Evaluation of autochthonous cultures to improve the cheese flavor: A case study in hard cheese model

Abstract

The characterization of autochthonous cultures based on their contribution to cheese flavor is an additional selection criterion for their use in cheese making. The objective of the present work was to assess the ability of three strains of mesophilic lactobacilli: Lactobacillus casei 72 (Lc72), L. paracasei 90 (Lp90), and L. plantarum 91 (Lp91), one strain of thermophilic lactobacillus: L. helveticus 209 (Lh209), and the thermophilic–mesophilic combinations, to grow and produce aroma compounds in a hard cheese model. Microbiological counts, pH, and the profiles of carbohydrates, organic acids, and volatile compounds were analyzed during incubation for 14 days at 37 ℃. The population of mesophilic lactobacilli reached levels around 8.0 log CFU ml⁻¹ at three days, but then decreased until ∼7.0 log CFU ml⁻¹ toward 14 days. Thermophilic lactobacillus population reached and maintained levels around 7.7 log CFU ml⁻¹ during incubation. Carbohydrates were absent in the hard cheese model, and so no change in the pH values and in the levels of lactic acid was detected. Mesophilic lactobacilli, inoculated individually or in association with Lh209, metabolized the citric acid and produced ethanoic acid. The profiles of volatile compounds of mesophilic lactobacilli (characterized mainly by butan-2-one, 3-hydroxybutan-2-one, 3-methylbutan-1-ol, hexan-1-ol, 2-phenylethanol, and ethanoic acid) were different from the profile of thermophilic lactobacillus Lh209 (characterized mainly by heptan-2-one, ethyl acetate, isoamyl hexanoate, pentan-1-ol, decanoic acid, and 2- and 3-methylbutanal). Cooperative effects in the production of compounds related to cheese flavor, such as 3-hydroxybutan-2-one, ethyl butanoate, ethanol, pentan-2-ol, hexan-1-ol, benzeneacetaldehyde, 2-phenylethanol, and heptanoic acid, were largely evidenced between Lh209 and Lp91; in a lesser extent, cooperative effects were also found for Lh209+Lp90 for the following compounds: 3-hydroxybutan-2-one, isoamyl acetate, and ethanoic acid. Of the mesophilic lactobacilli strains evaluated, Lp91 and Lp90 would be interesting candidates for its use as adjunct cultures in hard cheeses to improve and diversify the flavor.

Keywords

Autochthonous cultures thermophilic and mesophilic lactobacilli hard cheese model volatile compounds profile

Introduction

During cheese ripening, several biochemical events occur, leading to the formation of a great number of flavor compounds which are associated with the enzymatic activities of the microbiota present in the cheese matrix (Steele et al., 2013). Particularly, in the long-ripening hard cheeses, thermophilic lactobacilli used as starter and adventitious mesophilic lactobacilli of non-starter lactic acid bacteria (NSLAB) origin coexist and develop dynamically (Gatti et al., 2014; Gobbetti et al., 2018; Montel et al., 2014). The main role of the starter is to act during cheese production, converting the lactose to lactic acid in few hours, with the consequent pH reduction; in addition, it could also contribute to the production of volatile compounds impacting in sensory characteristics of the cheese (Helinck et al., 2004; Klein et al., 2001). On the other hand, the influence of NSLAB in the cheese ripening is strain-dependent. In previous studies, several strains of NSLAB demonstrated good properties for their use as adjunct cultures in cheeses, because they were able to control the growth of potentially harmful microorganisms and contribute to the development of a mature and improved flavor, especially in cheeses made with pasteurized milk (Gobbetti et al., 2015; Hynes et al., 2003; Irigoyen et al., 2007; Rynne et al., 2007). In particular, the use of selected NSLAB strains with key enzymatic activities involved in the production of volatile compounds is an interesting strategy to improve, diversify, and modulate the cheese flavor (Settanni and Moschetti, 2010). However, in order to assure the usefulness of this strategy, the technological performance of the adjunct culture, the interaction with the starter, and the specific impact on the maturation profiles of cheeses must be previously studied (Sgarbi et al., 2013; Smit et al., 2009; Stefanovic et al., 2017). Some studies of the associative effect between lactococci (used as starter) and mesophilic lactobacilli (used as adjunct culture) in cheese flavor development have been reported (Amárita et al., 2001; Ayad et al., 2000; Kieronczyk et al., 2004), while there are no studies about the cooperation between mesophilic lactobacilli and thermophilic lactobacilli, according to our knowledge.

Many works about the evaluation of the role of the cultures in the cheese flavor development and their impact on the quality have been carried out in cheese made in pilot scale. Although these studies are the most representative, they demand high economic resources and take a considerable time. Alternatively, simpler model systems that mimic the ripening conditions of the cheese have been proposed to evaluate the effect of different cultures in a rapid, economical, and reproducible way (Budinich et al., 2011; Milesi et al., 2011; Sgarbi et al., 2013; Stefanovic et al., 2017).

The objective of the present work was to assess the ability of three strains of mesophilic lactobacilli: Lactobacillus casei 72 (Lc72), L. paracasei 90 (Lp90), and L. plantarum 91 (Lp91), one strain of thermophilic lactobacillus: L. helveticus 209 (Lh209), and the thermophilic–mesophilic combinations, to grow and produce aroma compounds in a hard cheese model. This approach is proposed as a selection step before the application of autochthonous cultures in cheese making.

Materials and methods

Strains

One strain of thermophilic lactobacillus: L. helveticus 209 (Lh209), and three strains of mesophilic lactobacilli: L. casei 72 (Lc72), L. paracasei 90 (Lp90), and L. plantarum 91 (Lp91), belonging to the INLAIN collection, were used. Lh209 was isolated from whey starters provided by Reggianito cheese-producing industries (Reinheimer et al., 1995, 1996) and the technological and biochemical properties have been studied in cheeses and model systems (Candioti et al., 2002; Cuffia et al., 2018; Hynes et al., 2003; Milesi et al., 2011; Perotti et al., 2005). Mesophilic lactobacilli were isolated from good quality Tybo cheeses and they have been characterized regarding their genetic and probiotic properties (Bude Ugarte et al., 2006; Burns et al., 2012). In addition, the technological performance in in vitro assays (salt tolerance, phage resistance, acidifying and proteolytic activities) and in soft and semi-hard cheese models (Briggiler-Marcó et al., 2007; Milesi et al., 2010; Peralta et al., 2014) has also been studied. The aminotransferase (AT) and glutamate dehydrogenase activities of the tested strains were previously determined (Peralta et al., 2016a).

