Advertisement

Gas production by Paucilactobacillus wasatchensis WDCO4 is increased in Cheddar cheese containing sodium gluconate

Open AccessPublished:March 10, 2022DOI:https://doi.org/10.3168/jds.2021-21617

      ABSTRACT

      Paucilactobacillus wasatchensis can use gluconate (GLCN) as well as galactose as an energy source and because sodium GLCN can be added during salting of Cheddar cheese to reduce calcium lactate crystal formation, our primary objective was to determine if the presence of GLCN in cheese is another risk factor for unwanted gas production leading to slits in cheese. A secondary objective was to calculate the amount of CO2 produced during storage and to relate this to the amount of gas-forming substrate that was utilized. Ribose was added to promote growth of Pa. wasatchensis WDC04 (P.waWDC04) to high numbers during storage. Cheddar cheese was made with lactococcal starter culture with addition of P.waWDC04 on 3 separate occasions. After milling, the curd was divided into six 10-kg portions. To the curd was added (A) salt, or salt plus (B) 0.5% galactose + 0.5% ribose (similar to previous studies), (C) 1% sodium GLCN, (D) 1% sodium GLCN + 0.5% ribose, (E) 2% sodium GLCN, (F) 2% sodium GLCN + 0.5% ribose. A vat of cheese without added P.waWDC04 was made using the same milk and a block of cheese used as an additional control. Cheeses were cut into 900-g pieces, vacuum packaged and stored at 12°C for 16 wk. Each month the bags were examined for gas production and cheese sampled and tested for lactose, galactose and GLCN content, and microbial numbers. In the control cheese, P.waWDC04 remained undetected (i.e., <104 cfu/g), whereas in cheeses A, C, and E it increased to 107 cfu/g, and when ribose was included with salting (cheeses B, D, and F) increased to 108 cfu/g. The amount of gas (measured as headspace height or calculated as mmoles of CO2) during 16 wk storage was increased by adding P.waWDC04 into the milk, and by adding galactose or GLCN to the curd. Galactose levels in cheese B were depleted by 12 wk while no other cheeses had residual galactose. Except for cheese D, the other cheeses with GLCN added (C, E and F) showed little decline in GLCN levels until wk 12, even though gas was being produced starting at wk 4. Based on calculations of CO2 in headspace plus CO2 dissolved in cheese, galactose and GLCN added to cheese curd only accounted for about half of total gas production. It is proposed that CO2 was also produced by decarboxylation of amino acids. Although P.waWDC04 does not have all the genes for complete conversion and decarboxylation of the amino acids in cheese, this can be achieved in conjunction with starter culture lactococcal. Adding GLCN to curd can now be considered another confirmed risk factor for unwanted gas production during storage of Cheddar cheese that can lead to slits and cracks in cheese. Putative risk factors now include having a community of bacteria in cheese leading to decarboxylation of amino acids and release of CO2 as well autolysis of the starter culture that would provide a supply of ribose that can promote growth of Pa. wasatchensis.

      Key words

      INTRODUCTION

      Paucilactobacillus wasatchensis has been identified as a component of the nonstarter lactobacilli in cheese that can cause unwanted gas formation during the aging of cheese (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ). Before division of the Lactobacillus genus into 27 different genera (
      • Zheng J.
      • Wittouck S.
      • Salvetti E.
      • Franz C.M.A. P
      • Harris H.M.B.
      • Mattarelli P.
      • O'Toole P.W.
      • Pot B.
      • Vandamme P.
      • Walter J.
      • Watanabe K.
      • Wuyts S.
      • Felis G.E.
      • Ganzle M.G.
      • Lebeer S.
      A taxonomic note on the genus Lactobacillus: Description of 23 novel genera, emended description of the genus Lactobacillus Beijerinck 1901, and union of Lactobacillaceae and Leuconostocaceae.
      ), it was designated as Lactobacillus wasatchensis (
      • Oberg C.J.
      • Oberg T.S.
      • Culumber M.D.
      • Ortakci F.
      • Broadbent J.R.
      • McMahon D.J.
      Lactobacillus wasatchensis sp. nov., a non-starter lactic acid bacteria isolated from aged Cheddar cheese.
      ). It is now grouped with other lactobacilli that do not utilize glucose and can metabolize only a few carbohydrates as part of the Paucilactobacillus genus. In addition to using 5-carbon sugars, Pa. wasatchensis can utilize some 6-carbon sugars such as galactose as an energy source and after removing one carbon atom as C02, the remaining 5-carbon molecule can be fermented in the pentose phosphate pathway (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Growth and gas production of a novel obligatory heterofermentative Cheddar cheese nonstarter lactobacilli species on ribose and galactose.
      ;
      • Oberg C.
      • Sorensen K.
      • Oberg T.
      • Young S.
      • Domek M.
      • Culumber M.
      • McMahon D.
      Gluconate metabolism and gas production by Paucilactobacillus wasatchensis WDC04.
      ). Using a starter culture for making Cheddar cheese that includes Streptococcus thermophilus leaves residual galactose in the cheese and this is a risk factor for having slit development in cheese when Pa. wasatchensis is part of the nonstarter lactic acid bacteria population (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Growth and gas formation by Lactobacillus wasatchensis, a novel obligatory heterofermentative nonstarter lactic acid bacterium, in Cheddar-style cheese made using a Streptococcus thermophilus starter.
      ). In addition to galactose, the type strain Pa. wasatchensis WDC04 (Pa.wWDC04) can also utilize fructose and N-acetylglucosamine, and has a marginal ability to utilize lactose and glucose (
      • McMahon D.J.
      • Bowen I.B.
      • Green I.R.
      • Domek M.J.
      • Oberg C.J.
      Growth and survival characteristics of Paucilactobacillus wasatchensis WDC04.
      ). N-acetylglucosamine from lysed starter culture cells would help promote growth of nonstarter lactobacilli in cheese depending on the species and strains present (
      • Møller C.O.A.
      • Christensen B.B.
      • Rattray F.P.
      Modelling the biphasic growth of non-starter lactic acid bacteria on starter-lysate as a substrate.
      ).
      Recently,
      • Oberg C.
      • Sorensen K.
      • Oberg T.
      • Young S.
      • Domek M.
      • Culumber M.
      • McMahon D.
      Gluconate metabolism and gas production by Paucilactobacillus wasatchensis WDC04.
      found that Pa. wasatchensis has the metabolic pathway to import and phosphorylate gluconate (GLCN) as a substrate to support growth. They showed that 4 distinct strains of Pa. wasatchensis isolated from cheese could all grow to high number in carbohydrate-restricted media with GLCN as the only added substrate and produce CO2. They proposed that the ability of bacteria such as Pa. wasatchensis to utilize GLCN relates to its original habitat in silage and other plant materials in which molds are present that convert glucose to GLCN as a way to limit growth of many bacteria that would compete for resources (
      • Goldberg I.
      • Rokem J.S.
      Organic and fatty acid production, microbial.
      ).
      Adding sodium GLCN (NaGLCN) to Cheddar cheese curd during salting is used as a way prevent formation of calcium lactate crystals during cheese storage even when lactate levels are high (
      • Phadungath C.
      • Metzger L.E.
      Effect of sodium gluconate on the solubility of calcium lactate.
      ). Gluconate forms a soluble complex with calcium and lactate so insufficient calcium lactate is then available to precipitate and form crystals. Based on anecdotal information, use of NaGLCN as a processing aid during cheesemaking is becoming more common in the United States, with levels of 0.5 to 1% (wt/wt) being used, and sometimes up to 1.5%. A side benefit has been that the solubilization of calcium results in the cheese being ready to convert into shreds and slices at shorter storage times.
      The initial aim of this research was to determine if adding NaGLCN to Cheddar cheese would increase the likelihood of such cheese having unwanted gas formation during storage. To accelerate gas production, ribose can be added to the cheese during salting (as previously shown by
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ) and addition of galactose can be used as a comparison. By calculating the amount of CO2 produced in the cheese bags, this can then be related to the amount of substrates (galactose and GLCN) utilized during storage to determine if there are other potential substrates in cheese that can be utilized by Pa. wasatchensis to generate unwanted gas production that may cause slits and cracks in the cheese.

