have exerted a masking effect on the bitterness generally associated with KCl (Homma et al., Reference Homma, Yamashita, Funaki, Ueda, Sakurai, Ishimaru and Asakura2012).
Discussion
Our aim was to reduce the salt content of hard cheeses without affecting their consumer characteristics. Hard cheeses were produced using the technology of Grana cheese (Grana-type) and salted using three brines containing different amounts of KCl (K-brines) and compared with control cheeses, salted with marine NaCl. Literature indicates that the total replacement of NaCl with KCl (or other salts such as MgCl2 or CaCl2) causes marked acidity and structural alterations in cheese (Guinee and O'Kennedy, Reference Guinee, O'Kennedy, Kilcast and Angus2007). Specifically, the lack of NaCl negatively affects flavor, texture, proteolysis and starter activity, as shown for Cheddar and other cheeses (Cruz et al., Reference Cruz, Faria, Pollonio, Bolini, Celeghini, Granato and Shah2011; Rulikowska et al., Reference Rulikowska, Kilcawley, Doolan, Alonso-Gomez, Nongonierma, Hannon and Wilkinson2013; Yanachkina et al., Reference Yanachkina, McCarthy, Guinee and Wilkinson2016). A partial reduction or replacement with other salts is more feasible and some attempts have been performed in cheese like cottage, Cheddar and Gouda. In previous works, the substitution of NaCl for mixtures of 50 : 50% of NaCl and KCl did not cause any biochemical, textural or microbiological changes, although some discrepancies among studies have been observed (Cruz et al., Reference Cruz, Faria, Pollonio, Bolini, Celeghini, Granato and Shah2011). For instance, in Caciotta cheese the partial substitution (55%) of NaCl with KCl did not affect flavor and texture (Quattrucci et al., Reference Quattrucci, Bruschi, Manzi, Aromolo and Panfili1997) whereas Van Den Berg et al. (Reference Van Den Berg, de Vries and Stadhouders1986) claimed that a substitution of only 20% of salt in Gouda cheese increases the risk of butyric acid fermentation. A concentration of KCl >1% was reported to cause considerable sourness perception and increased proteolysis (Cruz et al. Reference Cruz, Faria, Pollonio, Bolini, Celeghini, Granato and Shah2011).
This study aimed to reduce the Na content in cheese by using brines with different amounts of KCl. Our results highlighted that, even at the highest concentration of K in K1 cheese (0.49 ± 0.04 g/100 g), no effect on water content, composition and sensory properties was observed. For a hard cheese like Grana, the NSLAB microbiota plays a recognized role in cheese ripening (Gatti et al., Reference Gatti, Bottari, Lazzi, Neviani and Mucchetti2014). The use of KCl did not affect the growth and viability of both thermophilic LAB and NSLAB. Cheese ripening showed a similar behavior between control and K-cheeses, as confirmed by the content of free pyroglutamic acid, an index related to Grana cheese ripening, that was in agreement with literature data (Mucchetti et al., Reference Mucchetti, Locci, Gatti, Neviani, Addeo, Dossena and Marchelli2000). The total content of free AAs was higher than that reported in Grana cheeses aged 9 months, and more similar to cheeses aged about 14 months (Resmini et al., Reference Resmini, Hogenboom, Pazzaglia and Pellegrino1993; Cattaneo et al., Reference Cattaneo, Hogenboom, Masotti, Rosi, Pellegrino and Resmini2008). This accelerated ripening can be attributed to the smaller size of the experimental cheeses, i.e. between 23–26 kg, compared to regular Grana cheeses, whose weight ranges from 24 to 40 kg. The smaller cheese size made it possible to equilibrate the temperature of the inner and central parts of the cheese, allowing a faster growth of the thermophilic LAB, thus accelerating their autolysis and the subsequent release of endocellular proteinases and peptidases (Giraffa et al., Reference Giraffa, Rossetti, Mucchetti, Addeo and Neviani1998). Furthermore, an accelerated growth of mesophilic LAB counts (~7 log10 CFU/g after 3 months; Table 2) may have contributed to the proteolysis. The accumulation of the single AAs in the cheeses was in agreement with typical values reported from Grana cheeses (Hogenboom et al., Reference Hogenboom, D'Incecco, Fuselli and Pellegrino2017) and no particular differences were observed between all samples. We can therefore assume that the proteolysis was not affected by the K-brines tested in this work. No defects were observed in K-cheeses, and they were not significantly distinguishable from controls in sensory analysis, even when the reduction of Na was ~50%. This result is of great commercial importance since the acceptability of KCl as a brine salting agent is questioned due to its pronounced bitter/chemical metallic aftertaste or the residual sour flavor conferred to cheese (Reddy and Marth, Reference Reddy and Marth1994; Zorrilla and Rubiolo, Reference Zorrilla and Rubiolo1999; Cruz et al., Reference Cruz, Faria, Pollonio, Bolini, Celeghini, Granato and Shah2011).
In conclusion, this is the first study about the reduction of Na in Grana-type cheese, which refers to a category of hard cheeses showing a relevant place in the world cheese market and a relatively high (1.4–1.8%) NaCl content. The reduction of Na of about 50% in the final product, which was successfully achieved with a brine containing almost equal concentration of Na and K, meets the WHO guidelines that recommend producing foods with a lower NaCl content. WHO also recommends a maximum dietary potassium intake of 3.9 g/d to people who tend to develop hyperkalemia. In this regard, our data showed that a portion of 100 g K-salted cheese may contribute to about 10% of the daily recommended intake (Table 1). Moreover, foods with high protein content (such as hard cheeses) are preferred compared to food with similar potassium intake and low protein content (such as milk and yogurts) since protein intake must be increased in hemodialysis and peritoneal dialysis patients (Cupisti et al., Reference Cupisti, Kovesdy, D'Alessandro and Kalantar-Zadeh2018).
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S0022029919000797.
