Species identification of raw and cooked bivalves using electrophoresis
Monique ETIENNE 1 *, Marc JÉRÔME, Joël FLEURENCE
RÉSUMÉ Identification d’espèce de bivalves crus et cuits par électrophorèse.
Trois techniques électrophorétiques, l’isoélectro-focalisation (IEF), l’électropho- rèse en gel de polyacrylamide-sodium dodecyl sulfate (PAGE-SDS) et l’IEF- urée , ont été utilisées pour l’identification d’espèce de bivalves : des
« baby-clams » du Vietnam (Meretrix lyrata), des palourdes croisées d’Europe (Ruditapes decussatus) et des coques communes (Cerastoderma edule). Les analyses ont été réalisées à partir de protéines extraites de différentes parties des mollusques, les muscles adducteurs, le manteau et le pied, provenant d’échantillons crus et cuits. Les muscles adducteurs ont donné les profils les plus discriminants. Les trois méthodes testées peuvent être utilisées pour l’identification des bivalves, mais en utilisant l’IEF, après cuisson, les profils électrophorétiques étaient modifiés tandis qu’avec le SDS-PAGE ou l’IEF-urée, les profils des échantillons crus et cuits étaient identiques. En conclusion, les deux techniques, SDS-PAGE et urée IEF, réalisées à partir des muscles adduc- teurs, crus ou cuits, ont donné des spectres, spécifique de l’espèce, permet- tant l’authentification des échantillons.
Mots clés : identification d’espèce, bivalve cuit, SDS-PAGE, urée IEF, protéine.
SUMMARY
Three electrophoretic techniques, isoelectric focusing (IEF), sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and urea-IEF were used for species identification of “baby-clams” from Vietnam (Meretrix lyrata), grooved carpet shell (Ruditapes decussatus) and the common edible cockle (Cerastoderma edule). Analysis were performed on proteins extracted from dif- ferent parts of the mollusks, adductor muscle, mantle, foot, of raw and cooked samples. The electrophoretic patterns of adductor muscle were the most dis- criminative. Although all three methods tested could be used for bivalve spe- cies identification, the electrophoretic profiles with IEF were modified after
1. Laboratoire biochimie des protéines et qualité, Département valorisation des produits, Ifremer, BP 21105, 44311 Nantes cedex 3, France.
* Correspondence
Monique.Etienne@ifremer.fr
cooking, whereas those with SDS-PAGE or urea-IEF were identical in raw and cooked samples and thus more suitable for sample authentication.
Key-words: species identification, cooked bivalve, SDS-PAGE, urea-IEF, pro- tein.
1 - INTRODUCTION
Intact fishery products can be identified by the morphological criteria clearly described by taxonomists, or from the genomic expression of sarcoplasmic proteins characteristic of a species, or directly from the DNA analysis (SOTELOet al., 1993; TORO, 1998; BOSSIER, 1999). Two main procedures can be used for fish species identification when the animal is not intact: protein analysis and molecular biology methods.
Electrophoretic separation of raw muscle proteins is a well-established pro- cedure for fish and shellfish identification (REHBEIN, 1990; MACKIE, 1997), provi- ding distinctive profiles which can then be compared with those of authentic species obtained using the same protein extraction, electrophoretic conditions and staining procedure (LUNDSTROM, 1980; DURANDet al., 1985; REHBEINet al., 1995). Analytical methods for cooked fish have been described in the literature (MACKIE, 1980; SCOBBIE et al., 1988; REHBEIN et al., 1992; CRAIG et al., 1995), and two methods have recently been standardised: sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE) followed by silver staining (PIÑEROet al., 1999) and urea-IEF with Coomassie staining (ETIENNEet al., 1999).
These methods, which were validated by a collaborative exercise (REHBEIN et al., 1999), can be applied to the meat of finfish, shark, squids and shrimps.