      MATERIALS AND METHODS

      Cheese Manufacture

      Cheddar cheese was made from 570-kg aliquots of pasteurized milk in an enclosed vat as described by
      • McMahon D.J.
      • Oberg C.J.
      • Drake M.A.
      • Farkye N.
      • Moyes L.V.
      • Arnold M.R.
      • Ganesan B.
      • Steele J.
      • Broadbent J.R.
      Effect of sodium, potassium, magnesium, and calcium salt cations on pH, proteolysis, organic acids, and microbial populations during storage of full fat Cheddar cheese.
      . The experimental design was to make a regular vat of cheese and on the following day an experimental vat that was inoculated with 2 mL of a Pa.wWDC04 cell suspension containing ∼1 × 1010 cfu/mL along with the starter culture in 3 replicates on separate occasions. A lactococcal starter culture (DVS850; Chr. Hansen Inc.) was used, calcium chloride solution (Nelson Jameson) was added at 0.11 mL/kg, single-strength annatto (DSM Food Specialties USA Inc.) added at 0.072 mL/kg of milk. Rennet was added and the curd allowed to set for 30 min then cut over 5 min (for a total of 60 revolutions of the cutting blades). The curd was heated to 39°C over 35 min, and then stirred for about another 35 min until curd pH reached 6.3 then pumped onto a drain table and the whey drained. The curd was then cheddared for ∼2 h until the curd pH reached 5.4 to 5.5.
      For the regular vat, the curd was milled and salted (at a rate of 2.95 kg per 1,000 kg of milk) on the drain table using 3 applications of salt over 15 min, then hooped and pressed. One block was retained from this vat and this cheese was designated as “No Pa.wWDC04” and used for comparison with cheese made with Pa.wWDC04 added to the milk.
      For the experimental vat, after the curd was milled it was separated into six 10-kg portions, placed into open plastic containers, and then salt, d-galactose (Alfa Aesar), d-ribose (WellBodyNaturals, LLC), and NaGLCN (Jungbunzlauer) added according to Table 1. Less salt was added when NaGLCN was also being added to maintain similar Na+ concentrations in the cheese and minimize an effect of Na+ concentration on microbial activity in the cheese. Each aliquot was added and manually mixed using 3 applications with 5 min between each application. The curd was allowed to stand for 10 min before being placed into plastic cheesecloth-lined stainless-steel hoops and pressed overnight (140 kPa, ∼18 h, ∼20°C). One portion of the curd did not receive any of the 6-carbon substrates (galactose or GLCN) or ribose and the cheese was designated as “No added substrate.”
      Table 1Percentage (wt/wt) of salt, ribose, galactose (Gal), and sodium gluconate (NaGLCN) added to cheese curd made with Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) added to milk with the starter culture
      CodeVatCheeseSalt %Ribose %Gal %NaGLCN %
      X1No Pa.w.WDC04
      A2No added substrate2.80
      None added.
      B20.5% Gal + Ribose
      Same Gal and ribose combination previously shown to induce gas production by Pa. wasatchensis WDC04 in Cheddar cheese (Ortakci et al., 2015b).
      2.800.500.50
      C21% NaGLCN2.551.0
      D21% NaGLCN + Ribose2.550.501.0
      E22% NaGLCN2.292.0
      F22% NaGLCN + Ribose2.290.502.0
      1 None added.
      2 Same Gal and ribose combination previously shown to induce gas production by Pa. wasatchensis WDC04 in Cheddar cheese (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ).

      Chemical Analysis

      Samples of cheese were frozen at 1, 4, 8, 12, and 16 wk, and kept frozen at −20°C until analyzed. Cheese pH and composition were measured on wk-1 cheese. Moisture was determined in triplicate by drying 2 to 4 g of shredded cheese for 18 h in a hot air oven at 95°C. Fat content was measured by Babcock method 15.083 (
      • Wehr H.M.
      • Frank J.F.
      Standard Methods for the Examination of Dairy Products.
      ). Mineral content was determined using inductively coupled plasma optical emission spectrometry (Thermo iCAP 6300, Thermo Fisher Scientific) after wet digestion. Protein was measured using dye absorption (Sprint; CEM Corporation). Cheese pH was measured using a glass electrode after mixing shredded cheese with distilled water in a 2:1 ratio. Salt was measured by adding 5.00 g of cheese to 98.20 g of distilled water and stomaching at 260 rpm for 4 min. The slurry was allowed to stand for 15 min, then filtered to remove cheese particles. Salt content was measured using a chloride analyzer (Model 926; Corning). Cheese samples for measurement of GLCN, galactose, and lactose concentration were prepared in the same manner as for salt and then measured using enzyme analysis kits K-GATE for GLCN and LACGAR for galactose and lactose (Megazyme Inc.).

      Microbial Enumeration

      Cheese samples [11 g added to 99 mL of sterile 2% (wt/wt) sodium citrate buffer] were stomached (260 rpm for 4 min), and serially diluted. Lactococci were enumerated by plating on M17-L agar and incubating aerobically at 30°C and counting colonies after 18 to 24 h as described by
      • Oberg C.J.
      • Moyes L.V.
      • Domek M.J.
      • Brothersen C.F.
      • McMahon D.J.
      Survival of probiotic adjunct cultures in cheese and challenges in their enumeration using selective media.
      to avoid counting colonies of nonstarter lactobacilli. Paucilactobacillus wasatchensis WDC04 was enumerated by plating on MRS agar (Sigma-Aldrich Inc.) was supplemented with 0.5% (wt/wt) ribose and 10 mg/L vancomycin (MRS+RV) and incubated anaerobically at 25°C using GasPak EZ pouches (VWR International LLC.). After 48 h, any fast-growing colonies were marked and the plates re-incubated for an additional 4 d with new GasPak EZ pouches inserted in the incubation jar and the slow-growing small colonies counted (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ).

      Cheese Packaging and Measurement of Gas Production

      To monitor gas formation during storage, 6 blocks of cheese from each treatment cheese in each of the 3 replicates, weighing 900 ± 0.05 g (∼5.7 × 17.5 × 9 cm) were placed in oxygen-barrier 25 × 30 cm plastic bags (Vilutis), vacuum evacuated to −85 kPa, and sealed 3 cm from the end of the bag. Additional blocks of each cheese were vacuum packed for microbial and chemical analysis. Cheese was stored at 12°C. When the cheese was 1, 4, 8, 12, and 16 wk old, the 6 packages of cheese were examined and the distance was measured from the block of cheese to the position at which the 2 sides of the plastic bag were still tightly held together by the vacuum in the bag. This was then expressed as head space height.

      Calculation of Carbon Dioxide Generated

      An estimation of the amount of CO2 produced during storage was calculated based upon solubility of CO2 in cheese, the volume of the headspace and the pressure inside the bag. Carbon dioxide solubility in the nonfat portion of cheese was based on the measurement by
      • Lamichhane P.
      • Sharma P.
      • Kelly A.L.
      • Risbo J.
      • Rattray F.P.
      • Sheehan J.J.
      Solubility of carbon dioxide in renneted casein matrices: Effect of pH, salt, temperature, partial pressure, and moisture to protein ratio.
      of 32 mmol/kg at 100 kPa atmospheric pressure and 13°C for an aqueous micellar casein matrix containing 62% moisture, 31% protein and 1.9% salt. Adjustments were made for the decrease in CO2 solubility that occurs with the increased protein and salt content of cheese relevant atmospheric pressure according to
      • Lamichhane P.
      • Sharma P.
      • Kelly A.L.
      • Risbo J.
      • Rattray F.P.
      • Sheehan J.J.
      Solubility of carbon dioxide in renneted casein matrices: Effect of pH, salt, temperature, partial pressure, and moisture to protein ratio.
      . Carbon dioxide solubility in the butterfat portion of cheese was calculated as 18.5 mmol/kg at 12°C and 85.5 kPa based upon
      • Truong T.
      • Palmer M.
      • Bansal N.
      • Bhandari B.
      Investigation of solubility of carbon dioxide in anhydrous milk fat by lab-scale manometric method.
      who reported solubility index for anhydrous milk fat of 39.0 and 11.9 mmol/kg/kPa at 24 and 4°C, respectively. Combining the aqueous and fat portion, CO2 solubility in Cheddar cheese at the atmospheric pressure at Utah State University (Logan, UT) and with storage temperature of 12°C was calculated as 21.7 mmol/kg.
      Because the cheese blocks were vacuum packaged to −85 kPa and the bags remained tightly around the cheese, it was assumed that the initial quantity of CO2 in the cheese was negligible. During storage, the plastic bags were observed to be pulled inward toward the cheese until the headspace height reached 3 cm, after which the interior of the bags was assumed to be at atmospheric pressure. The pressure inside the bag was calculated by interpolation (i.e., 0 kPa when headspace height < 1 mm and 8.5 kPa with headspace height ≥ 3 cm). All of the gas inside the bags was assumed to be CO2 and the amount of CO2 per unit volume was calculated based upon pressure inside the bags in a similar manner. The amount of CO2 produced during storage was considered as the quantity dissolved in the cheese plus the quantity in the headspace and expressed as millimoles.

      Statistical Analysis

      Cheese composition data were analyzed using a mixed model in which treatment is the fixed factor and replicate vats were considered the random factor. Contrasts of interests between treatments were performed as shown in Table 2. Changes in galactose, lactose and gluconate levels as well as starter culture and Pa.wWDC04 numbers during storage were analyzed with a mixed model in which treatment, storage time and their interaction were fixed factors and replicate vats and treatment by vat were random factors. Relative gas height and gas volume were analyzed by the above mixed model with unstructured covariances for the repeated measures on the same bag of cheese over the storage (using 6 bags of cheese per treatment per replicate). All analyses were performed using PROC GLIMMIX of SAS/STAT 15.1 (SAS Institute Inc.) and least square means and standard errors calculated. Significance was specified at 0.05 level. Tendencies were considered at 0.10 > P > 0.05. Contrasts in Table 2 at interested time points were tested.
      Table 2Individual contrasts used to test the effect of various treatments applied during manufacture of cheese with Paucilactobacillus wasatchensis WDC04 as shown in Table 1
      ComparisonContrast
      Addition of Pa. wasatchensis WDC04 to milkX vs. A
      Addition of Pa. wasatchensis WDC04 to milkX vs. (A, B, C, D, E, F)
      Addition of substrates (galactose, gluconate, or ribose)A vs. (B, C, D, E, F)
      Addition of galactose and ribose to curdA vs. B
      Addition of gluconate to curdA vs. (C, D) vs. (E, F)
      Level of gluconate added to curd(C, D) vs. (E, F)
      Adding 1% sodium gluconate to curdA vs. (C, D)
      Adding 2% sodium gluconate to curdA vs. (E, F)
      Addition of ribose (with galactose or gluconate)(A, C, E) vs. (B, D, F)
      Addition of ribose (with gluconate)(C, E) vs. (D, F)

