Aspergillus carbonarius growth and ochratoxin A production on a synthetic grape medium in relation to environmental factors


Sonia Marín, Food Technology Department, University of Lleida, Av. Alcalde Rovira Roure 191, 25198. Lleida, Spain (e-mail:


Aims:  The effects of water activity (0·90–0·99 aw), temperature (15–37°C), and their interaction on growth and ochratoxin A (OTA) production by eight isolates of Aspergillus carbonarius were investigated on synthetic nutrient medium (SNM) with composition similar to grapes.

Methods and Results:  Growth data were modelled by an multiple linear regression and response surface models were obtained. Aspergillus carbonarius grew much faster at 30°C than at the other temperature levels tested and its growth rate increased with increasing aw, maximum growth rate being between 0·95 and 0·99 aw. In general, isolates grew faster at 35–37°C than at 20°C, although no significant differences were found between these temperatures. OTA accumulation was also favoured by high aw levels, and although it was observed in the whole range of temperatures, maximum amounts were detected at 20°C. No OTA was found at the most unfavourable growth conditions.

Conclusions:  Optimum aw level for growth seems to correspond with optimum for OTA production, meanwhile the most propitious temperature for the toxin production was below the best one for growth.

Significance and Impact of the Study:  Prediction of A. carbonarius growth would allow estimating their presence and therefore, the OTA production, as it was found that conditions for the toxin production were more limited than those permitting growth.


Ochratoxin A (OTA) is an isocumarin derivative of the secondary metabolism of different filamentous fungi. It is a genotoxic carcinogen for animals and humans. Sources of this mycotoxin can be vegetal materials (cereals, coffee, beer, wine and fruit juices) as well as food products based on animal tissues (Varga et al. 2001). Moreover, the occurrence of OTA in grapes, raisins, wine and other derivatives has gained increasing attention in the last few years.

Fungi belonging to the genera Aspergillus and Penicillium are described as the main OTA producers. It has been observed that in grapes, Aspergillus section Nigri (black aspergilli) are dominant in the field before harvest and some of its members have been reported to be potential OTA-producing fungi in grapes (Da Rocha Rosa et al. 2002; Sage et al. 2002; Battilani et al. 2003; Magnoli et al. 2003; Bellíet al. 2004a). In particular A. carbonarius seems to be confirmed as the target pathogen, especially because a high percentage (75–100%) of its isolates are able to produce the toxin (Cabañes et al. 2002; Bellíet al. 2005).

Growth of ochratoxigenic species such as A. ochraceus and OTA formation, has been shown to be affected by temperature, water activity (aw) and other factors in different substrates such as cereals (Lee and Magan 2000; Pardo et al. 2004). However, A. carbonarius isolates have been rarely studied, as they were recently described as OTA producers. Preliminary studies with a relatively low number of isolates showed that aw and temperature play an important role in the development and OTA production by these fungi (Mitchell et al. 2003; Bellíet al. 2004b,c).

In this study, eight isolates of A. carbonarius isolated from wine grapes were grown on synthetic media, with composition similar to grapes, at different combinations of aw and temperature to determine the marginal and optimum conditions for mycelial growth and OTA production. A suitable model for prediction of A. carbonarius growth as a function of aw and temperature was also obtained.

Materials and methods


Eight OTA-producing isolates of A. carbonarius obtained from wine grapes from four European countries: Italy (W9, W198), France (W37, W38), Portugal (W89, W104) and Spain (W120, W128) were used in this study. Isolates were supplied by the Faculty of Agriculture (Università Cattolica del Sacro Cuore, Piacenza, Italy); the Institut National Polytechnique de Toulouse (École Nationale Supérieure Agronomique de Toulouse, France); the Departamento Engenharia Biologica (Universidade do Minho, Braga, Portugal); the Departament de Sanitat i d'Anatomia Animals (Facultat de Veterinària, Univ. Autònoma de Barcelona, Spain) and the Departament de Tecnologia d'Aliments (Escola Tècnica Superior d'Enginyeria Agrària, Universitat de Lleida, Spain), where samples of each isolate are held in their culture collection.