Despite the increasing number of processed “clam” species on the market, only few electrophoretic tests have been developed for species identification of bivalves (CODURI, 1972; SECCHI et al., 1982; WIGGIN et al., 1983; KITTS et al., 1997). In the context of an inquiry by the European Commission to determine whether electrophoresis can be used to identify species from frozen cooked bivalve meat, experiments were performed on raw and cooked bivalve samples using methods established and standardised in EC projects (FAR UP 3-783 and FAIR CT 95-1227).
After preliminary studies showed that the adductor muscle was more sui- table for analysis than the mantle or foot, electrophoretic profiles were obtained from three bivalve species and compared before and after cooking to determine whether it was possible to identify processed (cooked) samples.
2 - MATERIALS AND METHODS
2.1 Bivalve samples
The species studied were (i) Meretrix lyrata, or “baby clams” from Vietnam, provided by the Vietnamese Ministry of Fisheries as two batches (1 kg each) of frozen raw baby clams (shell on) and two batches (1 kg each) of frozen cooked clam meat; (ii) Ruditapes decussatus, a grooved carpet shell, which was bought raw (shell on) at a local market (2 kg); and (iii) Cerastoderma edule, a common edible cockle also bought raw (shell on) at a local market (2 kg). Specimens of (ii) and (iii) were cooked live in a pan without added water and the meat remo- ved. All samples were stored deep-frozen at – 20°C until analysis, which was performed on the adductor muscle, mantle and foot of both raw and cooked animals.
Meretrix lyrata and Ruditapes decussatus belong to two different genus of the same family, Veneridae family, whereas Cerastoderma edule belongs to another family, Cardiidae family. Both families, Veneridae and Cardiidae, are included in the same order, Veneroida, of Lamellibranch class.
2.2 Analytical methods
Three standard operating procedures defined previously in EC projects were used. Isoelectric focusing (IEF), the usual method for identifying raw fish meat, was applied to water-soluble fish protein extract (REHBEINet al., 1995), whereas sodium dodecyl sulphate-polyacrylamide gel electrophoresis (SDS-PAGE) and urea-IEF have been used for the authentication of processed fish (PIÑEROet al., 1999; ETIENNEet al., 1999).
The same basic approach consisting of a flat-bed electrophoresis, Multiphor II Electrophoresis System (Pharmacia Biotech) run at least at 2000 V, a thermostatic circulator, a homogeniser (Ultraturrax), a centrifuge (20000 g), a spectrophotome- ter with quartz cuvettes for measurements at 280 nm, a rocking platform, a gel air-dryer and an image-analysis system, was used for all analyses.
The protein determination of the extracts was made using the same tech- nique for the three electrophoretic methods, the OD280-procedure. The principle of this determination is based on the assumption that if a solution gives an absorbance at 280nm of 1, this means that the protein concentration is 1 mg/ml. Fish muscle extracts, bovine serum albumin (BSA) standard solution (10 mg/mL in 0.2% (w/v) SDS) and reagent control without protein (extraction solution) were diluted 20 fold with 0.2% (w/v) SDS. The absorbance was mea- sured at 280 nm in a spectrophotometer using quartz cuvettes and the protein content of the extracts (mg protein/ml) was calculated using the equation:
Psample= [Asample– Ablank]* 20, (20 is the dilution factor). As a control, the absor- bance was read also for BSA standard solution and the 0.2% SDS solution, and the difference (ABSA– ASDS) should be close to 0.33. This protein determination requires that the extracts were not frozen before analysis.
Since in previous experiences and publications no pattern differences were observed among different batches of the same species, if the same cooking times were applied (WIGGIN and KRYNOWEK, 1983) and even with different
cooking times (PIÑERO et al., 1999; ETIENNE et al., 1999; REHBEIN et al. 1999), two protein extracts of each raw batch and each cooked batch were performed per kind of analysis.