      RESULTS AND DISCUSSION

      Initial Proximate Cheese Composition

      Having the cheese with no added Pa.wWDC04 made on a separate day, and using different levels of additions (as shown in Table 1) there were some differences in moisture, protein, fat, mineral content, and pH of the cheese when measured 1 wk after manufacture (Table 3). However, these differences were assumed not to influence growth of Pa.wWDC04 or production of CO2 during cheese storage.
      Table 3Mean composition of Cheddar cheese made with Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) added to milk and included during salting with either no added substrates or 0.5% galactose (Gal) and 0.5% ribose, 1% or 2% sodium gluconate (NaGLCN) with or without ribose as described in Table 1, compared with cheese made with no Pa.wWDC04 added as an adjunct culture
      Data reported from 3 independent replicates.
      CheeseMoistureMFFS
      MFFS = moisture-on-fat-free-basis.
      FDB
      FDB = fat-on-dry-basis.
      ProteinSalt
      Measured by chloride analysis.
      S/M
      S/M = salt-in-moisture, calculated as salt/(salt + moisture).
      CaPO4
      Calculated as PO4 from measurement of P.
      NaNa/M
      Calculated as Na/(Na+moisture).
      pH
      Measured on 7-d-old cheese.
      No Pa.wWDC0434.753.153.224.61.895.170.741.530.681.925.25
      No added substrate35.454.153.424.71.804.850.751.540.531.495.35
      0.5% Gal + 0.5% ribose35.554.654.323.91.714.600.721.480.601.685.31
      1% NaGLCN36.155.053.824.71.724.540.751.540.641.765.39
      1% NaGLCN + 0.5% ribose36.354.853.124.31.814.760.751.540.501.405.38
      2% NaGLCN36.554.552.123.41.694.420.711.470.601.645.50
      2% NaGLCN + 0.5% ribose37.056.152.523.61.714.420.681.440.681.805.44
      Pooled SE0.320.821.150.440.070.040.020.040.070.230.03
      1 Data reported from 3 independent replicates.
      2 MFFS = moisture-on-fat-free-basis.
      3 FDB = fat-on-dry-basis.
      4 Measured by chloride analysis.
      5 S/M = salt-in-moisture, calculated as salt/(salt + moisture).
      6 Calculated as PO4 from measurement of P.
      7 Calculated as Na/(Na+moisture).
      8 Measured on 7-d-old cheese.

      Moisture

      Cheese composition was in the range for Cheddar cheese with means of 34.7 to 37.0% moisture, 52.1 to 54.3% fat-on-dry basis, and 1.69 to 1.89% salt (Table 3). Moisture was significantly influenced by the treatments as shown in Table 4. Cheese obtained from the regular vat (i.e., No Pa.wWDC04 cheese) had lower moisture (34.7%; P = 0.037) than the 35.4% moisture in the corresponding cheese made with Pa.wWDC04 added to milk but without any substrates added to the curd during salting (i.e., no added substrates cheese). Adding galactose and ribose did not significantly increase moisture (P = 0.80). Adding NaGLCN did increase cheese moisture (P = 0.016), especially when 2% NaGLCN was added (P < 0.001) and the mean moisture content was 36.5%. This was expected as ionic calcium forms a soluble complex with GLCN and lactate ions, resulting in the movement of calcium (and corresponding phosphate) from insoluble calcium phosphate bound to the casein into the soluble ionic form to maintain ionic calcium at its solubility limit in cheese. The lower moisture in the cheese made without adding Pa.wWDC04 was most likely because the curd was salted on the drain table, whereas the other cheeses were salted and substrates added as 10-kg portions placed into plastic tubs as described in Table 1.
      Table 4Probability (P) values from one-way ANOVA of overall cheese treatment effects and contrasts (as described in Table 2) for composition and pH of d-7 cheese
      ParameterOverallAddition of Pa.wWDC04
      Paucilactobacillus wasatchensis WDC04.
      Addition of Galactose and RiboseAddition of GLCN
      Gluconate.
      Level of GLCNAdding 1% NaGLCN
      1% or 2% sodium gluconate
      Adding 2% NaGLCNAddition of RiboseAddition of Ribose with GLCN
      Moisture<0.001
      P < 0.05
      0.037
      P < 0.05
      0.800.016
      P < 0.05
      0.040
      P < 0.05
      0.034
      P < 0.05
      <0.001
      P < 0.05
      0.210.18
      Fat0.005
      P < 0.05
      0.590.290.120.008
      P < 0.05
      0.490.006
      P < 0.05
      0.920.39
      FDB
      FDB = fat-on-dry-basis.
      0.120.730.270.590.043
      P < 0.05
      1.000.140.640.82
      Salt by Cl0.007
      P < 0.05
      0.077
      0.01 > P ≥ 0.05.
      0.069
      0.01 > P ≥ 0.05.
      0.048
      P < 0.05
      0.065
      0.01 > P ≥ 0.05.
      0.540.041
      P < 0.05
      0.770.10
      S/M
      S/M = salt-in-moisture, calculated as salt/(salt+moisture).
      <0.001
      P < 0.05
      0.017
      P < 0.05
      0.061
      0.01 > P ≥ 0.05.
      0.006
      P < 0.05
      0.015
      P < 0.05
      0.140.002
      P < 0.05
      0.860.24
      MFFS0.001
      P < 0.05
      0.054
      0.01 > P ≥ 0.05.
      0.260.150.260.077
      0.01 > P ≥ 0.05.
      0.014
      P < 0.05
      0.033
      P < 0.05
      0.056
      pH<0.001
      P < 0.05
      0.011
      P < 0.05
      0.290.012
      P < 0.05
      0.018
      P < 0.05
      0.140.003
      P < 0.05
      0.310.61
      0.01 > P ≥ 0.05.
      Lactose0.260.130.450.860.730.800.610.360.56
      Galactose<0.001
      P < 0.05
      0.031
      P < 0.05
      <0.001
      P < 0.05
      0.890.230.660.95<0.001
      P < 0.05
      0.75
      Gluconate<0.001
      P < 0.05
      <0.001
      P < 0.05
      0.12
      Calcium0.270.670.310.670.030
      P < 0.05
      0.990.120.290.56
      Sodium0.220.180.360.270.250.720.200.990.51
      Phosphate0.360.900.280.710.044
      P < 0.05
      0.990.160.300.61
      1 Paucilactobacillus wasatchensis WDC04.
      2 Gluconate.
      3 1% or 2% sodium gluconate
      4 FDB = fat-on-dry-basis.
      5 S/M = salt-in-moisture, calculated as salt/(salt+moisture).
      * P < 0.05
      ** 0.01 > P ≥ 0.05.

      Salt and Sodium

      Having a higher moisture content when the curd was salted in tubs (compared with salting on the drain table) suggests there was better whey expulsion when salting the curd on the drain table. This was possibly because better continuous stirring of the curd on the drain table allowed better drainage and removal of whey than in the plastic tubs. Salting the larger mass of curd (∼70 kg) also appeared to allow more salt retention as the cheese salted in the tubs had lower salt content (1.69­–1.81%) compared with 1.89% (Table 3). Just salting in the tubs showed a tendency (P = 0.077) to decrease salt retention with adding galactose and ribose having a tendency (P = 0.069) to further decrease salt retention. The lower salt content in cheeses with added NaGLCN (P = 0.016), and especially with 2% added NaGLCN (P < 0.001) was expected as less salt was added.
      Having the lowest moisture and the highest salt content also means that the cheese made without added Pa.wWDC04 had the highest mean salt/moisture content of 5.17% (P = 0.017). The experimental cheeses had mean salt/moisture contents of 4.42 to 4.85% with no significant effect upon addition of galactose and ribose (P = 0.61), whereas the cheese with 2% NaGLCN was added (P = 0.002). For cheeses made with NaGLCN added to the curd, the amount of salt added was reduced proportionally so that although salt levels as measured by Cl ion concentration were significantly lower (P = 0.048), there was no difference in the sodium levels between the cheese (P = 0.22). There was more variation in the sodium analysis (CV = 10.9%) than in the salt measurement (CV = 3.4%) with mean sodium content of the cheeses ranging from 0.51 to 0.68%. Keeping the Na+ ion level constant is important as the Na+ ion concentration induces stress responses in bacteria upon salting of the curd (
      • McMahon D.J.
      • Oberg C.J.
      • Drake M.A.
      • Farkye N.
      • Moyes L.V.
      • Arnold M.R.
      • Ganesan B.
      • Steele J.
      • Broadbent J.R.
      Effect of sodium, potassium, magnesium, and calcium salt cations on pH, proteolysis, organic acids, and microbial populations during storage of full fat Cheddar cheese.
      ).

      Calcium and Phosphate

      Except for the cheeses with 2% NaGLCN added, there were no overall significant differences in calcium (P = 0.22) and phosphate (P = 0.36) content of the cheeses. Adding 1% NaGLCN did not decrease calcium phosphate levels in the cheese. However, there was a decrease in calcium content when 2% NaGLCN was added; presumably gluconate forms a complex with calcium and lactate (
      • Phadungath C.
      • Metzger L.E.
      Effect of sodium gluconate on the solubility of calcium lactate.
      ), and when 2% NaGLCN was added, there was enough calcium solubilized that some could have been lost with the whey during pressing of the curd.