Medium and water activity modification

Studies were carried out in vitro using synthetic nutrient medium (SNM) composed of d(+)-glucose, 70 g; d(−)-fructose, 30 g; l(+)-tartaric acid, 7 g; l(−)-malic acid, 10 g; (NH4)2SO4, 0·67 g; (NH4)2HPO4, 0·67 g; KH2PO4, 1·5 g; MgSO4·7H2O, 0·75 g; NaCl, 0·15 g; CaCl2, 0·15 g; CuCl2, 0·0015 g; FeSO4·7H2O, 0·021 g; ZnSO4, 0·0075 g; (+) catechin, 0·05 g; agar, 20 g; water, ca 1000 ml; and pH 4·2 reached with KOH (2 n) (modified from Delfini 1982). This medium simulates grape composition at veraison and its aw was 0·99. SNM was modified to 0·90, 0·93 and 0·95 aw by adding different amounts of glycerol (Bellíet al. 2004c). Determinations were performed using a water activity meter (AquaLab CX-2; Decagon, Pullman, WA, USA). Autoclaved media (20 ml) were poured into 9 cm diameter sterile plastic Petri dishes and used immediately.

Inoculation, incubation and mycelium measurement

Spore suspensions (106 spores ml−1) from each fungal isolate grown on SNM were prepared and used to needle-inoculate the centre of the SNM Petri plates. Plates were incubated at different temperatures (15, 20, 30, 35 and 37°C) in sealed polyethylene bags in order to maintain a constant relative humidity (RH) level. Diameters of the growing colonies were measured daily with the aid of a binocular magnifier, for up to 30 days. Treatments were repeated three times.

OTA analysis

Additional SNM plates were prepared as in the growth experiment and inoculated. OTA was extracted by a variation of Bragulat et al. (2001) method. Three plugs (diameter: 6 mm) were removed from the inner, middle and outer area of the colony, after 7 days of incubation at different combinations of temperature (15, 20, 30, 35 and 37°C) and aw (0·90, 0·93, 0·95 and 0·99) levels. Plugs were introduced in a vial containing 1 ml of methanol. After 60 min, the extracts were shaken and filtered (MillexR SLHV 013NK; Millipore, Bedford, MA, USA) into another vial and stored at 4°C until the analysis by high-performance liquid chromatography with fluorescence detection (Waters 474, Milford, MA, USA) (λexc 330 nm; λem 460 nm). Acetonitrile–water–acetic acid (57 : 41 : 2) (1·0 ml min−1) was the mobile phase and a C18 column (Waters Spherisorb 5 m, ODS2, 4·6 mm × 250 mm) was used. The injection volume was 25 l and the retention time was 7·1 min. The detection limit was 0·02 g OTA g−1 of SNM, based on a signal-to-noise ratio of 3 : 1. The OTA standard was from A. ochraceus (Sigma-Aldrich, Steinheim, Germany). The standard solution was made in methanol and concentration confirmed by using an u.v. spectrophotometer. Treatments were performed in triplicate.

Statistical treatment of the results

The regression lines of colony diameters against days after inoculation were calculated for each aw and temperature combination with Microsoft excel program (version 2002; Microsoft, Munich, Germany) and were used to obtain the growth rate under each treatment conditions.

Temperature and aw effects on mycelial growth and OTA production were statistically analysed with SAS software (version 8.02; SAS Institute, Inc., Cary, NC, USA) by analysis of variance followed by either LSMEAN or Student–Newman–Keuls multiple-range tests. Statistical significance was judged at P < 0·0001.

Polynomial multiple linear regressions (MLR) and the resulting response surface models (RSM) were obtained for prediction of A. carbonarius growth rates, with the Unscrambler software (version 7.6; CAMO ASA, Oslo, Norway), including the significant factors, interactions and quadratic terms.


Temperature and aw influence on mycelial growth

Analysis of variance revealed that the factors aw and temperature and their interaction, had a significant influence on mycelial growth of the eight A. carbonarius isolates tested (P < 0·0001). No significant differences among the isolates tested were observed (data not shown).

The combined effects of aw and temperature on growth rates (mm day−1) are shown in Table 1. Optimum temperature for growth was 30°C for all aw levels tested, except for isolate W9 at 0·99 aw, W37 at 0·90 and 0·99 aw and W120 at 0·93 aw, where growth was faster at 35°C than at 30°C, and W128 and W198 at 0·99 aw, where maximum growth was detected at 20°C.

Table 1.  Growth rates (mm day−1) of eight Aspergillus carbonarius isolates at different aw and temperatures on synthetic nutrient medium ± s.e.
IsolateawGrowth rates (mm day−1)
  1. NG, no growth.

