COBISS: 1.08 Agris category code: Q01, U30
THE USE OF ELECTRONIC NOSE TO DIFFERENTIATE ENTIRE MALES FROM IMMUNOCASTRATES (ASSOCIATION WITH LEVELS OF ANDROSTENONE AND SKATOLE)
Carolina PUGLIESE 1 2, Martin SKRLEP 3, Antonella MACAGNANO 4, Emiliano ZAMPETTI 4, Nina BATOREK 3, Etienne LABUSSIÈRE 5, Marjeta CANDEK-POTOKAR 3
ABSTRACT
The aim of this work was to test the ability of the electronic nose (designed and fabricated by CNR-Institute for Microelectronics and Microsystems Rome, Italy) for the differentiation of subcutaneous fat samples from entire or immunocastrated pigs. Seven pigs immunocastrated with Improvac* (IC) and eight entire males (EM) from genotype Pietrain x (Large White x Landrace) were used. After the slaughter, samples of backfat were taken for the determination of androstenone and skatole concentrations. Thighs from experimental pigs (n = 30) were processed to dry-cured ham Kraški pršut. After the "riposo" phase, subcutaneous fat was sampled from thighs for the analysis with electronic nose. The device is based on eight gravimetric sensors (quartz crystal microbalances-QCMs) oscillating at 20 MHz of fundamental frequency and coated with a proper set of organic polymers and macromolecules. Principal Component Analysis (PCA) was used to discriminate IC and EM, based on the signal acquired with electronic nose. The first two components explained 93.5% of variance. The discrimination of EM from IC was good but not flawless, because four EM samples were false negative, but only one pig was false negative for both samples/repetitions. It can be concluded that the results are promising and that the tested electronic nose is able to discriminate fat samples of EM, however, due to the methodological uncertainties (sampling fat after "riposo") results should be considered preliminary and further work is needed.
Key words: boar taint / pig / immunocastration / electronic nose / principal component analysis
1 INTRODUCTION
Surgical castration of piglets is a routine practice in pig production. It is used as a preventive measure to avoid unpleasant smell, the so-called boar taint, which occurs in non-castrated males at the onset of puberty. The two main compounds are held responsible for boar taint, an-drostenone, related mostly to "urine", and skatole, related mostly to "manure" (Gunn et al., 2004, Dijksterhuis et al. 2000). Castration as practiced nowadays (without analgesia/anaesthesia) is presently being questioned in the EU and there is a strong initiative to end it until 2018. One of the alternatives of surgical castration is the so-
called immunocastration, i.e. the vaccination against gonadotropin-releasing-hormone (GnRH), which effectively disrupts the reproductive hormonal axis and thus reduces the levels of compounds (androstenone and skatole) responsible for boar taint (Batorek et al., 2012). Boar taint is perceived by human nose when the meat or fat is heated. Because of the differences in human sensibility and due to non-harmonized methods, the perception of boar taint does not perfectly match the levels of androstenone and skatole concentrations in pork (Bon-neau et al., 1992). To detect boar taint, sensory analysis or chemical determination of responsible compounds can be used; however, chemical analysis is time-consuming,
1	DiSPAA, Animal Science Section. University of Firenze, Via delle Cascine, 5, IT-50144 Firenze
2	Corresponding author, e-mail: carolina.pugliese@unifi.it
3	Agricultural Institute of Slovenia, Hacquetova ul. 17, SI-1000 Ljubljana
4	CNR-Institute for Microelectronics and Microsystems, Via del Fosso del Cavaliere, IT-00100 Roma
5	INRA, UMR 1348 Pegase, F-35590 Saint-Gilles, France; Agrocampus Ouest, UMR 1348 Pegase, F-35000 Rennes
expensive and not adapted for use on the slaughter line. In this respect, the use of the electronic nose for odour detection is of interest. The objective of the study was to test the electronic nose developed by CNR-Institute for Microelectronics and Microsystems Rome, Italy for its ability to differentiate back fat samples of entire from im-munocastrated male pigs.