      Cheese pH

      There were significant differences (P < 0.001) in wk-1 pH of the cheeses (Table 2). The mean pH of regular cheese (No Pa.wWDC04) at 5.24 was lower than the other cheeses (P = 0.011). Adding galactose did not significantly change cheese pH (P = 0.29), whereas adding NaGLCN caused the cheese pH to be higher (P = 0.012) especially when 2% NaGLCN was added (P = 0.003). Compared with cheese A (no added substrate), adding 1% NaGLCN raised the mean cheese pH by 0.04 units to pH 5.39, whereas adding 2% NaGLCN gave a mean pH of 5.45. Rather than any inhibition of starter culture fermentation this appears to be a function of calcium solubilization by GLCN causing a rise in pH. Sodium gluconate forms a neutral solution (pH 7) and together with uptake of H+ ions when calcium phosphate is released from the caseins would result cheese pH increasing. This also corresponds with the cheese with 2% NaGLCN added losing calcium (P = 0.030) and phosphate (P = 0.044) during pressing (Table 2).

      Initial Substrate Levels

      Galactose

      When Cheddar cheese is made using only L. lactis starter cultures there is no residual galactose as L. lactis utilizes both the glucose and galactose moieties of lactose. Except for cheese B, the mean galactose levels were accordingly ≤ 0.03% (Figure 1). In cheese B (which had added galactose) the initial mean galactose was 0.29%. Thus, almost half of the added galactose was not absorbed into the curd and lost during pressing. Retention of galactose in cheese B varied between replicates with values of 0.17, 0.21, and 0.49% and the reason for this was not determined and assumed to be a function of inadvertent differences in handling the curd during salting.
      Figure thumbnail gr1
      Figure 1Galactose content during 16 wk storage at 12°C of a control Cheddar cheese made without adding Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) to the milk and cheeses made with Pa.wWDC04 added and included during salting either no added substrates, 0.5% galactose (Gal) and 0.5% ribose (Rib), or 1 or 2% sodium gluconate (NaGLCN) with or without 0.5% ribose. Data reported as the mean ± SE of 3 independent replicates.

      Lactose

      All of the cheeses had similar initial mean lactose content (P = 0.26) ranging from 0.25 to 0.52%. This was similar to the 0.6% lactose reported for Cheddar cheese by
      • Shakeel-Ur-Rehman
      • Waldron D.
      • Fox P.F.
      Effect of modifying lactose concentration in cheese curd on proteolysis and in quality of Cheddar cheese.
      ,
      • Hou J.
      • Hannon J.A.
      • McSweeney P.L.H.
      • Beresford T.P.
      • Guinee T.P.
      Effect of curd washing on composition, lactose metabolism, pH, and the growth of non-starter lactic acid bacteria in full-fat Cheddar cheese.
      ).

      Gluconate

      Slightly less than half of the GLCN added to the milled curd was retained in the cheese after pressing. For the cheeses that had 1% or 2% NaGLCN added, the mean (± SE) initial level of GLCN in the cheese was 0.38% (± 0.02%) and 0.84% (± 0.04%), respectively (Figure 2). When ribose was also added, the levels tended to be slightly higher (but not significantly different, P = 0.12). No GLCN was detected in any of the cheese without added GLCN, which was expected as GLCN is not a natural component of cheese.
      Figure thumbnail gr2
      Figure 2Gluconate concentration in Cheddar cheese during 16 wk storage at 12°C made with Paucilactobacillus wasatchensis WDC04 added to milk and with 1% or 2% sodium gluconate (NaGLCN) added during salting with or without addition of 0.5% ribose. Data reported as the mean ± SE of 3 independent replicates.

      Cheese Microbiology

      Initial Microbial Levels

      There was no significant difference in mean starter culture counts either between vats (P = 0.75) or based on substrate addition (P = 0.99; Table 5). The pooled mean count of lactococci in the cheese at wk 1 was 1.8 × 108 cfu/g. The amount of added Pa.wWDC04 in the milk was between 2 × 104 and 5 × 104 cfu/mL (Table 6) and during Cheddar cheese manufacture there is a 10-fold concentration of bacteria as milk is concentrated and converted into cheese (
      • Poudel R.
      • Thunell R.K.
      • Oberg C.J.
      • Overbeck S.
      • Oberg T.S.
      • Lefevre M.
      • McMahon D.J.
      Comparison of growth and survival of single strains of Lactococcus lactis and Lactococcus cremoris during Cheddar cheese manufacture.
      ). The level of Pa.wWDC04 in Cheeses A to F was thus expected to be about 105 cfu/g although this was not able to be determined because of interference in enumerating the Pa.wWDC04 colonies due to overgrowth from starter culture lactococci. This was corrected at wk 8 by including vancomycin in the medium to prevent lactococcal growth during the 7-d incubation required for Pa.wWDC04 colonies to be large enough to count on MRS+RV agar plates.
      Table 5Contrast probabilities (P) values from two-way ANOVA (as described in Table 2) for lactococcal starter culture numbers
      Storage time (wk)Addition of Paucilactobacillus wasatchensis WDC04 (X vs. ABCDEF)
      Cheeses A, B, C, D, E, F, and X as described in Table 1.
      Addition of substrates (A vs. BCDEF)
      10.750.99
      40.019
      Indicates significant differences (P < 0.05) during 16 wk storage at 12°C.
      0.48
      80.002
      Indicates significant differences (P < 0.05) during 16 wk storage at 12°C.
      0.21
      12<0.001
      Indicates significant differences (P < 0.05) during 16 wk storage at 12°C.
      0.011
      Indicates significant differences (P < 0.05) during 16 wk storage at 12°C.
      16<0.001
      Indicates significant differences (P < 0.05) during 16 wk storage at 12°C.
      <0.001
      Indicates significant differences (P < 0.05) during 16 wk storage at 12°C.
      1 Cheeses A, B, C, D, E, F, and X as described in Table 1.
      * Indicates significant differences (P < 0.05) during 16 wk storage at 12°C.
      Table 6Pooled mean number of Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) added to milk and final numbers in cheese after storage at 12°C for cheeses described in Table 1 with galactose (Gal), ribose, and sodium gluconate (NaGLCN) added during salting compared with a cheese made with no Pa.wWDC04 added as an adjunct culture
      Data reported from 3 independent replicates
      CheeseTreatmentsMilk
      Cheeses were made using the same batches of milk.
      (cfu/mL)
      Cheese
      Data pooled for cheese sampled at wk 8, 12, and 16.
      (cfu/g)
      No Pa.wWDC04No Pa w.WDC04 added
      Dash indicates not measured but typically <1 × 101 cfu/mL in milk converted into cheese at Utah State University creamery (Ortakci et al., 2015a).
      <1 × 104
      No added substrate 1% NaGLCN 2% NaGLCNPa w.WDC04 added but no ribose added2 × 1041.6 × 107
      0.5% Gal + 0.5% ribose 1% NaGLCN + 0.5% ribose 2% NaGLCN + 0.5% ribosePa.w.WDC04 and ribose added2 × 1042.9 × 108
      1 Data reported from 3 independent replicates
      2 Cheeses were made using the same batches of milk.
      3 Data pooled for cheese sampled at wk 8, 12, and 16.
      4 Dash indicates not measured but typically <1 × 101 cfu/mL in milk converted into cheese at Utah State University creamery (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Growth and gas production of a novel obligatory heterofermentative Cheddar cheese nonstarter lactobacilli species on ribose and galactose.
      ).

      Paucilactobacillus wasatchensis

      Between wk 8 and wk 16, there was no significant effect of storage time on the number of Pa.wWDC04 in the cheeses (P = 0.78) or any interaction between cheese treatment and storage time (P = 0.80). The only significant difference was that adding ribose to the curd resulted in significantly higher numbers of Pa.wWDC04 in cheese (P < 0.001). Consequently, the Pa.wWDC04 numbers were pooled based upon ribose addition as shown in Table 6. That ribose is the preferred substrate for supporting Pa. wasatchensis growth is demonstrated by Pa.wWDC04 being 10-fold higher in numbers by wk 8 when ribose was added (108 cfu/g compared with 107 cfu/g in the other cheeses). This preference for ribose was previously observed by
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      and applies whether the other substrate is galactose or GLCN.
      In the cheese made without adding Pa.wWDC04 to the milk, no Pa. wasatchensis was not detected and remained below 104 cfu/g throughout the 16 wk of storage (Table 6). Low levels of Pa. wasatchensis are known to be present in cheese made in the university's creamery as it was from a blown bag of the university's cheese that Pa.wWDC04, which is the type strain for Pa. wasatchensis, was initially isolated (
      • Oberg C.J.
      • Oberg T.S.
      • Culumber M.D.
      • Ortakci F.
      • Broadbent J.R.
      • McMahon D.J.
      Lactobacillus wasatchensis sp. nov., a non-starter lactic acid bacteria isolated from aged Cheddar cheese.
      ). As yet, it has not been isolated from milk in the creamery but is assumed to be there at <101 cfu/mL, and <102 cfu/g in young cheese. It's entry into the cheese vat is assumed to be environmental as Pa.wWDC04 does not survive pasteurization (
      • McMahon D.J.
      • Bowen I.B.
      • Green I.R.
      • Domek M.J.
      • Oberg C.J.
      Growth and survival characteristics of Paucilactobacillus wasatchensis WDC04.
      ). Numbers of this obligatory heterofermentative nonstarter lactic acid bacteria are usually maintained at a low enough level in the cheese made in this facility to keep unwanted gas production below a level that the bags of cheese become puffy.