W90·90NG0·84 ± 0·092·86 ± 0·082·00 ± 0·061·49 ± 0·06
0·930·65 ± 0·073·03 ± 0·095·37 ± 0·193·92 ± 0·173·42 ± 0·21
0·952·09 ± 0·034·26 ± 0·068·76 ± 0·225·79 ± 0·124·45 ± 0·16
0·992·86 ± 0·095·61 ± 0·207·35 ± 0·167·69 ± 0·275·69 ± 0·21
W370·90NG0·94 ± 0·071·73 ± 0·312·21 ± 0·051·53 ± 0·09
0·930·34 ± 0·092·05 ± 0·125·05 ± 0·173·18 ± 0·423·42 ± 0·09
0·951·52 ± 0·034·03 ± 0·087·75 ± 0·195·68 ± 0·165·24 ± 0·15
0·992·76 ± 0·103·63 ± 0·166·08 ± 0·266·74 ± 0·415·87 ± 0·31
W380·90NG0·71 ± 0·041·47 ± 0·121·27 ± 0·140·91 ± 0·09
0·930·42 ± 0·072·53 ± 0·084·96 ± 0·243·74 ± 0·073·24 ± 0·16
0·951·81 ± 0·044·11 ± 0·047·02 ± 0·315·99 ± 0·224·81 ± 0·18
0·992·62 ± 0·095·94 ± 0·167·93 ± 0·247·78 ± 0·315·84 ± 0·17
W890·90NG0·78 ± 0·101·94 ± 0·131·16 ± 0·070·93 ± 0·09
0·930·25 ± 0·062·36 ± 0·073·78 ± 0·233·11 ± 0·133·44 ± 0·15
0·951·86 ± 0·044·15 ± 0·076·75 ± 0·364·75 ± 0·154·70 ± 0·16
0·992·86 ± 0·105·00 ± 0·256·83 ± 0·136·42 ± 0·344·32 ± 0·11
W1040·90NG0·94 ± 0·052·63 ± 0·101·34 ± 0·121·32 ± 0·08
0·930·69 ± 0·082·49 ± 0·114·99 ± 0·193·74 ± 0·273·53 ± 0·17
0·952·01 ± 0·024·32 ± 0·067·65 ± 0·144·66 ± 0·154·84 ± 0·22
0·992·50 ± 0·105·63 ± 0·127·44 ± 0·375·55 ± 0·645·11 ± 0·13
W1200·90NG1·02 ± 0·112·47 ± 0·102·43 ± 0·152·21 ± 0·13
0·930·77 ± 0·062·98 ± 0·074·51 ± 0·094·84 ± 0·104·31 ± 0·09
0·952·13 ± 0·055·20 ± 0·077·75 ± 0·165·96 ± 0·234·49 ± 0·12
0·992·90 ± 0·157·02 ± 0·2410·1 ± 0·416·05 ± 0·255·72 ± 0·20
W1280·90NG0·60 ± 0·061·88 ± 0·101·20 ± 0·080·82 ± 0·04
0·930·50 ± 0·071·91 ± 0·075·38 ± 0·192·59 ± 0·121·94 ± 0·21
0·951·53 ± 0·033·90 ± 0·058·29 ± 0·104·36 ± 0·363·39 ± 0·12
0·993·37 ± 0·127·42 ± 0·336·54 ± 0·155·83 ± 0·164·84 ± 0·16
W1980·90NG0·63 ± 0·101·92 ± 0·171·11 ± 0·121·36 ± 0·13
0·931·12 ± 0·033·17 ± 0·045·90 ± 0·233·67 ± 0·173·87 ± 0·04
0·952·32 ± 0·024·45 ± 0·118·01 ± 0·205·49 ± 0·194·44 ± 0·08
0·992·91 ± 0·117·13 ± 0·296·80 ± 0·195·69 ± 0·244·82 ± 0·10

Growth rates presented a peak at 30°C at 0·95 aw, and at 0·99 aw in the case of W38, W89 and W120 isolates, being superior of 10 mm day−1 for this last one.

No significant differences were found between 20, 35 and 37°C, although in general, isolates grew faster at 35 and 37°C than at 20°C.