Preparation of hard cheese model

Preparation of the hard cheese model was performed according to Cuffia et al. (2018). Briefly, Reggianito cheeses, made in the pilot plant of our institute, using Lh209 as starter, were processed after three months of ripening with distilled water (1:1 w/v) using a food processor. Then, the suspension was centrifuged at 8000 g/15 min/5 ℃.The aqueous phase was extracted, filtered, and the NaCl content and pH were adjusted at 4% w/v and 5.25, respectively. Finally, it was sterilized by heat (70 ℃ for 30 min). These cheese extracts were used as hard cheese model.

Experimental design

The influence of the lactobacilli strains, individually and in thermophilic–mesophilic combinations, was studied in three independent experiments (E1, E2, and E3) using the hard cheese model. Four treatments were made in each experiment: extracts inoculated with Lh209, extracts inoculated with each of the mesophilic lactobacilli, extracts inoculated with each of the combinations thermophilic–mesophilic lactobacilli, and extracts without inoculation (controls) (Table 1). The level of inoculation of each strain was 10⁵ CFU ml⁻¹. Then, all extracts were incubated at 37 ℃ for 14 days. The experiments were carried out in duplicate using two independent cultures of each strain for inoculation.

Table 1.

Experimental design

Treatment of inoculation	Experiment 1 (E1)	Experiment 2 (E2)	Experiment 3 (E3)
Without inoculation	Control	Control	Control
Thermophilic lactobacilli	Lh209	Lh209	Lh209
Mesophilic lactobacilli	Lc72	Lp90	Lp91
Thermophilic–mesophilic lactobacilli	Lh209+Lc72	Lh209+Lp90	Lh209+Lp91

pH and microbiological analysis

The pH was measured at 3, 7, and 14 days of incubation with a pH meter (Orion 3-star benchtop, Thermo Fisher Scientific Inc., USA) by direct immersion of the electrode.

Counts of mesophilic and thermophilic lactobacilli at 3, 7, and 14 days of incubation were made by spread-plate method (MRS-agar) using 0.1 ml of the decimal dilutions obtained in sterile peptone water. The plates were incubated for 48 h at 37 and 43 ℃, respectively. Thermophilic lactobacilli only grew at 43 ℃, while at 37 ℃ both mesophilic and thermophilic lactobacilli were able to grow, but in this case mesophilic lactobacilli were differentiated from Lh209 according to the different morphology of the colonies: pin-point opaque colonies with small center and transparent halo-shaped border for Lh209 and small, smooth, and creamy colonies for the three strains of mesophilic lactobacilli (Bude Ugarte et al., 2006; Reinheimer et al., 1996).

Carbohydrates and organic acids

The content of organic acids (lactic, citric, ethanoic, and butanoic) and carbohydrates (lactose, glucose, and galactose) in the extracts at 14 days of incubation was analyzed by high-performance liquid chromatography (with the UV and IR detectors in series) according to Peralta et al. (2016b). Samples were centrifuged (5000 g/10 min/10 ℃), and the supernatants were diluted in the mobile phase (1:4 v/v), filtered through 0.45 µm membranes (Millex, Millipore, São Paulo, Brazil) and a volume of 60 µl was injected into the chromatograph.

Volatile compounds

The profiles of volatile compounds were determined at the end of incubation (14 days). Solid-phase microextraction (SPME) was applied as sampling technique, using a divinylbenzene/carboxen/polydimethylsiloxane fiber (DVB/Car/PDMS 50/30 µm, Supelco, USA). Prior to analysis, frozen samples were thawed at 4 ℃ overnight. Aliquots (10 ml) were transferred to a screw-top glass vial (40 ml) containing a micro stirring bar and capped with a cap with a silicon rubber/teflon membrane and a hole enabling SPME sampling. Vial was introduced in an aluminum block placed on a heater/stirrer (IKA, USA) and preheated at 40 ℃ for 10 min with an agitation speed of 250 r min⁻¹. Then, the fiber was exposed to the headspace for 30 min at 40 ℃. Separation, identification, and semi-quantitative analysis of volatile compounds by GC-FID/MS were performed as described by Wolf et al., 2016). Three replicates were analyzed for each sample and the results were expressed as peak areas (arbitrary units).

Statistical analysis

One-way analysis of variance with fixed factors was conducted on the results of pH, carbohydrates, organic acids, and volatile compounds using Minitab 18 software (Minitab Inc., State College, PA, USA). When differences were significant (p < 0.05), the means were compared using Tukey's test. In the case of microbiological analyses, Student's t-test was applied. In addition, volatile compounds were subjected to a multivariate analysis, specifically, principal component analysis (PCA) using XLSTAT 2014.5 software (Addinsoft SARL).

Results and discussion

pH and microbiological analysis

The population of mesophilic lactobacilli either inoculated individually or together with Lh209 increased until 7.7–8.4 log CFU ml⁻¹ on the third day, remained at similar levels at seven days of incubation (except for Lc72 individually inoculated, which grew 0.5 log CFU ml⁻¹), and finally decreased until ∼7.0 log CFU ml⁻¹ toward 14 days (Figure 1). Thermophilic lactobacilli population reached levels of 7.3–8.0 log CFU ml⁻¹ (data not shown), which were maintained during incubation period. No microbial growth was observed in the control extracts. The good survival of the lactobacilli reflects their ability to use alternative energy sources (free fatty acids, short peptides, amino acids (AAs), citrate) instead of simple sugars which were absent in the extracts (Skeie et al., 2008; Stefanovic et al., 2017).

Figure 1.