      Starter Culture

      Both cheese treatment (P = 0.042) and storage time (P < 0.001) significantly affected starter culture numbers in the cheese with no interaction between time and treatment (P = 0.54). During the 16 wk of storage at 12°C, the number of lactococci cultured from the cheese decreased 3 to 4 logs (Figure 3). The extent of this decrease was greater in cheeses made with Pa.wWDC04 added to the milk than in cheese made with no added Pa.wWDC04. This difference was significant (P = 0.019) by 4 wk and continued through 16 wk (P < 0.001; Table 5). Adding any of the substrates (galactose, ribose or GLCN) to the curd further influenced the decrease in lactococci but significant differences (P < 0.05) were not observed until 12 wk (Table 5). By 16 wk, the mean lactococci numbers in the cheese with 2% NaGLCN and 0.5% ribose added to the curd dropped to 1.1 × 104 cfu/g. In comparison, mean lactococcal numbers were 1.2 × 105 cfu/g for the cheese with no added substrates.
      Figure thumbnail gr3
      Figure 3Lactococcal starter culture numbers during 16 wk storage at 12°C of a control Cheddar cheese made without adding Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) to the milk and cheeses made with Pa.wWDC04 added and included during salting either no added substrates, 0.5% galactose (Gal) and 0.5% ribose (Rib), or 1% or 2% sodium gluconate (NaGLCN) with or without 0.5% ribose. Data reported as the mean ± SE of 3 independent replicates.
      In the cheese with 0.5% galactose and 0.5% ribose added, the mean lactococcal number by wk 16 was 2.6 × 104 cfu/g. Having high numbers of Pa. wasatchensis initially in the cheese, and adding substrates that promote its growth, thus appear to increase the lysis of starter culture bacteria or their conversion into a nonculturable state. This faster drop in lactococcal plate count numbers during storage when high numbers of Pa.wWDC04 are present has previously been observed for both L. lactis starter cultures (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ) and Streptococcus thermophilus starter cultures (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Growth and gas formation by Lactobacillus wasatchensis, a novel obligatory heterofermentative nonstarter lactic acid bacterium, in Cheddar-style cheese made using a Streptococcus thermophilus starter.
      ), especially when elevated storage conditions (12°C) were utilized.

      Observed Headspace Height

      During storage, both cheese treatment and storage time significantly (P < 0.001) affected gas production as measured by height of the headspace above the cheese. There was also a significant cheese by storage time interaction (P < 0.001). When averaged over time, both addition of Pa.wWDC04 to the milk during cheese manufacture and the addition of galactose, ribose or GLCN as substrates had significant effects (P < 0.001) on gas production during cheese storage (Table 7).
      Table 7Contrast probabilities (P) values from 2-way ANOVA (cheese codes as described in Table 2) for gas production in cheese during 16 wk of storage at 12°C
      Cheeses A, B, C, D, E, F, X as described in Table 1.
      ParameterP-value
      Addition of Pa.wWDC04 (X vs. A)Addition of galactose and ribose (A vs. B)Addition of GLCN (A vs. CDEF)GLCN concentration (CD vs. EF)Addition of ribose with GLCN (CE vs. DF)
      Averaged over time
       Headspace height<0.001*0.004*<0.001*0.021*<0.001*
       Volume of CO2<0.001*0.001*0.001*0.084**0.023*
      After 16 wk
       Headspace height<0.001*0.001*<0.001*<0.001*<0.001*
       Volume of CO2<0.001*0.009*<0.001*0.170.001*
      1 Cheeses A, B, C, D, E, F, X as described in Table 1.
      In the control cheese (no added Pa.wWDC04 and without any substrates added), there was no observable gas production during storage (Figure 4A) with the bag tightly held to the surface of the cheese (<6 mm headspace height) even at 16 wk. Just adding 104 cfu/mL of Pa.wWDC04 to the milk during cheese manufacture, without adding any substrates, increased gas production and by 16 wk the headspace height above the cheese was 24 mm. This confirms our previous research (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ) that having a high number of Pa. wasatchensis nonstarter lactobacilli initially in the cheese is a risk factor for unwanted gas production during storage. It would be worthwhile in large scale Cheddar cheese manufacture to monitor NSLAB initially in cheese using a plating medium such as MRS+RV incubated anaerobically at 23 to 30°C to enumerate fast and slow growing lactobacilli.
      Figure thumbnail gr4
      Figure 4Headspace height in cheese bags during 16 wk storage at 12°C of (A) a control Cheddar cheese made without adding Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) to the milk and cheeses made with Pa.wWDC04 added and included during salting either no added substrates, 0.5% galactose (Gal), and 0.5% ribose, or (B) 1% or 2% sodium gluconate (NaGLCN) with or without 0.5% ribose. Data reported as the mean ± SE of 3 independent replicates.
      Adding galactose plus ribose to the Pa.wWDC04-inoculated cheese resulted in a further increase in gas production (P = 0.001; Figure 4A). This increase was apparent after just 4 wk of storage when the mean headspace height was 13 mm. Gas production continued through 12 wk of storage (mean headspace height of 38 mm) and then gas production slowed so at wk 16 the mean headspace height had only increased to 40 mm.
      When 1% NaGLCN was added to the curd, the headspace height above the cheese was not significantly different from that of the Pa.wWDC04-inoculated cheese with no substrates added, for the first 8 wk of storage (Figure 4). However, gas production continued throughout storage with the mean headspace height for the 1% NaGLCN cheese reaching 34 mm by wk 16 compared with only 24 mm when no substrate was added. When 2% NaGLCN was added, gas production further increased, but at 12 wk it was still less than the cheese with galactose and ribose added (31 mm compared with 38 mm, respectively). However, when compared with galactose as the substrate, gas production continued throughout storage when GLCN was added. By 16 wk the 2% NaGLCN cheese was similar to the cheese with galactose and ribose, and had a mean headspace height of 46 mm.
      Adding 0.5% ribose along with GLCN (which increased growth of Pa.wWDC04; Table 6) initially did not cause more gas production. After 8 wk storage, the higher number of Pa.wWDC04 in the GLCN + ribose cheese produced more gas and by 16 wk the headspace height was significantly increased because of ribose addition (P < 0.001). Increasing the amount of GLCN added to the curd also significantly increased gas production (P < 0.001). Cheeses with 1% NaGLCN and 2% NaGLCN added as well as 0.5% ribose had mean headspace heights at 16 wk of 75 and 100 mm, respectively. Unlike when galactose and ribose were added, there was no leveling out of gas production and, based on the slope of the lines in Figure 4, the substrate causing gas production in the GLCN cheeses had not been exhausted. Because all the cheeses with added ribose had a higher number of Pa.wWDC04 present after 8 wk, the difference in gas production was attributed to having more 6-carbon substrate (either galactose or GLCN) being converted into ribulose-5-P for use in the pentose phosphate pathway for energy production resulting in release of CO2 (
      • Oberg C.
      • Sorensen K.
      • Oberg T.
      • Young S.
      • Domek M.
      • Culumber M.
      • McMahon D.
      Gluconate metabolism and gas production by Paucilactobacillus wasatchensis WDC04.
      ).