Growth at 0·95 and 0·99 aw was significantly higher than at 0·93 aw, and this one significantly higher than at 0·90 aw, where little growth was detected in few of the isolates. Minimum growth rates (<4 mm day−1) were obtained at 15°C regardless of aw level and isolates tested. No growth was observed at this temperature at 0·90 aw, and it was very low (<1·5 mm day−1) at 0·93 aw.

The proposed model that allows predicting growth rates of A. carbonarius is shown in Fig. 1. It is suitable for any A. carbonarius isolate, as no significant differences were found between the eight isolates. The percentage of variance explained by the model (R2 coefficient) varied between 0·86 and 0·96.

Figure 1.

Response surface contour plot showing the effect of aw and temperature on the growth rates (mm day−1) of Aspergillus carbonarius (W128) isolate, on synthetic nutrient medium (R2 = 0·86)

Temperature and aw influence on OTA production

All single factors (aw, temperature and isolate) and the interaction isolate × temperature had a significant effect on OTA production (Table 2).

Table 2.  Analysis of variance of the effects of aw and temperature on ochratoxin A production by eight isolates of Aspergillus carbonarius after 7 days growing on synthetic nutrient medium
FactorsMean squareF-value
  1. *Significant (P < 0·0001); ns, notsignificant.

aw × T6·571·07ns
Isolate × T42·426·89*
aw × Isolate4·490·73ns
aw × Isolate × T6·711·09ns

Maximum amounts of OTA were detected at 0·95 aw for W120 and at 0·99 aw for the other isolates tested (Table 3). OTA production decreased with decreasing aw, although no significant differences were found between 0·95 and 0·93 aw and between 0·93 and 0·90 aw. No OTA was detected at 0·90 aw and only two isolates (W37 and W120) produced low amounts at 0·93 aw at 20°C.

Table 3.  Mean ochratoxin A concentrations (μg g−1) produced by eight isolates of Aspergillus carbonarius incubated at different aw and temperatures on SNM for 7 days
IsolateawTemperature (°C)
  1. NG, no growth; DL detection limit (0·02 g OTA g−1 SNM); SNM, synthetic nutrient medium.

  2. *Mean value of three replicates.

0·95NG0·19 ± 0·160·12 ± 0·21<DL<DL
0·99<DL0·57 ± 0·500·18 ± 0·31<DL<DL
0·93NG0·22 ± 0·38<DL<DL<DL
0·95NG0·25 ± 0·43<DL<DL<DL
0·992·28 ± 3·954·95 ± 7·321·47 ± 1·780·47 ± 0·81<DL
0·95NG6·09 ± 6·340·10 ± 0·19<DL<DL
0·993·66 ± 4·179·13 ± 2·141·88 ± 0·952·03 ± 1·923·11 ± 1·81
0·95NG0·71 ± 0·76<DL<DL<DL
0·99<DL0·21 ± 0·130·77 ± 0·381·52 ± 2·641·09 ± 1·55
0·95NG<DL0·03 ± 0·06<DL<DL
0·99<DL0·91 ± 1·130·37 ± 0·37<DL0·96 ± 1·66
0·93NG14·00 ± 19·990·03 ± 0·03<DL<DL
0·95NG20·49 ± 7·071·70 ± 0·08<DL<DL
0·9914·25 ± 4·2510·60 ± 8·5110·36 ± 6·421·75 ± 1·532·60 ± 0·94
0·95NG2·21 ± 1·95<DL<DL<DL
0·99<DL2·55 ± 0·731·84 ± 2·291·79 ± 3·101·17 ± 1·41
0·95NG0·08 ± 0·13<DL<DL<DL
0·99<DL<DL1·69 ± 1·461·21 ± 1·78<DL

OTA was detected under the whole range of temperatures tested, although at 15°C only three isolates (W37, W38 and W120) produced OTA at 0·99 aw. At 35 and 37°C, OTA was only detected at the highest aw level tested. Temperature was statistically significant only for the higher OTA-producing isolates (W38 and W120), where a peak production of OTA occurred at 20°C followed by 15°C. Significantly less OTA was found at 30, 35 and 37°C, with no significant differences among them. For the other isolates tested and in general, 20°C seems also the optimum temperature for producing OTA after 7 days of incubation, although no significant differences were observed between temperatures.