2 MATERIAL AND METHODS
Immunocastrated (IC, N = 7), and entire males (EM, N = 8) of the genetic line Pietrain x (Large White x Landrace) were used. Experimental pigs come from the trial conducted at INRA UMR 1348 PEGASE (one IC pig died in the course of the experiment). Pigs of IC group received two vaccinations with Improvac®, the first one (VI) at 11 weeks of age and the second one (V2) at 16 weeks of age. Pigs were slaughtered at 24 weeks of age. Backfat was sampled to assess the concentrations of androstenone and skatole (Pauly et al., 2008). One day after slaughter, 30 thighs (16 EM + 14 IC) were harvested and sent to Slovenia to be processed to dry-cured ham "Kraški pršut". After "riposo" phase, the samples of subcutaneous fat were taken from left and right thighs (according to two methods differing in the content of salt) and analysed with an electronic nose designed and fabricated by CNR-Institute for Microelectronics and Microsystems Rome, Italy. Measurements were performed on headspace of sealed bottles (100 mL) that contained 4 g of fat (12 hours after thawing) which were placed on a disk of silicone treated filter paper (Whatman). The electronic nose consisted in eight gravimetric sensors - quartz crystal microbalances-QCMs - oscillating
at 20 MHz of fundamental frequency and coated with a proper set of organic polymers and macromolecules. The sensors recorded a frequency shift related to the mass of volatile compounds absorbed on their surfaces according to Sauerbrey's equation (Af = -Cf x Am, Af -the observed frequency change, Am the change in mass per unit area and Cf the constant factor depending on the crystal used). Sample measurements were carried out at room temperature until a dynamic equilibrium was reached between the sensors and the volatile compounds absorbed and no further frequency change was recorded (steady state). Thirty minutes were adopted to restore the sensors by flowing filtered and dry air throughout the measurement chamber (10 cc) before initiating a new measurement. Cluster analysis based on androstenone and skatole (with SAS 9.2, SAS Institute Inc., Cary, NC, USA) was made to confirm a discrimination between EM and IC. The signals/responses (from eight microbalances) of the electronic nose were analysed by Principal Component Analysis (PCA) using statistical software (PLS Toolbox by Eigenvector Research, Inc., Wenatchee, WA, USA). The responses were calculated as the difference between frequencies during the measurement and cleaning phase. Autoscaling of the data was performed prior to PCA.
3 RESULTS AND DISCUSSION
Cluster analysis (Fig. 1) was used to validate "group status" (IC or EM) based on androstenone and skatole concentration. It can be noted that two groups were well separated, however some subgroups within the main ones could be detected, in particular within EM. One IC
Figure 1: Cluster analysis based on androstenone and skatole concentration 168 Acta agriculturae Slovenica, Supplement 4 - 2013
Figure 2: Scoreplot of PCA from electronic nose measures. Sample labels indicate the level of androstenone (pg/g fat).
pig was also slightly apart from the others with androstenone level (0.47 ^g/g fat) being close to the sensory perception threshold in fat phase (0.5-1.0 ^g/g fat; Haugen, 2012) and is suspicious for being a non-responder.
Discrimination of EM and IC based on the signal acquired by the electronic nose was good (Fig. 2, 3). Figures 2 and 3 show the PCA plots as related to the level of androstenone or skatole, respectively. The first two components (PC1, PC2) explained a very high part of the variance (93.5%). The component PC2 was able to distinguish EM (positive values) from IC (negative values). However, the distinction was not faultless; there were some EM samples that were false negatives, i.e. positioned in the group of IC. However, only one pig
was false negative for both samples/repetitions. It is also worth noting that two EM samples with very high androstenone and skatole concentrations were found on the border of the cloud. On the other hand, there were no false positive cases, i.e. no IC samples positioned in the group of EM. The same observations can be made when considering the levels of skatole (Fig. 3). All the samples, which were above the threshold limit (0.2 ^g/g fat) were classified in the EM group.
It is worth noting that in the present study the samples of fat were taken after salting and "riposo" phase, from the experiment where two salting levels were applied, which could influence the results. Indeed, salt is known to have prooxidant effect on lipids (Min and Ahn,
I (75.76%)
Figure 3: Scoreplot of PCA from electronic nose measurements. Sample labels indicate the level of skatole (pg/g fat).