Evolution of the population of mesophilic lactobacilli during the incubation: extracts inoculated only with the mesophilic lactobacilli (black bars) and extracts with mesophilic–thermophilic combination (grey bars). (A) Lactobacillus casei 72 (Experiment 1), (B) Lactobacillus paracasei 90 (Experiment 2), (C) Lactobacillus plantarum 91 (Experiment 3). Note: Different letters at each sampling time indicate significant differences (p < 0.05).

At three and seven days, counts of Lc72 and Lp90 were significantly higher (p < 0.05) in the extracts inoculated individually with these strains than in the extracts inoculated together with Lh209 (Figure 1(a) and (b)). This could be due to a competitive effect between mesophilic and thermophilic lactobacilli in the first days of incubation, which could indicate that Lh209 would have better capacity to metabolize the nutrients of the medium in comparison to mesophilic lactobacilli (Gatti et al., 2014). However, the final levels of mesophilic lactobacilli were high.

In relation to pH, the values for the inoculated extracts were similar to those of the controls in the three experiments and remained constant during incubation; the values were in the range of 5.02–5.10.

Profiles of carbohydrates and organic acids

Glucose, lactose, and galactose were not detected in any of the extracts. Lactic acid was the predominant (71–74% of the total acids), followed by butanoic (∼18%) and by ethanoic and citric acids (3–7% for both acids) (Table 2). The levels of lactic acid were similar (p > 0.05) in inoculated extracts for the three experiments. The absence of changes in the pH values and lactic acid content is related to the absence of easily fermentable carbohydrates in the extracts, the main substrate for the production of lactic acid by lactic acid bacteria (LAB) (McSweeney, 2011).

Table 2.

Concentration of organic acids in the extracts at the end of incubation (expressed as mg 100 ml⁻¹ of extract)

	Lactic	Citric	Ethanoic	Butanoic
Experiment 1
Control	1290.6 ± 58.4	121.9 ± 1.8a	49.8 ± 1.7c	288.1 ± 10.8b
Lh209	1281.9 ± 33.3	122.7 ± 0.7a	70.1 ± 0.2b	295.3 ± 3.3b
Lc72	1438.5 ± 80.0	70.3 ± 9.6b	141.3 ± 17.0a	378.2 ± 10.4a
Lh209+Lc72	1316.2 ± 90.0	66.4 ± 1.1b	129.1 ± 2.7a	346.4 ± 28.1a
Experiment 2
Control	1288.5 ± 72.7	117.5 ± 2.5a	51.1 ± 2.2c	269.3 ± 9.9b
Lh209	1291.1 ± 29.3	129.4 ± 1.2a	72.2 ± 1.1b	291.9 ± 3.7b
Lp90	1484.2 ± 101.9	73.6 ± 9.2b	132.5 ± 11.3a	360.3 ± 26.7a
Lh209+Lp90	1382.1 ± 27.7	61.7 ± 5.5b	121.9 ± 2.1a	344.9 ± 3.1a
Experiment 3
Control	1293.3 ± 61.1	120.3 ± 2.3a	47.7 ± 2.3c	280.2 ± 11.6
Lh209	1277.8 ± 47.7	120.1 ± 1.3a	74.2 ± 1.4b	289.9 ± 2.9
Lp91	1222.7 ± 137.1	54.9 ± 10.7b	116.4 ± 9.6a	311.9 ± 34.4
Lh209+Lp91	1251.5 ± 158.2	66.4 ± 14.3b	122.0 ± 12.4a	316.5 ± 36.4

Values are means ± standard deviation of two incubation replicas made from independent cultures.

Different letters (a, b, c) in the same column for each experiment indicate the presence of significant differences (p ≤ 0.05).

Lh209 increased the levels of ethanoic acid. This strain did not metabolize citric acid, which was expected because thermophilic lactobacilli are considered citrate-negative (McSweeney, 2011). In contrast, in the extracts inoculated with the mesophilic lactobacilli strains, individually or in association with Lh209, a consumption of this compound was verified, which was correlated with an increase in the content of ethanoic acid. The degradation of citrate with the concomitant production of ethanoic acid by mesophilic lactobacilli strains was reported previously (Sgarbi et al., 2013). In addition, ethanoic acid can also be synthesized via AA catabolism such as serine and aspartic acid (Liu, 2003), which is related with the enzymatic activities found for these strains (Peralta et al., 2016a).

On the other hand, the extracts with Lc72 and Lh209 + Lc72, and Lp90 and Lh209 + Lp90 had higher contents of butanoic acid in relation to control extracts and those inoculated only with Lh209 (Table 2). Butanoic acid is an important flavor compound in cheeses, formed mainly from metabolism of milk fat (Collins et al., 2003), or from AA deamination (Buffa et al., 2004).

Profiles of volatile compounds

A total of 30 compounds—five ketones, five esters, eight alcohols, two aldehydes, four aromatic compounds, and six acids (Tables 3 to 5), which are related to the characteristic flavor of Reggianito cheeses (Wolf et al., 2010)—were identified in all extracts.

Table 3.

Volatile compounds in the extracts of experiment 1 at the end of incubation (expressed as areas in arbitrary units)