The highest mean OTA concentrations of 9·13 g g−1 SNM at 0·98 aw and 20°C, and of 20·5 g g−1 SNM at 0·95 aw and 20°C were detected for W38 and W120 isolates respectively. The lowest OTA producers were the Italian (W9, W198) and the Portuguese (W89, W104) isolates, producing <2 g g−1, meanwhile the amount of OTA produced by W37 and W128 isolates was up to 5 g g−1.


The present study has focused on A. carbonarius, the main fungal species responsible for OTA accumulation in grapes, and puts the emphasis on the effect of ecological factors on the toxin production as a measure to predict further contamination in the field.

Aspergillus carbonarius growth was found to occur optimally at 30°C. This temperature was previously reported as the temperature that allows maximum growth rates for four A. carbonarius isolates (Bellíet al. 2004b). Similarly, they found that A. carbonarius grew faster at higher aw (0·98–0·995), whereas no significant differences in growth rates were detected at 0·93 and 0·95 aw. Other studies suggested that the optimum temperature for A. carbonarius growth was between 25 and 30°C (Leong et al. 2004) or between 25 and 35°C, depending on the isolate, but most of the cases growing faster at 35°C than at 25°C (Mitchell et al. 2003). Contrary to us, they found a different behaviour of the isolates tested in terms of aw, with optimum varying from 0·93 to 0·987 aw. However, in most of the cases, optimum was 0·98 aw, similarly to our results. The greatest tolerance to low aw in all the isolates studied was found at temperatures close to the optimum ones.

The range of temperatures allowing A. carbonarius growth proposed in other studies is similar to the one showed in this one (15–37°C). In general, growth at temperatures below 15°C is not common and it is only possible at the highest aw levels (Mitchell et al. 2003; Bellíet al. 2004b). Aspergillus carbonarius is sensitive to higher temperatures compared with the A. niger aggregate (Bellíet al. 2004b), showing much reduced growth at 42°C (Leong et al. 2004).

Provided that aw of grapes is 0·95–0·99 and mean temperature levels in the vineyards from June to September may vary between 20 and 26°C, and ca 70 days with maximum temperature higher or equal to 30°C (INM 2003), ca 60% of the time in the period preceding harvest (4 months), would give suitable conditions for A. carbonarius growth and therefore for OTA production.

In this study, OTA was analysed after an incubation period of 7 days, as it is usually enough to allow OTA detection, although some A. carbonarius isolates required more days to reach maximum production in YES and CYA (Bragulat et al. 2001). Five days resulted in the maximum OTA accumulation for A. carbonarius growing on SNM, dropping off after 10 days, but no information was given between these two incubation periods (Bellíet al. 2004c).

Optimum temperature for OTA production was observed at 20°C for most of the isolates. Mitchell et al. (2003) found that OTA production by A. carbonarius was greater at 25°C than at 20°C. Similar to us, they reported that more toxin was produced at higher aw levels.

High RH and temperature levels in the field during the months preceding harvest could also favour OTA production in grapes. Moreover, other factors such as the situation of the vineyards, meteorological extreme conditions like hailstone or the fluctuation of the environmental factors between day and night, could also influence positively or negatively in the final OTA content in wine grapes, although their effect is not known yet.

Testing isolates of different countries of origin may have explained the different adaptation of the isolates to different climatic conditions. However, no significant conclusions have been drawn about this aspect from the results, as all isolates behaved similarly in terms of growth and in OTA production at the aw and temperature conditions tested, although the minimum amounts of OTA were detected in the Italian and Portuguese isolates.

Predicting the growth of A. carbonarius could be important to estimate the production of OTA by these species, as the conditions for OTA formation are more restricted than those for growth, as reported before for other mycotoxins (Marín et al. 1999). Therefore, if growth is restricted, OTA will not be present.

Finally, further investigations are required to determine the ability of A. carbonarius to produce OTA when growing on natural substrates as grapes, instead of on SNM, to observe if they behave similarly or the penetration of the grapes skin pose a hurdle to them in an approach to better understand growth conditions and OTA production by these fungi.


The authors acknowledge financial support from the Catalonian Government (Direcció General de Recerca, Generalitat de Catalunya), the Spanish Government (CICYT, Comisión Interministerial de Ciencia y Tecnología, AGL 2001 2974-C05–02 project) and the EC, Quality of Life Programme (QoL), Key Action 1 (KA1) on Food, Nutrition and Health (QLRT-2000–01761) project.