2.2.1 Isoelectric focusing analysis of water-soluble proteins Protein extract
Extraction of water-soluble proteins: 500 mg of meat, from 5 to 10 indivi- duals, were homogenised in 1.0 mL of pre-cooled distilled water using the Ultraturrax homogeniser in an ice bath. The mixture was clarified by centrifuga- tion (20 min, 20000 g, 4°C), and the supernatant was collected and stored at + 4°C until analysis (24 h maximum).
Preparation of samples for native IEF: Protein concentrations of the sample extracts were adjusted to 8 mg/mL with distilled water.
Preparation of the pI calibration kit: The content of one vial of broad pI kit, pH 3.5/9.3 (Pharmacia Biotech, 17-0471-01) was solubilised in 100 µl of distilled water.
Electrophoresis conditions
Instrument and gel preparation: The gel support plate was cooled to 15°C.
IEF electrode strips (Pharmacia Biotech, 18-1004-40) were soaked in the follo- wing solutions: anode 1 M H3PO4, cathode 1 M NaOH. The gel [Ampholine PAG plates 3.5 / 9.5 (Pharmacia Biotech, 80-1124-80)] was placed on the cooling plate, and the soaked electrode strips were applied to the gel.
Sample application and running conditions: 10 µL of sample or of pI mar- ker were applied 10 mm from the cathode using sample application pieces (Pharmacia Biotech, 80-1129-46). Running conditions were 1500 V, 50 mA, 30 W and 1.5 h.
Fixation and Coomassie staining
The gel was placed in fixing solution [11.6% (w/v) TCA, 3.4% (w/v) sulphosa- licylic acid] for 30 to 60 min maximum, washed in de-staining solution for 5 min [ethanol/acetic acid/water (50/16/134) (v/v/v)], stained for 10 min in staining solution [0.1% (w/v) Coomassie Brillant blue R-250 (Sigma, B-0149) dissolved in de-staining solution] pre-heated to 60°C. After a further de-staining step, the gel was soaked in the preserving solution [1% (w/v) glycerol in de-staining solution]
for 1 h, covered with a cellophane preserving sheet and dried in a gel air-drying system. The gels were then scanned.
2.2.2 Urea-IEF (CleanGel) analysis Protein extract
Protein extraction: 500 mg of flesh, from 5 to 10 individuals, were homoge- nised with 4 ml of extraction solution [8 M urea, 0.1 M 1,4-dithiothreitol (DDT), 20 mM sodium phosphate, pH 6.5] using an Ultraturrax homogeniser. The mix- ture was kept at room temperature for at least 30 min, and un-dissolved mate- rial was then removed by centrifugation (20°C, 15 min, 20000 g).
Preparation of samples for urea-IEF: The protein concentrations of the sample extracts were adjusted to 8 mg/mL with the extraction solution.
Preparation of the pI calibration kit: The pI markers used were parvalbumin dry matters (PADM) prepared according to a procedure described by REHBEIN (2000). The PADM mixture gave 7 bands in urea-IEF.
Electrophoresis conditions
Re-hydration of CleanGel IEF: CleanGel (Pharmacia Biotech, 18-1035-32) was re-hydrated in 21 mL of 8 M urea, 0.5% (w/v) Servalyte 2-4, 2% (w/v) Ser- valyte 4-6 and 0.5% (w/v) Servalyte 4-9T.
Instrument and gel preparation: The gel support plate was cooled to 15°C.
Electrode wicks (Boehringer Ingelheim Bioproducts, 42942) were then cut to a suitable length for the gel and soaked with an appropriate volume of anode fluid 3 (0.025 M aspartic acid, 0.025 M glutamic acid, 10 mM CaCl2) or cathode fluid 10 (2 M ethylenediamine, 0.025 M arginine, 0.025 M lysine). The gel was placed on the cooling plate, and the soaked electrode wicks were applied.