      Moles of CO2 Produced

      Although measuring headspace height above the cheese is a simple nondestructive method for comparing gas production between vacuum packaged blocks of cheese, it doesn't take into account that not all the CO2 produced is expelled from the cheese. Carbon dioxide is soluble in water and fat and so a portion remains in the cheese up to its solubility limit. Solubility also follows Henry's law and is proportional to the pressure inside the bag. Compared with the
      • Lamichhane P.
      • Sharma P.
      • Kelly A.L.
      • Risbo J.
      • Rattray F.P.
      • Sheehan J.J.
      Solubility of carbon dioxide in renneted casein matrices: Effect of pH, salt, temperature, partial pressure, and moisture to protein ratio.
      study, the aqueous matrix of Cheddar cheese has increased protein (0.66 protein:moisture ratio compared with 0.50) and increased salt (4.5% salt/moisture compared with 3.0%), which would slightly lower CO2 solubility in the nonfat portion of Cheddar cheese to 27 mmol/kg (at 100 kPa). And with an atmospheric pressure of 85.5 the CO2 solubility in the nonfat portion of Cheddar cheese in our study would only be 23 mmol/kg.
      Twice as much CO2 is dissolved in liquid fats compared with solid fats (
      • Truong T.
      • Palmer M.
      • Bansal N.
      • Bhandari B.
      Investigation of solubility of carbon dioxide in anhydrous milk fat by lab-scale manometric method.
      ) with solubility increasing in the liquid phase as temperature is lowered, but because in milk fat more of the triglycerides solidify as cheese is cooled, overall CO2 solubility decreases. Pressure has a similar effect on solubility of CO2 in lipids as it does in the aqueous environments, so solubility decreases in proportion to pressure.
      • Truong T.
      • Palmer M.
      • Bansal N.
      • Bhandari B.
      Investigation of solubility of carbon dioxide in anhydrous milk fat by lab-scale manometric method.
      determined that the solubility index for CO2 in anhydrous milk fat was 39.0 and 11.9 mmol/kg/kPa at 24 and 4°C, respectively. Based upon the solubility of CO2 in anhydrous milk fat determined by
      • Truong T.
      • Palmer M.
      • Bansal N.
      • Bhandari B.
      Investigation of solubility of carbon dioxide in anhydrous milk fat by lab-scale manometric method.
      the milk fat in cheese at 12°C and 85.5 kPa was calculated to be 18.5 mmol/kg. Combining the aqueous and fat portion, CO2 solubility in Cheddar cheese at the atmospheric pressure and storage temperature of 12°C at Utah State University was calculated as 21.7 mmol/kg.
      Because the 900-g cheese blocks were initially vacuum packaged at −8.5 kPa, it was assumed that the initial CO2 solubility was zero and any CO2 already in the cheese was negligible. During storage, any CO2 produced by bacteria in the cheese would remain in the cheese up to its solubility limit (which increases as the pressure inside the bag increases) with any excess CO2 leaving the cheese and being entrapped inside the impermeable plastic bag surrounding the cheese block. We observed that at some measurement times, the pressure inside the bag remained lower than atmospheric pressure and the bag was still held tightly against the cheese block (Figure 5). When the headspace height reached 3 cm above the cheese block, it was observed that this no longer occurred. It was assumed that after enough gas had been produced to generate this much headspace, the pressure inside the bag was the same as the pressure outside the bag. With further CO2 production, the pressure remained the same with the bag being loose around the cheese so just the volume increased. In this 16-wk study, the bags were large enough to accommodate all the gas expelled from the cheese into the headspace without causing the bags to bulge from having higher pressure inside the bags than outside.
      Figure thumbnail gr5
      Figure 5Schematic of gas formation inside vacuum sealed bags of cheese during storage.
      Using a headspace height of 3 cm at which the plastic bag was no longer observed to be pulled tight against the cheese, the headspace volume was calculated by decreasing the width of the headspace proportionally (as shown in Figure 5). Also, because the pressure inside the bag when the headspace height was <3 cm was less than atmospheric pressure, the CO2 solubility in the cheese was likewise decreased in proportion to the headspace height. Based on these calculations, the total amount of CO2 produced in the cheese during storage was then calculated.
      Initially during storage, CO2 was produced faster in the cheese containing galactose and ribose than any of the other cheeses (Figure 6A). After the first 8 wk, this cheese had produced a mean of 20 mmol of CO2. During the last 8 wk, gas production slowed down and by 16 wk reached only a mean of 26 mmol of CO2 being produced, which was still significantly greater (P = 0.009) than in the cheese with no added substrates. In comparison, the cheeses with added GLCN (Figure 6B) did not exhibit any slowing in the rate of CO2 production throughout the 16 wk and reached mean levels up to 37 mmol when 2% NaGLCN and ribose were added.
      Figure thumbnail gr6
      Figure 6Carbon dioxide produced in 900 g of Cheddar cheese during 16 wk storage at 12°C of (A) a control Cheddar cheese made without adding Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) to the milk and cheeses made with Pa.wWDC04 added and included during salting either no added substrates, 0.5% galactose (Gal) and 0.5% ribose, or (B) 1% or 2% sodium gluconate (NaGLCN) with or without 0.5% ribose. Data reported as the mean ± SE of 3 independent replicates.
      Adding ribose along with GLCN to the cheese increased the amount of CO2 produced by 16 wk (P = 0.023). Because of large variations between replicates when both GLCN and ribose were added to the curd (with CV of 8–10%), gas production at 16 wk was not significantly different (P = 0.17) based on GLCN concentration. Although by averaging over time, the contrast based on GLCN concentration was at P = 0.084, so there was a tendency for GLCN concentration to influence gas production, but it was not as significant as adding ribose, although with longer storage the difference may become more apparent. Adding ribose along with GLCN so that Pa.wWDC04 cell concentrations were 10 times higher, resulted in about 35% more CO2 being produced than in the corresponding cheeses without added ribose.

      Substrate Utilization During Storage

      Galactose

      During the first 8 wk of storage, galactose was consistently consumed with the mean galactose level decreasing to 0.06% (Figure 1). With virtually no more galactose available, there was little further gas production in these cheeses after 8 wk (Figure 6A). Because we did not have a method for measuring ribose concentration in the cheese, we were not able to determine its initial level in the cheese or when depletion of ribose occurred.
      Presumably, because ribose is the preferred carbohydrate by Pa. wasatchensis for energy production (
      • Oberg C.J.
      • Oberg T.S.
      • Culumber M.D.
      • Ortakci F.
      • Broadbent J.R.
      • McMahon D.J.
      Lactobacillus wasatchensis sp. nov., a non-starter lactic acid bacteria isolated from aged Cheddar cheese.
      ), then ribose would have been consumed by 8 wk as well. When ribose is present, galactose can be co-utilized for other cellular needs such as cell wall growth (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Growth and gas production of a novel obligatory heterofermentative Cheddar cheese nonstarter lactobacilli species on ribose and galactose.
      ) without any CO2 being produced while still enabling cell multiplication. In a model system, if all the galactose was consumed before ribose was depleted then gas production was not observed (
      • Green I.R.
      • Oberg C.J.
      • Broadbent J.R.
      • Thunell R.K.
      • McMahon D.J.
      Galactose-positive adjunct cultures prevent gas formation by Paucilactobacillus wasatchensis WDC04 in a model gas production test.
      ). Rather, gas production was observed after the presence of ribose had promoted growth of Pa. wasatchensis to high numbers and there was still some galactose remaining. Extensive gas production appears to occur as the cells transition from exponential growth into a stationary phase (
      • Green I.R.
      • Oberg C.J.
      • Broadbent J.R.
      • Thunell R.K.
      • McMahon D.J.
      Galactose-positive adjunct cultures prevent gas formation by Paucilactobacillus wasatchensis WDC04 in a model gas production test.
      ). If the same observations apply in cheese then the observed gas production would indicate both growth of Pa.wWDC04 to high numbers and concomitant depletion of ribose in the cheese.

      Lactose

      About half of the residual lactose in the cheese was depleted during the 16 wk of storage (Figure 7) at which time the pooled mean (± SE) lactose content had dropped to 0.15% ± 0.03%. Paucilactobacillus wasatchensis WDC04 showed only limited growth on lactose (
      • McMahon D.J.
      • Bowen I.B.
      • Green I.R.
      • Domek M.J.
      • Oberg C.J.
      Growth and survival characteristics of Paucilactobacillus wasatchensis WDC04.
      ). Based upon the genome sequence for Pa.wWDC04 (GenBank Accession number NZ_AWTT00000000, https://www.ncbi.nlm.nih.gov;
      • Oberg C.J.
      • Oberg T.S.
      • Culumber M.D.
      • Ortakci F.
      • Broadbent J.R.
      • McMahon D.J.
      Lactobacillus wasatchensis sp. nov., a non-starter lactic acid bacteria isolated from aged Cheddar cheese.
      ), Pa.wWDC04 has a functioning β-galactosidase enzyme but lacks a transport mechanism for bringing lactose into the cell. The presence of residual lactose in the cheese not consumed by the starter culture bacteria during the first week of storage would provide a carbohydrate source for other nonstarter lactobacilli in Cheddar cheese that can utilize lactose such as Lacticaseibacillus paracasei, Latilactobacillus curvatus, and Lactiplantibacillus plantarum (
      • Jordan K.N.
      • Cogan T.M.
      Identification and growth of non-starter lactic acid bacteria in Irish Cheddar cheese.
      ).
      Figure thumbnail gr7
      Figure 7Lactose content during 16 wk storage at 12°C of a control Cheddar cheese made without adding Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) to the milk and cheeses made with Pa.wWDC04 added and included during salting either no added substrates, 0.5% galactose (Gal) and 0.5% ribose (Rib), or 1% or 2% sodium gluconate (NaGLCN) with or without 0.5% ribose. Data reported as the mean ± standard error of 3 independent replicates.

      Gluconate

      Although Pa. wasatchensis can utilize and grow to high cell numbers using GLCN (
      • Oberg C.
      • Sorensen K.
      • Oberg T.
      • Young S.
      • Domek M.
      • Culumber M.
      • McMahon D.
      Gluconate metabolism and gas production by Paucilactobacillus wasatchensis WDC04.
      ), there was little change in GLCN levels for the first 8 wk of storage (Figure 2). The GLCN level in cheese that had 1% NaGLCN added (but no ribose) remained constant through all 16 wk of storage at 0.4%. This implies that the gas production in this cheese resulted from utilization of some other substrate in the cheese and not from GLCN. However, the mean amount of gas produced in this cheese by 16 wk (24 mmol) was higher than that produced in the cheese with no substrates added (9 mmol) and similar to the cheese with galactose and ribose added (26 mmol) as shown in Figure 6.
      For the cheeses with GLCN added, both the substrate addition and storage time significantly (P < 0.001) affected GLCN levels in cheese. There was also a significant interaction between these 2 variables (P = 0.004). Adding ribose along with 1% NaGLCN did result in some utilization of GLCN and lowered (P < 0.001) the GLCN to 0.1% by wk 16. In the cheeses with 2% NaGLCN added to the curd, addition of ribose was not a significant effect and the GLCN levels decreased from an initial 0.9% to 0.7% during the 16 wk of storage.