Volatile compounds	Treatment
Volatile compounds	Control	Lh209	Lc72	Lh209+Lc72
Ketones
Propan-2-one	45,240 ± 3310	54,900 ± 4533	54,569 ± 702	46,710 ± 3193
Butan-2-one	19,210 ± 2397c	21,357 ± 740c	64,884 ± 4520a	52,569 ± 2826b
Butane-2,3-dione	142,274 ± 4017a	138,708 ± 4918a	109,362 ± 3209b	130,662 ± 8036a,b
Heptan-2-one	103,203 ± 13,850b	154,056 ± 6072a	103,662 ± 1429b	136,285 ± 7069a,b
3-Hydroxybutan-2-one	45,125 ± 1344b	55,448 ± 5494b	65,506 ± 2937a	53,639 ± 2657b
Esters
Ethyl acetate	20,884 ± 2102b	65,118 ± 4023a	18,418 ± 2304b	52,098 ± 5581a
Ethyl butanoate	35,056 ± 837	34,431 ± 5953	34,651 ± 1885	37,122 ± 3727
Isoamyl acetate	37,504 ± 2040a,b	32,727 ± 1840b	35,648 ± 2782a,b	42,424 ± 1499a
Ethyl hexanoate	43,089 ± 5946	25,995 ± 4031	42,121 ± 4315	31,608 ± 1171
Isoamyl hexanoate	19,182 ± 1991c	36,926 ± 1107a	24,060 ± 91b	23,880 ± 67b
Alcohols
Ethanol	31,037 ± 5719	21,140 ± 1730	25,494 ± 497	33,783 ± 4280
Pentan-2-ol	33,675 ± 535b	45,003 ± 374a	24,061 ± 4167c	210,09 ± 2580c
Butan-1-ol	166,073 ± 11,835	129,689 ± 13,756	140,733 ± 1331	126,554 ± 4142
3-Methylbutan-1-ol	20,467 ± 2097b	40,408 ± 1092b	140,706 ± 2993a	136,918 ± 16,819a
Pentan-1-ol	54,436 ± 7262b	138,339 ± 6672a	35,539 ± 6220b	44,567 ± 2506b
Hexan-1-ol	11,488 ± 1734b	12,882 ± 672a,b	18,541 ± 1328a	15,275 ± 1585a,b
2-Ethylhexan-1-ol	50,532 ± 7537a,b	58,460 ± 2810a	57,042 ± 2089a	33,357 ± 4560b
2,3-Butanediol	230,048 ± 5586b,c	195,327 ± 3645c	284,099 ± 8597a	239,316 ± 14,091b
Aldehydes
2-Methylbutanal	98,139 ± 10,893b	156,821 ± 12,452a	22,773 ± 2364c	18,428 ± 921c
3-Methylbutanal	219,466 ± 5066b	272,488 ± 12,958a	72,469 ± 5318c	64,688 ± 6328c
Aromatics
Benzaldehyde	1,655,106 ± 169,257a	189,483 ± 8065b	215,785 ± 10,034b	173,613 ± 14,535b
Benzeneacetaldehyde	72,469 ± 5167b	88,322 ± 9834a,b	102,145 ± 2018a	87,883 ± 3018a,b
2-Phenylethanol	45,392 ± 1210d	79,683 ± 2029c	143,987 ± 3280a	123,762 ± 4979b
Phenol	171,833 ± 8492a	151,028 ± 9484a	118,933 ± 6591b	74,805 ± 1826c
Acids
Ethanoic	116,794 ± 6793c	89,889 ± 2352c	330,498 ± 2440a	242,865 ± 30,024b
Butanoic	1,851,892 ± 63,367b	2,022,358 ± 233,471b	2,506,215 ± 100,020a	2,184,254 ± 73,401a,b
Hexanoic	107,320 ± 7948c	149,048 ± 20,704b	204,204 ± 12,044a	167,793 ± 5739a
Heptanoic	71,365 ± 7505a	64,071 ± 6214a,b	41,991 ± 2342b	79,889 ± 7555a
Octanoic	64,996 ± 1258b	76,898 ± 8943a,b	90,830 ± 865a	83,186 ± 1282a,b
Decanoic	36,853 ± 4467b	81,696 ± 734a	51,615 ± 6838b	49,314 ± 1332b

Values are means ± standard deviation of two incubation replicas made from independent cultures.

Different letters (a, b, c, d) in the same row indicate the presence of significant differences (p ≤ 0.05).

Table 4.

Volatile compounds in the extracts of experiment 2 at the end of incubation (expressed as areas in arbitrary units)

Volatile compounds	Treatment
Volatile compounds	Control	Lh209	Lp90	Lh209+Lp90
Ketones
Propan-2-one	45,322 ± 4120b	55,674 ± 5176b	64,248 ± 1462a	65,739 ± 737a
Butan-2-one	18,111 ± 2117b	22,332 ± 1198b	54,552 ± 5690a	64,568 ± 1719a
Butane-2,3-dione	144,224 ± 4212a	140,121 ± 5546a	139,019 ± 3711a	109,453 ± 1876b
Heptan-2-one	102,109 ± 14,121c	152,430 ± 6395a	146,184 ± 971a,b	108,864 ± 13,533b,c
3-Hydroxybutan-2-one	43,558 ± 1120c	55,551 ± 7222b,c	63,151 ± 4769b	73,787 ± 1570a
Esters
Ethyl acetate	21,120 ± 2407b	66,232 ± 5277a	20,139 ± 2699b	21,238 ± 2117b
Ethyl butanoate	35,126 ± 1109	34,199 ± 4322	44,189 ± 3627	32,760 ± 4020
Isoamyl acetate	37,622 ± 2265b	33,459 ± 2231b	37,908 ± 260b	44,904 ± 736a
Ethyl hexanoate	41,546 ± 6157	24,521 ± 5659	35,063 ± 2777	28,590 ± 3684
Isoamyl hexanoate	19,100 ± 1007c	37,871 ± 1132a	27,274 ± 971b	28,008 ± 1577b
Alcohols
Ethanol	32,421 ± 3612a,b	20,220 ± 2324b	32,939 ± 5128a,b	42,632 ± 520a
Pentan-2-ol	31,220 ± 721b	44,720 ± 549a	38,826 ± 3423a,b	41,058 ± 1961a,b
Butan-1-ol	166,987 ± 13,225	123,037 ± 15,111	172,581 ± 5110	169,324 ± 10,232
3-Methylbutan-1-ol	21,559 ± 1919c	42,387 ± 1138c	432,242 ± 29,827a	310,661 ± 41,136b
Pentan-1-ol	55,832 ± 6160c	141,762 ± 7734a	56,528 ± 979c	92,411 ± 4600b
Hexan-1-ol	11,500 ± 2571b	13,199 ± 954b	48,134 ± 6541a	39,381 ± 2582a
2-Ethylhexan-1-ol	52,473 ± 8540	57,633 ± 4174	52,207 ± 1493	71,556 ± 7040
2,3-Butanediol	241,667 ± 27,811	189,665 ± 31,397	248,314 ± 21,478	278,249 ± 26,308
Aldehydes
2-Methylbutanal	91,231 ± 9112b	150,332 ± 15,897a	31,591 ± 5459c	23,213 ± 1840c
3-Methylbutanal	231,562 ± 5131b	269,796 ± 14,353a	113,884 ± 1268c	68,928 ± 2904c
Aromatics
Benzaldehyde	1,698,543 ± 181,434a	191,288 ± 7321b	205,177 ± 11,108b	178,849 ± 14,024b
Benzeneacetaldehyde	69,125 ± 4298	85,646 ± 7564	88,547 ± 10,998	87,841 ± 2651
2-Phenylethanol	44,127 ± 1555c	76,621 ± 2887b	115,015 ± 8761a	131,643 ± 12,197a
Phenol	168,222 ± 6122a,b	155,333 ± 8966b	136,047 ± 3389b	214,193 ± 20,279a
Acids
Ethanoic	121,690 ± 7751c	87,342 ± 3120c	254,977 ± 4953b	315,400 ± 13,152a
Butanoic	1,872,440 ± 74,565	2,170,290 ± 245,645	2,385,662 ± 27,383	2,488,160 ± 199,858
Hexanoic	106,878 ± 5561b	151,299 ± 25,402b	203,323 ± 567a	207,251 ± 25,894a
Heptanoic	72,387 ± 9207	60,045 ± 8843	71,163 ± 3546	77,095 ± 5474
Octanoic	66,921 ± 1333b	78,662 ± 9117a,b	89,254 ± 45a,b	91,514 ± 8806a
Decanoic	34,324 ± 4142c	82,167 ± 1194a	55,047 ± 2956b	55,553 ± 5951b