2005) and because the samples were taken app. three months later than the samples for androstenone and skatole determination this could explain why in some cases, two samples from the same animal were ambiguously classified, once as EM and once as IC. Another explanation could be that other aromatic compounds are involved in boar taint. As suggested by Vestergaard et al. (2006), there are other major compounds present in the head-space of fat samples which can be involved or can interfere with compounds responsible for boar taint. various electronic devices to detect either boar taint or responsible compounds (in particular androstenone) have been tried in and validated in the past (Bourrou-net et al., 1995; Di Natale et al., 2003; Vestergaard et al., 2006; Kirsching et al., 2012). As concluded by Haugen (2006) in his review, the potential of electronic noses is promising, but there is still a need for further research and development.
4	CONCLUSION
Results obtained in the present study show that the electronic nose which was tested has the potential to discriminate EM. However, there were some false negative samples that were classified as IC despite relatively high (0.8-1.2 ^g/g) androstenone levels. The reason might be due to the methodological circumstances, and thus the present results should be considered merely preliminary. Further studies are needed to test the ability of electronic nose to detect boar taint, in particular in view of other (presently unknown) compounds potentially involved or interfering.
5	REFERENCES
Batorek N., Candek-Potokar M., Bonneau M., Van Milgen J. 2012. Meta-analysis of the effect of immunocastration on production performance, reproductive organs and boar taint compounds in pigs. Animal, 6: 1330-1338
Bonneau M., Le Denmat M., Vaudelet J.C., Veloso Nunes J.R., Mortensen A.B., Mortensen H.P. 1992. Contributions of fat androstenone and skatole to boar taint: 1. Sensory attributes of fat and pork meat. Livestock Production Science, 32: 63-80
Bourrounet B., Talou T., Gaset A. 1995. Application of a multi-gas-sensor device in the meat industry for boar-taint detection. Sensors and Actuators B, 26-27: 250-254 Dijksterhuis G.B., Engel B., Walstra P., Font i Furnols M., Ager-hem H., Fischer K., Oliver M.A., Claudi-Magnussen C., Siret F., Béague M.P., Homer D.B., Bonneau M. 2000. An international study on the importance of androstenone and skatole for boar taint: II. Sensory evaluation by trained panels in seven European countries. Meat Science, 54: 261-269 Di Natale C., Pennazza G., Macagnano A., Martinelli E., Pao-lesse R., D'Amico A. 2003. Thickness shear mode resonator sensors for the detection of androstenone in pork fat. Sensors and Actuators B, 91: 169-174 Gunn M., Allen P., Bonneau M., Byrne D.V., Cinotti S., Fre-driksen B., et al. 2004. Welfare aspects of the castration of piglets. Scientific report on the scientific panel for animal health and welfare on a request from the Commission related to welfare aspects of the castration of piglets (Question No. EFSA-Q-2003-091). The EFSA Journal, 91 Haugen J.E. 2006. The use of chemical sensor array technology, the electronic nose, for detection of boar taint. Acta Veterinaria Scandinavica, 48(Suppl 1): S15 Haugen J.E. 2012. Boar taint detection methods: state of the art and critical review. EC meeting on alternatives to surgical castration of male pigs, 17. December 2012, Brussels, Belgium http://ec.europa.eu/dgs/health_consumer/infor-mation_sources/docs/ahw/17122012_haugen_presenta-tion_en.pdf (14. may. 2013) Min B., Ahn D.U. 2005. Mechanism of Lipid Peroxidation in Meat and Meat Products -A Review. Food Science and Biotechnology, 14, 1: 152-163 Vestergaard J.S., Haugen J.E., Byrne D.V. 2006. Application of an electronic nose for measurements of boar taint in entire male pigs. Meat Science, 74: 564-577 Kirsching A, Bázár G., Házás Z., Romvári R. 2012. Classification of meat with boar taint using an electronic nose. Acta Agriculturae Slovenica, Supplement 3: 99-103