      Other Substrates

      Based upon the concentrations of galactose and GLCN initially present in the cheeses, and how much remained after the 16 wk of storage it was possible to calculate the amount of CO2 theoretically produced from these substrates (Table 8). Because these 6-carbon substrates must have one carbon removed to enter the pentose phosphate pathway (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Growth and gas production of a novel obligatory heterofermentative Cheddar cheese nonstarter lactobacilli species on ribose and galactose.
      ;
      • Oberg C.
      • Sorensen K.
      • Oberg T.
      • Young S.
      • Domek M.
      • Culumber M.
      • McMahon D.
      Gluconate metabolism and gas production by Paucilactobacillus wasatchensis WDC04.
      ), one mole of CO2 would be produced for each mole of galactose or GLCN metabolized for energy by Pa. wasatchensis. Interestingly, in each cheese there was more CO2 produced than could be attributed to the amount of galactose or GLCN in the cheese. For the cheeses with Pa.wWDC04 added as an adjunct culture, the amount of gas attributed to some other substrate ranged from 14 to 24 mmol (Table 8). Using a generic MW of 190 for this unknown 6-carbon substrate, this represents a concentration in the cheese from 0.3 to 0.5% (wt/wt). Even with the cheese containing no added substrates, 17.5 mmol of CO2 was generated due to substrates already present in the cheese.
      Table 8Predicted and mean (± SE) amount of CO2 (mmol) produced in 900 g of Cheddar cheese after 16 wk of storage at 12°C based upon quantity
      Data reported from 3 independent replicates
      (mmol) of galactose (Gal) and gluconate (GLCN) initially present in the cheese and 1:1 molar ratio production of CO2 when converted into ribulose-5-P by Paucilactobacillus wasatchensis WDC04 (Pa.wWDC04) and an estimated amount of unknown substrate responsible for the excess CO2 production
      Data reported from 3 independent replicates
      CheeseInitial mean level
      Calculated from levels measured in 1-wk-old cheese for each replicate.
      16-wk mean level
      Calculated from levels measured in 16-wk-old cheese for each replicate.
      CO2Unknown substrate
      GalGLCNGalGLCNPredicted
      Calculated from levels measured in 1-wk-old cheese for each replicate.
      Actual
      No Pa.wWDC040.8
      Not detected. *P < 0.05 **0.01 > P ≥ 0.05.
      0.60.23.8 (0.2)3.6
      No added substrate1.41.418.9 (2.0)17.5
      0.5% Gal + 0.5%ribose14.41.912.626.7 (1.0)14.1
      1% NaGLCN1.419.219.11.525.5 (0.5)24.0
      1% NaGLCN + 0.5%ribose1.221.60.14.917.832.8 (2.8)15.0
      2% NaGLCN1.443.036.18.327.6 (0.6)19.3
      2% NaGLCN + 0.5%ribose1.447.70.334.214.637.2 (3.6)22.6
      1 Data reported from 3 independent replicates
      2 Calculated from levels measured in 1-wk-old cheese for each replicate.
      3 Calculated from levels measured in 16-wk-old cheese for each replicate.
      4 Not detected.*P < 0.05**0.01 > P ≥ 0.05.
      Other potential substrates that could account for gas production in the cheese by Pa.wWDC04 include other components of lysed starter culture cells, fermentation of citrate present in the cheese, and decarboxylation of amino acids. We have observed growth of Pa.wWDC04 on cell lysate (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ) and when high numbers of Pa.wWDC04 were present in the cheese there was a more rapid decrease in starter culture numbers in the cheese as previously shown for cheese made using lactococcal (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
      ) or streptococcal (
      • Ortakci F.
      • Broadbent J.R.
      • Oberg C.J.
      • McMahon D.J.
      Growth and gas formation by Lactobacillus wasatchensis, a novel obligatory heterofermentative nonstarter lactic acid bacterium, in Cheddar-style cheese made using a Streptococcus thermophilus starter.
      ) starter cultures. It has not been determined if this is due to some antagonistic action by Pa. wasatchensis or simply a competition for resources when nonstarter lactobacilli are initially present in high numbers in the cheese.
      In a previous study, Pa.wWDC04 was observed to grow on the N-acetylglucosamine component of bacterial cell walls components but did not utilize N-acetylmuramic acid (
      • McMahon D.J.
      • Bowen I.B.
      • Green I.R.
      • Domek M.J.
      • Oberg C.J.
      Growth and survival characteristics of Paucilactobacillus wasatchensis WDC04.
      ). Growth was slightly less than when Pa.wWDC04 is grown with galactose. However, there is not enough bacterial mass in cheese to account for the amount of CO2 produced even if all the carbon was considered to be a 6-carbon substrate [using typical mass of a bacterial cell of 1 × 10−12 g (
      • Liu L.
      The mass of a bacterium.
      ) and 11% of its wet weight being carbon (
      • Romanova N.D.
      • Sazhin A.F.
      Relationships between the cell volume and the carbon content of bacteria.
      )]. In a 900-g block of cheese containing 1010 cells/g, the maximum amount of substrates obtained from cell lysis could be no more than 0.5 g, or 0.25 mmol. This is negligible in comparison to the ∼20 mmol of excess CO2 that was produced in the cheese in this study.
      In Cheddar cheese made at the university creamery, initial approximate organic acids concentrations are 500 mM lactic acid, 25 mM formic acid, 15 mM propionic acid, and 15 mM citric acid (
      • McMahon D.J.
      • Oberg C.J.
      • Drake M.A.
      • Farkye N.
      • Moyes L.V.
      • Arnold M.R.
      • Ganesan B.
      • Steele J.
      • Broadbent J.R.
      Effect of sodium, potassium, magnesium, and calcium salt cations on pH, proteolysis, organic acids, and microbial populations during storage of full fat Cheddar cheese.
      ). However, Pa.wWDC04 does not utilize citrate to form CO2 (Weschler et al., 2021) and production of CO2 from any other organic acids appears unlikely.
      The most likely source of CO2 production by Pa.wWDC04 is decarboxylation of amino acids formed as the casein in cheese is hydrolyzed by residual coagulant along with starter and nonstarter bacterial proteinases and peptidases. Addition of Pa.wWDC04 into milk in the manufacture of Swiss cheese has been found to produce excessive gas during warm room storage and result in elevated levels of biogenic amines (∼28 mmol/kg) mainly as cadaverine and putrescine, as well as some histamine and tyramine (
      • Wechsler D.
      • Berthoud H.
      • Irmler S.
      • Dreier M.
      • Shani N.
      • Guggisberg D.
      • Portmann R.
      • Badertscher R.
      • Loosli F.
      • Bisig W.
      • Bütikofer U.
      • Häni W.
      • Fröhlich-Wyder M.T.
      Formation of biogenic amines by Lactobacillus wasatchensis in experimental Swiss-type cheeses and related opening defects. In IDF International Cheese Science and Technology Symposium. International Dairy Federation.
      ). The substrates thought to produce these amines are lysine, ornithine, histidine, and tyrosine, respectively. As storage time continues, there would be continued hydrolysis of proteins and peptides in cheese leading to continued formation of amino acids as precursors for conversion to amines and continued production of CO2. Growth of nonstarter lactobacilli in cheese occurs through utilization of amino acids as well as cell contents from lysed starter culture (
      • Møller C.O.A.
      • Christensen B.B.
      • Rattray F.P.
      Modelling the biphasic growth of non-starter lactic acid bacteria on starter-lysate as a substrate.
      ).
      Genomic analysis of Pa.wWDC04 found no genes involved in direct decarboxylation of free amino acids, but there are genes involved in the transport and decarboxylation of ornithine (
      • Romano A.
      • Trip H.
      • Lonvaud-Funel A.
      • Lolkema J.S.
      • Lucas P.M.
      Evidence of two functionally distinct ornithine decarboxylation systems in lactic acid bacteria.
      ). Ornithine, however, is not an amino acid coded by DNA and so is not a constituent of any of the caseins but is generated by breakdown of arginine directly by the enzyme arginase or through the arginine deiminase (ADI) pathway (
      • Majsnerowska M.
      • Noens E.E.E.
      • Lolkema J.S.
      Arginine and citrulline catabolic pathways encoded by the arc gene cluster of Lactobacillus brevis ATCC 367.
      ). Curiously, Pa.wWDC04 has no genes encoding the enzymes needed to convert arginine to ornithine, either directly or through the ADI pathway.
      However, starter culture L. lactis strains typically have an active ADI pathway (whereas L. cremoris strains typically do not) and produce citrulline which is then further converted into ornithine and exported from the cell via an arginine/ornithine antiport system to bring arginine into the cell (
      • Noens E.E.
      • Kaczmarek M.B.
      • Żygo M.
      • Lolkema J.S.
      ArcD1 and ArcD2 arginine/ornithine exchangers encoded in the arginine deiminase pathway gene cluster of Lactococcus lactis..
      ). It has been shown that utilization of arginine by lactococci is initiated to produce ATP as the level of lactose is depleted during cheese storage (
      • Ganesan B.
      • Stuart M.R.
      • Weimer B.C.
      Carbohydrate starvation causes a metabolically active but nonculturable state in Lactococcus lactis..
      ). This could make ornithine readily available for further decarboxylation by a nonstarter lactobacilli such as Pa.wWDC04 and result in the gas production observed in this study (Figure 8).
      Figure thumbnail gr8
      Figure 8Putative symbiotic biochemical pathway for the generation of putrescine from arginine by Lactococcus lactis and Paucilactobacillus wasatchensis. Free arginine from the extracellular environment is brought into the L. lactis cell using arcD (arginine/ornithine antiporter) and converted into citrulline and ammonia by arcA (arginine deiminase) then to ornithine and carbamoyl-phosphate by arcB (ornithine carbamoyltransferase), which is exported out of the cell using arcD. The newly generated ornithine is then brought into the Pa. wasatchensis cell using potE (putrescine/ornithine antiporter) and converted into CO2 and putrescine by ODC (ornithine decarboxylase) and exported to the extracellular environment by gntP.