Values are means ± standard deviation of two incubation replicas made from independent cultures.

Different letters (a, b, c) in the same row indicate the presence of significant differences (p ≤ 0.05).

Table 5.

Volatile compounds in the extracts of experiment 3 at the end of incubation (expressed as areas in arbitrary units)

Volatile compounds	Treatment
Volatile compounds	Control	Lh209	Lp91	Lh209+Lp91
Ketones
Propan-2-one	41,678 ± 5111c	53,445 ± 5476b,c	65,468 ± 3384a,b	74,981 ± 4707a
Butan-2-one	18,765 ± 3766b	24,543 ± 1210b	63,579 ± 4596a	25,771 ± 3782b
Butane-2,3-dione	139,878 ± 7892a	135,668 ± 5150a	128,740 ± 6346a	67,128 ± 2652b
Heptan-2-one	100,012 ± 11,120b	152,239 ± 6223a	111,902 ± 699b	42,280 ± 832c
3-Hydroxybutan-2-one	43,376 ± 3522b	58,878 ± 7766b	60,259 ± 5056b	122,308 ± 1482a
Esters
Ethyl acetate	18,865 ± 3361c	65,456 ± 5100a	21,055 ± 2583c	43,888 ± 2966b
Ethyl butanoate	37,765 ± 1009b	36,599 ± 6540b	44,085 ± 3847b	130,679 ± 21,514a
Isoamyl acetate	37,025 ± 4323a,b	32,021 ± 2223b	31,440 ± 1509b	43,799 ± 2958a
Ethyl hexanoate	45,650 ± 5181a	26,755 ± 5199b	53,093 ± 6037a	24,597 ± 2637b
Isoamyl hexanoate	18,767 ± 2321b	38,008 ± 1989a	23,929 ± 3729b	31,722 ± 3394a
Alcohols
Ethanol	30,987 ± 4232b,c	23,987 ± 1821b	40,349 ± 1694b	193,509 ± 3590a
Pentan-2-ol	32,567 ± 1002b	48,902 ± 1129b	33,245 ± 2376b	96,378 ± 22,932a
Butan-1-ol	169,656 ± 10,239	127,745 ± 9328	170,357 ± 8832	157,069 ± 6016
3-Methylbutan-1-ol	19,998 ± 2432c	42,981 ± 998b	160,840 ± 3283a	21,084 ± 387c
Pentan-1-ol	55,443 ± 6262b	137,767 ± 5466a	25,934 ± 4399c	58,046 ± 2716b
Hexan-1-ol	10,199 ± 2447c	11,223 ± 423c	32,272 ± 4368b	80,595 ± 644a
2-Ethylhexan-1-ol	52,333 ± 6540c	56,690 ± 1992a,b	65,530 ± 3759a	30,053 ± 988d
2,3-Butanediol	231,215 ± 6532b,c	191,114 ± 2986b	275,769 ± 12,663a	298,897 ± 29,535a
Aldehydes
2-Methylbutanal	95,675 ± 11,876b	160,090 ± 9229a	30,315 ± 580c	24,603 ± 2016c
3-Methylbutanal	222,399 ± 4321b	269,657 ± 13,556a	115,343 ± 5764c	103,458 ± 7863c
Aromatics
Benzaldehyde	1,690,766 ± 12,891a	191,776 ± 7665b	220,437 ± 12,421b	268,747 ± 38,636b
Benzeneacetaldehyde	70,333 ± 4220b,c	89,943 ± 8902b	67,202 ± 9254c	244,895 ± 4435a
2-Phenylethanol	47,742 ± 2133b	82,221 ± 3188b	107,109 ± 7699b	2,062,866 ± 126,964a
Phenol	169,878 ± 7201a	15,745 ± 7890a	86,792 ± 1718b	89,519 ± 17,080b
Acids
Ethanoic	111,546 ± 5539b	90,122 ± 3311c	431,318 ± 2240a	128,491 ± 4087b
Butanoic	189,976 ± 5598a	2,113,278 ± 19,190a	2,399,120 ± 207,877a	87,873 ± 2461b
Hexanoic	102,229 ± 8180a	151,999 ± 3655a	158,622 ± 22,637a	58,674 ± 7698b
Heptanoic	73,454 ± 5567b	63,178 ± 5512b	64,943 ± 2003b	182,789 ± 2007a
Octanoic	62,769 ± 1007	79,833 ± 9120	78,953 ± 9630	74,482 ± 6205
Decanoic	38,821 ± 5547b	85,565 ± 543a	48,413 ± 5691b	45,232 ± 742b

Values are means ± standard deviation of two incubation replicas made from independent cultures.

Different letters (a, b, c) in the same row indicate the presence of significant differences (p ≤ 0.05).

In experiment 1 (Table 3), Lh209 was characterized by the production of compounds such as heptan-2-one, ethyl acetate, isoamyl hexanoate, pentan-1-ol, pentan-2-ol, 2- and 3-methylbutanal, and decanoic acids. The production of volatile compounds by L. helveticus strains has been little studied in comparison with other LAB. However, their ability to produce aroma compounds through different metabolic pathways has been demonstrated in cheeses and in cheese models (Collins et al., 2003; Helinck et al., 2004; Klein et al., 2001). In fact, the formation of 2- and 3-methylbutanal could be correlated with the AT activity toward branched AAs (leucine and isoleucine) (Peralta et al., 2016a). Esterases of LAB can catalyze ester-forming reactions such as esterification and alcoholysis (reactions between medium or short chain fatty acids and partial glycerides with alcohols, respectively) (Holland et al., 2005). The information related to the esters production by L. helveticus strains is scarce. Methylketones are also important flavor compounds, produced by different LAB, including L. helveticus (Klein et al., 2001). Primary alcohols, such as pentan-1-ol, are formed by the reduction of aldehydes derived from AAs and fatty acids, while secondary alcohols, such as pentan-2-ol, are formed from methyl ketones (Collins et al., 2003). Lc72 showed the ability to produce butan-2-one, 3-hydroxybutan-2-one, 3-methylbutan-1-ol, hexan-1-ol, 2,3-butanediol, 2-phenylethanol, benzeneacetaldehyde, and ethanoic, butanoic, hexanoic, and octanoic acids. Contrary, this strain consumed 2- and 3-methylbutanal, both when was inoculated alone or combined with Lh209. This fact showed the ability of Lc72 to metabolize these aldehydes to their respective alcohols (e.g. 3-methylbutan-1-ol) or other volatile compounds. In addition, benzaldehyde was also consumed in the Lc72, Lh209, and Lc72+Lh209 extracts, and phenol in the Lc72 and Lh209+Lc72 extracts. When Lh209 and Lc72 were grown together increased levels of butan-2-one, isoamyl hexanoate, ethyl acetate, 3-methylbutan-1-ol, 2-phenylethanol, and ethanoic and hexanoic acids were found. The levels of these compounds were similar or intermediate to those in extracts inoculated with individual strains. Thus, no cooperative effect was observed in the production of volatile compounds for the combination of Lh209 and Lc72. A cooperative or associative effect takes place when the production of compounds in the extract inoculated with both strains exceeds to that found in the extracts inoculated with the individual strains.

In the experiments 2 and 3, the behavior of Lh209 (Tables 4 and 5) was similar to that described in the E1. Additionally, a diminution of benzaldehyde, and 2- and 3-methylbutanal in extracts with mesophilic lactobacilli (Lp90 and Lp91), inoculated alone and with Lh209, was also observed in these experiments.

In E2, Lp90 had the ability to synthesize propan-2-one, butan-2-one, heptan-2-one, 3-hydroxybutan-2-one, isoamyl hexanoate, 3-methylbutan-1-ol, hexan-1-ol, 2-phenylethanol, and ethanoic, hexanoic, and decanoic acids. These compounds were also increased in the Lh209 + Lp90 extracts, suggesting that Lp90 led the production of volatile compounds when both strains were together. In addition, cooperative effects between Lh209 and Lp90 were found for 3-hydroxybutan-2-one, isoamyl acetate, and ethanoic acid.

In E3 (Table 5), Lp91 produced propan-2-one, butan-2-one, hexan-1-ol, 3-methylbutan-1-ol, ethylhexan-1-ol, 2,3-butanediol, and ethanoic acid. Similar or intermediate levels for propan-2-one, ethyl acetate, and 2,3-butanediol were detected in the Lh209+Lp91 extracts in comparison to extracts inoculated with the individual strains. Cooperative effects were detected for 3-hydroxybutan-2-one, ethyl butanoate, ethanol, pentan-2-ol, hexan-1-ol, benzeneacetaldehyde, 2-phenylethanol, and heptanoic acid.

Among the volatile compounds detected in the extracts, 3-hydroxybutan-2-one has an important role in the flavor development of fermented milk products and it is produced by the reduction of butane-2,3-dione that can be generated via citrate metabolism (Le Bars and Yvon, 2008). In our work, a diminution of the citrate levels and an increase of 3-hydroxybutan-2-one were detected in the extracts inoculated with Lc72 and Lp90, and with the combination of Lh209 + Lp90 and Lh209 + Lp91.

Many authors have reported the production of volatile compounds by strains of L. casei, L. paracasei, and L. plantarum through different mechanisms, mainly associated with AA and citrate metabolism (Gobbetti et al., 2015; Le Bars and Yvon, 2008; Milesi et al., 2010; Pogačic et al., 2015), although the specific metabolic pathways of NSLAB in aroma compounds formation are still poorly understood.

In addition, volatile compounds profiles produced by the individual lactobacilli strains and by the three-pairing thermophilic–mesophilic lactobacilli were analyzed jointly by a multivariate approach.