      CONCLUSIONS

      Inclusion of high numbers of the obligatory heterofermentative nonstarter lactobacilli Pa.wWDC04 as an adjunct culture to the milk, along with addition of ribose and GLCN to the cheese curd caused high levels of gas production in cheese during 16 wk of storage at 12°C. Extent of gas production (measured as the height of the headspace above the cheese block) was as follows: addition of GLCN + ribose > addition of GLCN = addition of galactose + ribose > no substrates added. Gas production in all these cheeses was greater than a control cheese made with no addition of Pa.wWDC04 or any substrates. Conversion of the 6-carbon substrates (galactose or GLCN) into a 5-carbon substrate by Pa.wWDC04 and release of CO2 only accounted for about half of the gas produced. We propose that the remaining CO2 comes from decarboxylation of amino acids and production of biogenic amines. Such decarboxylation cannot be completed by Pa.wWDC04 on its own but can occur on a microbial community basis.

      ACKNOWLEDGMENTS

      This research was funded by the BUILD Dairy program of the Western Dairy Center (Utah State University, Logan) with financial support from Dairy West (Meridian, ID) and regional dairy processing companies. We thank the Aggie creamery staff for providing the pasteurized milk and assistance in cheesemaking. This research was also supported by the Utah Agricultural Experiment Station, Utah State University, and approved as journal paper number 9399. The authors have not stated any conflicts of interest.

      REFERENCES

        • Ganesan B.
        • Stuart M.R.
        • Weimer B.C.
        Carbohydrate starvation causes a metabolically active but nonculturable state in Lactococcus lactis..
        Appl. Environ. Microbiol. 2007; 73 (17293521): 2498-2512
        • Goldberg I.
        • Rokem J.S.
        Organic and fatty acid production, microbial.
        in: Schaechter M. Encyclopedia of Microbiology. 3rd ed. Elsevier, 2009: 421-442
        • Green I.R.
        • Oberg C.J.
        • Broadbent J.R.
        • Thunell R.K.
        • McMahon D.J.
        Galactose-positive adjunct cultures prevent gas formation by Paucilactobacillus wasatchensis WDC04 in a model gas production test.
        J. Dairy Sci. 2021; 104 (34275634): 10540-10549
        • Hou J.
        • Hannon J.A.
        • McSweeney P.L.H.
        • Beresford T.P.
        • Guinee T.P.
        Effect of curd washing on composition, lactose metabolism, pH, and the growth of non-starter lactic acid bacteria in full-fat Cheddar cheese.
        Int. Dairy J. 2012; 25: 21-28
        • Jordan K.N.
        • Cogan T.M.
        Identification and growth of non-starter lactic acid bacteria in Irish Cheddar cheese.
        Irish J. Ag. Food Res. 1993; 32: 47-55
        • Lamichhane P.
        • Sharma P.
        • Kelly A.L.
        • Risbo J.
        • Rattray F.P.
        • Sheehan J.J.
        Solubility of carbon dioxide in renneted casein matrices: Effect of pH, salt, temperature, partial pressure, and moisture to protein ratio.
        Food Chem. 2021; 336 (32771897)127625
        • Liu L.
        The mass of a bacterium.
        in: Elert G. The Physics Factbook. 2003
        • Majsnerowska M.
        • Noens E.E.E.
        • Lolkema J.S.
        Arginine and citrulline catabolic pathways encoded by the arc gene cluster of Lactobacillus brevis ATCC 367.
        J. Bacteriol. 2018; 200 (29712874): e00182-e00e18
        • McMahon D.J.
        • Bowen I.B.
        • Green I.R.
        • Domek M.J.
        • Oberg C.J.
        Growth and survival characteristics of Paucilactobacillus wasatchensis WDC04.
        J. Dairy Sci. 2020; 103 (32747094): 8771-8781
        • McMahon D.J.
        • Oberg C.J.
        • Drake M.A.
        • Farkye N.
        • Moyes L.V.
        • Arnold M.R.
        • Ganesan B.
        • Steele J.
        • Broadbent J.R.
        Effect of sodium, potassium, magnesium, and calcium salt cations on pH, proteolysis, organic acids, and microbial populations during storage of full fat Cheddar cheese.
        J. Dairy Sci. 2014; 97 (24913647): 4780-4798
        • Møller C.O.A.
        • Christensen B.B.
        • Rattray F.P.
        Modelling the biphasic growth of non-starter lactic acid bacteria on starter-lysate as a substrate.
        Int. J. Food Microbiol. 2021; 337 (33171308)108937
        • Noens E.E.
        • Kaczmarek M.B.
        • Żygo M.
        • Lolkema J.S.
        ArcD1 and ArcD2 arginine/ornithine exchangers encoded in the arginine deiminase pathway gene cluster of Lactococcus lactis..
        J. Bacteriol. 2015; 197 (26324452): 3545-3553
        • Oberg C.J.
        • Moyes L.V.
        • Domek M.J.
        • Brothersen C.F.
        • McMahon D.J.
        Survival of probiotic adjunct cultures in cheese and challenges in their enumeration using selective media.
        J. Dairy Sci. 2011; 94: 2220-2230
        • Oberg C.J.
        • Oberg T.S.
        • Culumber M.D.
        • Ortakci F.
        • Broadbent J.R.
        • McMahon D.J.
        Lactobacillus wasatchensis sp. nov., a non-starter lactic acid bacteria isolated from aged Cheddar cheese.
        Int. J. Syst. Evol. Microbiol. 2016; 66 (26475452): 158-164
        • Oberg C.
        • Sorensen K.
        • Oberg T.
        • Young S.
        • Domek M.
        • Culumber M.
        • McMahon D.
        Gluconate metabolism and gas production by Paucilactobacillus wasatchensis WDC04.
        J. Dairy Sci. 2021; 104 (34304875): 10586-10593
        • Ortakci F.
        • Broadbent J.R.
        • Oberg C.J.
        • McMahon D.J.
        Growth and gas production of a novel obligatory heterofermentative Cheddar cheese nonstarter lactobacilli species on ribose and galactose.
        J. Dairy Sci. 2015; 98 (25795482): 3645-3654
        • Ortakci F.
        • Broadbent J.R.
        • Oberg C.J.
        • McMahon D.J.
        Late blowing of Cheddar cheese induced by accelerated ripening and ribose and galactose supplementation in presence of a novel obligatory heterofermentative nonstarter Lactobacillus wasatchensis.
        J. Dairy Sci. 2015; 98 (26298753): 7460-7472
        • Ortakci F.
        • Broadbent J.R.
        • Oberg C.J.
        • McMahon D.J.
        Growth and gas formation by Lactobacillus wasatchensis, a novel obligatory heterofermentative nonstarter lactic acid bacterium, in Cheddar-style cheese made using a Streptococcus thermophilus starter.
        J. Dairy Sci. 2015; 98 (26364109): 7473-7482
        • Phadungath C.
        • Metzger L.E.
        Effect of sodium gluconate on the solubility of calcium lactate.
        J. Dairy Sci. 2011; 94 (21943735): 4843-4849
        • Poudel R.
        • Thunell R.K.
        • Oberg C.J.
        • Overbeck S.
        • Oberg T.S.
        • Lefevre M.
        • McMahon D.J.
        Comparison of growth and survival of single strains of Lactococcus lactis and Lactococcus cremoris during Cheddar cheese manufacture.
        J. Dairy Sci. 2022; 105: 2069-2081
        • Romano A.
        • Trip H.
        • Lonvaud-Funel A.
        • Lolkema J.S.
        • Lucas P.M.
        Evidence of two functionally distinct ornithine decarboxylation systems in lactic acid bacteria.
        Appl. Environ. Microbiol. 2012; 78 (22247134): 1953-1961
        • Romanova N.D.
        • Sazhin A.F.
        Relationships between the cell volume and the carbon content of bacteria.
        Oceanol. Acad. Sci. USSR. 2010; 50: 522-530
        • Shakeel-Ur-Rehman
        • Waldron D.
        • Fox P.F.
        Effect of modifying lactose concentration in cheese curd on proteolysis and in quality of Cheddar cheese.
        Int. Dairy J. 2004; 14: 591-597
        • Truong T.
        • Palmer M.
        • Bansal N.
        • Bhandari B.
        Investigation of solubility of carbon dioxide in anhydrous milk fat by lab-scale manometric method.
        Food Chem. 2017; 237 (28764051): 667-676
        • Wechsler D.
        • Berthoud H.
        • Irmler S.
        • Dreier M.
        • Shani N.
        • Guggisberg D.
        • Portmann R.
        • Badertscher R.
        • Loosli F.
        • Bisig W.
        • Bütikofer U.
        • Häni W.
        • Fröhlich-Wyder M.T.
        Formation of biogenic amines by Lactobacillus wasatchensis in experimental Swiss-type cheeses and related opening defects. In IDF International Cheese Science and Technology Symposium. International Dairy Federation.
        http://link.ira.agroscope.ch/en-US/publication/46924
        Date: 2021
        Date accessed: August 24, 2021
        • Wehr H.M.
        • Frank J.F.
        Standard Methods for the Examination of Dairy Products.
        17th ed. America Public Health Association, 2004
        • Zheng J.
        • Wittouck S.
        • Salvetti E.
        • Franz C.M.A. P
        • Harris H.M.B.
        • Mattarelli P.
        • O'Toole P.W.
        • Pot B.
        • Vandamme P.
        • Walter J.
        • Watanabe K.
        • Wuyts S.
        • Felis G.E.
        • Ganzle M.G.
        • Lebeer S.
        A taxonomic note on the genus Lactobacillus: Description of 23 novel genera, emended description of the genus Lactobacillus Beijerinck 1901, and union of Lactobacillaceae and Leuconostocaceae.
        Int. J. Syst. Evol. Microbiol. 2020; 70: 2782-2858