For that, PCA was applied in order to identify groupings of the samples and visualize associations or differences between strains or combinations of strains, using the correlation matrix (centered and scaled values). A two-PC model representing 78.1% of the cumulative total variance of the system (eigenvalues >1) was selected; seven compounds or variables were eliminated since their communalities were low (<0.75). PC1 represented 51.5%, while PC2 contributed 26.6% of the total variance. As visualized in the scores plot (Figure 2(a)), the samples were grouped in several groups in the plane defined by the two first PC. It was appreciated that the inoculated extracts with individual strains and with the combinations of thermophilic–mesophilic lactobacilli were distinguished perfectly from the controls. Lc72, Lp90, and Lp91 extracts were quite similar and completely different from Lh209 extracts. PC1 separated the samples clearly into two groups: the Lh209+Lp91 extracts were located at the positive end and were characterized by the compounds that contributed positively on this PC (pentan-2-ol, heptanoic acid, ethyl butanoate, benzeneacetaldehyde, 2-phenylethanol, ethanol, 3-hydroxybutan-2-one, and hexan-1-ol); the rest of the samples was grouped at the negative end and were associated with butanoic, hexanoic, octanoic, and ethanoic acids; butane-2-one; butane-2,3-dione; heptan-2-one; and 3-methylbutah-1-ol. Four groups were distinguished from the positive end toward the negative end on the PC2 axis: 1—Lh209; 2—control; 3—Lh209+Lc72 and Lh209+Lp91; and 4—Lc72, Lp90, Lp91, and Lh209+Lp90, respectively. The two branched chain aldehydes (2- and 3-methylbutanal) were positively related to PC2, whereas ethanoic acid, 3-methylbutan-1-ol, and butan-2-one contributed negatively to this PC (Figure 2(b)). Results obtained show that the production of volatile compounds of the three strains of mesophilic lactobacilli inoculated individually or combined with Lh209 was clearly distinguished from the thermophilic lactobacilli; associative effects were mainly observed for Lh209+Lp91.

Figure 2.

Score plot (a) and loading plot (b) (PC1 vs. PC2) of the PCA applied on the volatile compounds of the hard cheese extracts.

Conclusion

The abilities of four autochthonous lactobacilli strains, three mesophilic: Lc72, Lp90, and Lp91, and one thermophilic: Lh209, and the thermophilic–mesophilic combinations, to grow and produce volatile compounds in a hard cheese model, were studied. The profiles of volatile compounds produced by mesophilic lactobacilli and thermophilic–mesophilic combinations were differenced from that of thermophilic lactobacilli. Cooperative effects in the production of some volatile compounds related to cheese flavor were detected, mainly for the combination of Lh209 and Lp91, but also for Lh209 and Lp90. This fact suggests that Lp91 and Lp90 would be interesting candidates as adjunct cultures in hard cheeses to improve and diversify the flavor. The selection of individual strains and the detection of associations between them, based on the production of volatile compounds, represent a potential tool for designing cultures in order to obtain cheeses with distinctive flavor attributes.

Supplemental Material

FST881512 Supplemental Material - Supplemental material for Evaluation of autochthonous cultures to improve the cheese flavor: A case study in hard cheese model

Supplemental material, FST881512 Supplemental Material for Evaluation of autochthonous cultures to improve the cheese flavor: A case study in hard cheese model by Facundo Cuffia, Carina V Bergamini, Érica R Hynes, Irma V Wolf and María C Perotti in Food Science and Technology International

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors acknowledge Consejo Nacional de Investigaciones Científicas y Técnicas (CONICET, Argentina) for the doctoral fellowship of Facundo Cuffia. The present work was financially supported by grants from Universidad Nacional del Litoral (CAI+D No. 501-201-10100082-LI).

References

Amárita

Requena

Taborda

Amigo

Peláez

(2001) Lactobacillus casei and Lactobacillus plantarum initiate catabolism of methionine by transamination. Journal of Applied Microbiology 90: 971–978.

Ayad

Verheul

Wouters

Smit

(2000) Application of wild starter cultures for flavor development in pilot plant cheese making. International Dairy Journal 10: 169–179.

Briggiler-Marcó

Capra

Quiberoni

Vinderola

Reinheimer

Hynes

(2007) Nonstarter Lactobacillus strains as adjunct cultures for cheese making: In vitro characterization and performance in two model cheeses. Journal of Dairy Science 90: 4532–4542.

Bude Ugarte

Guglielmotti

Giraffa

Reinheimer

Hynes

(2006) Nonstarter lactobacilli isolated from soft and semi hard Argentinean cheeses: Genetic characterization and resistance to biological barriers. Journal of Food Protection 69: 2983–2991.

Budinich

Perez-Díaz

Cai

Rankin

Broadbent

Steele

(2011) Growth of Lactobacillus paracasei ATCC 334 in a cheese model system: A biochemical approach. Journal of Dairy Science 94: 5263–5277.

Buffa

Guamis

Saldo

Trujillo

(2004) Changes in organic acids during ripening of cheeses made from raw, pasteurized or high-pressure-treated goats' milk. LWT – Food Science and Technology 37: 247–253.

Burns

Cuffia

Milesi

Vinderola

Meinardi

Sabbag

, et al.(2012) Technological and probiotic role of adjunct cultures of non-starter lactobacilli in soft cheeses. Food Microbiology 30: 45–50.

Candioti

Hynes

Quiberoni

Palma

Sabbag

Zalazar

(2002) Reggianito Argentino cheese: Influence of Lactobacillus helveticus strains isolated from natural whey cultures on cheese making and ripening processes. International Dairy Journal 12: 923–931.

Collins

McSweeney

Wilkinson

(2003) Lipolysis and free fatty acid catabolism in cheese: A review of current knowledge. International Dairy Journal 13: 841–866.

10.

Cuffia

Bergamini

Wolf

Hynes

Perotti

(2018) Characterization of volatile compounds produced by Lactobacillus helveticus strains in a hard cheese model. Food Science and Technology International 24: 67–77.

11.

Gatti

Bottari

Lazzi

Neviani

Mucchetti

(2014) Invited review: Microbial evolution in raw-milk, long-ripened cheeses produced using undefined natural whey starters. Journal of Dairy Science 97: 573–591.

12.

Gobbetti

De Angelis

Di Cagno

Mancini

Fox

(2015) Pros and cons for using non-starter lactic acid bacteria (NSLAB) as secondary/adjunct starters for cheese ripening. Trends in Food Science and Technology 45: 167–178.

13.

Gobbetti

Di Cagno

Calasso

Neviani

Fox

De Angelis

(2018) Drivers that establish and assembly the lactic acid bacteria biota in cheeses. Trends in Food Science and Technology 78: 244–254.

14.

Helinck

Le Bars

Moreau

Yvon

(2004) Ability of thermophilic lactic acid bacteria to produce aroma compounds from amino acids. Applied and Environmental Microbiology 70: 3855–3861.

15.

Holland

Liu

Crow

Delabre

Lubbers

Bennett

, et al.(2005) Esterases of lactic acid bacteria and cheese flavour: Milk fat hydrolysis, alcoholysis and esterification. International Dairy Journal 15: 711–718.

16.

Hynes

Bach

Lamberet

Ogierb

Son

Delacroix-Buchet

(2003) Contribution of starter lactococci and adjunct lactobacilli to proteolysis, volatile profiles and sensory characteristics of washed-curd cheese. Le Lait 83: 31–43.

17.

Irigoyen

Ortigosa

Jaunsaras

Oneca

Torre

(2007) Influence of an adjunct culture of Lactobacillus on the free amino acids and volatile compounds in a Roncal-type ewes-milk cheese. Food Chemistry 100: 71–80.

18.

Kieronczyk

Skeie

Langsrud

Le Bars

Yvon

(2004) The nature of aroma compounds produced in a cheese model by glutamate dehydrogenase positive Lactobacillus INF15D depends on its relative aminotransferase activities towards the different amino acids. International Dairy Journal 14: 227–235.

19.

Klein

Maillard

Thierry

Lortal

(2001) Conversion of amino acids into aroma compounds by cell-free extracts of Lactobacillus helveticus. Journal of Applied Microbiology 91: 404–411.

20.

Le Bars

Yvon

(2008) Formation of diacetyl and acetoin by Lactococcus lactis via aspartate catabolism. Journal of Applied Microbiology 104: 171–177.

21.

Liu S (2003) Practical implications of lactate and pyruvate metabolism by lactic acid bacteria in food and beverage fermentations. International Journal of Food Microbiology 83: 115–131.

22.

McSweeney P. (2011). Biochemistry of cheese ripening. In: Rogisnski H, Fuquay J and Fox P (eds) Encyclopedia of Dairy Sciences. Vol. 1. London: Academic Press, pp.667–675.

23.

Milesi

Bergamini

Hynes

(2011) Production of peptides and free amino acids in a sterile extract describes peptidolysis in hard-cooked cheeses. Food Research International 44: 765–773.

24.

Milesi

Wolf

Bergamini

Hynes

(2010) Two strains of nonstarter lactobacilli increased the production of flavor compounds in soft cheeses. Journal of Dairy Science 93: 5020–5031.

25.

Montel

Buchin

Mallet

Delbes-Paus

Vuitton

Desmasures

, et al.(2014) Traditional cheeses: Rich and diverse microbiota with associated benefits. International Journal of Food Microbiology 177: 136–154.

26.

Peralta

Bergamini

Hynes

(2016a) Aminotransferase and glutamate dehydrogenase activities in lactobacilli and streptococci. Brazilian Journal of Microbiology 41: 741–748.

27.

Peralta

Wolf

Bergamini

Perotti

Hynes

(2014) Evaluation of volatile compounds produced by Lactobacillus paracasei I90 in a hard-cooked cheese model using solid-phase microextraction. Dairy Science and Technology 94: 73–81.

28.

Peralta

Wolf

Perotti

Bergamini

Hynes

(2016b) Formation of volatile compounds, peptidolysis and carbohydrate fermentation by mesophilic lactobacilli and streptoccocci cultures in a cheese extract. Dairy Science and Technology 96: 603–621.

29.

Perotti

Bernal

Meinardi

Zalazar

(2005) Free fatty acid profiles of Reggianito Argentino cheese produced with different starters. International Dairy Journal 15: 1150–1155.

30.

Pogačic

Maillard

Leclerc

Hervé

Chaut

Yee

, et al.(2015) A methodological approach to screen diverse cheese-related bacteria for their ability to produce aroma compounds. Food Microbiology 46: 145–153.

31.

Reinheimer

Quiberoni

Tailliez

Binetti

Suárez

(1996) The lactic acid microflora of natural whey starters used in Argentina on hard cheese production. International Dairy Journal 6: 869–879.

32.

Reinheimer

Suárez

Bailo

Zalazar

(1995) Microbiological and technological characteristics of natural whey cultures for Argentinean hard cheese production. Journal of Food Protection 54: 796–799.

33.

Rynne

Beresford

Kelly

Guinee

(2007) Effect of milk pasteurization temperature on age related changes in lactose metabolism, pH and the growth of non-starter lactic acid bacteria in half-fat Cheddar cheese. Food Chemistry 100: 375–382.

34.

Settanni

Moschetti

(2010) Non-starter lactic acid bacteria used to improve cheese quality and provide health benefits. Food Microbiology 27: 691–697.

35.

Sgarbi

Lazzi

Tabanelli

Gatti

Neviani

Gardini

(2013) Nonstarter lactic acid bacteria volatilomes produced using cheese components. Journal of Dairy Science 96: 4223–4234.

36.

Skeie

Kieronczyk

Eidet

Reitan

Olsen

Østlie

(2008) Interaction between starter bacteria and adjunct Lactobacillus plantarum INF15D on the degradation of citrate, asparagine and aspartate in a washed-curd cheese. International Dairy Journal 18: 169–177.

37.

Smit

Engels

Smit

(2009) Branched chain aldehydes: Production and breakdown pathways and relevance for flavour in foods. Applied Microbiology and Biotechnology 81: 987–999.

38.

Steele

Broadbent

Kok

(2013) Perspectives on the contribution of lactic acid bacteria to cheese flavor development. Current Opinion in Biotechnology 24: 135–141.

39.

Stefanovic

Thierry

Maillard

Bertuzzi

Rea

Fitzgerald

, et al.(2017) Strains of the Lactobacillus casei group show diverse abilities for the production of flavor compounds in 2 model systems. Journal of Dairy Science 100: 1–12.

40.

Wolf

Peralta

Candioti

Perotti

(2016) The role of propionibacteria in the volatile profile of Pategrás cheeses. Dairy Science and Technology 96: 551–567.

41.

Wolf I, Perotti M, Bernal S and Zalazar C. (2010). Study of the chemical composition, proteolysis, lipolysis and volatile compounds profile of commercial Reggianito Argentino cheese: Characterization of Reggianito Argentino cheese. Food Research International 43: 1204–1211.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.04 MB