Sample application and running conditions: The IEF settings were: pre- focusing (500 V, 8 mA, 8 W, 30 min), sample entrance (500 V, 8 mA, 8 W, 20 min), and focusing (2000 V, 14 mA, 14 W, 5000 Vh). After pre-focusing, the applicator strip [7 × 1 mm, silicon rubber (Boehringer Ingelheim Bioproducts, 42989)] was positioned 2 cm in front of the cathodic wicks; 7.5 µl of each sample extract were placed into the slots of the strip, and 10 µl of pI marker solution were applied under the same conditions.
Fixation and Coomassie staining
At the end of the run, at 5000 Vh, the proteins were fixed and stained with Coomassie SERVA Violet 17 dye (Boehringer Ingelheim Bioproducts, 35072).
The gel was placed in fixing solution [20% (w/v) TCA] for 30 min, washed in de- staining solution [methanol/acetic acid/water (25/10/65) (v/v/v)] for 30 min, and stained with [0.1% (w/v) SERVA Violet 17 dissolved in de-staining solution] for 10 min. After a de-staining step, the gel was soaked in the preserving solution [1% (w/v) glycerol (87%)] for 10 min, covered with a cellophane preserving sheet and dried in a gel air-drying system. The gels were then scanned.
2.2.3 SDS-PAGE analysis Protein extract
Protein extraction: 300 mg of flesh, from 3 to 7 individuals, were homogeni- sed (Ultraturrax) in 4 mL of extraction solution [2% (w/v) SDS, 0.1 M DTT, 60 mM Tris-HCl, pH 7.5]. Samples were then boiled in a water bath (100°C) for 2 min, homogenised for 30 s while hot, and centrifuged (20000 g at 20°C for 15 min).
Preparation of samples for SDS-PAGE: The protein concentration of the samples was adjusted to 0.3 mg/mL with Laemmli buffer [4.8% (w/v) SDS, 1 mM EDTA, 0.1 M DTT, 20% (v/v) glycerol, 125 mM Tris-HCl, pH 6.8, 0.05%
(w/v) bromophenol blue].
Preparation of the molecular weight calibration kit: The molecular weight marker [Novex Mark 12 (LC 5677)], was diluted 7 times with Laemmli buffer. The staining indicator [1.6 mg bovine plasma albumin (Sigma, A 7517) and 1.6 mg egg white lysozyme (Sigma, L 4631) in 1 mL of extraction solution] was diluted 784 times with Laemmli buffer in two steps (2 times 1/28).
Electrophoresis conditions
Instrument and gel preparation: The gel support plate was cooled to 15°C and the gel [ExcelGel SDS Homogeneous, 15% (Pharmacia Biotech, 80-1262- 01)] was positioned on the plate with the wells on the cathode side. The cathode and anode buffer strips (Pharmacia Biotech, 17-1342-01) were positio- ned respectively above the wells and the on the other side of the gel.
Sample application and running conditions: 10 µL of samples, together with the molecular weight marker mixture and the staining indicator, were drop- ped into the wells of the gel. Running conditions were 600 V, 30 mA and 30 W.
Once the bromophenol front began to enter the anode electrode strip, electro- phoresis was continued for another 20 min and then stopped.
Silver staining
The proteins were fixed and stained using the Silver Staining Kit, Protein (Pharmacia Biotech, Plusone 17-1150-01). The gel was soaked in preserving solution [1% (w/v) glycerol (87%)] for 20 min, covered with a cellophane preser- ving sheet and dried in a gel air-drying system. The gels were then scanned.
Figure 1
Electrophoretic profiles of different bivalve parts obtained by native IEF (on PAG Plate 3.5/9.5) and SDS-PAGE (on ExcelGel SDS Homogeneous 15%) lane 1 : adductor muscle; lane 2 : foot; lane 3 : mantle.
bivalve species : Ruditapes decussatus, grooved carpet shell.
3 - RESULTS AND DISCUSSION
3.1 Determination of the bivalve part to be used for authentication analysis
The electrophoretic patterns obtained from extracts of adductor muscle gave more characteristic bands than those from the mantle or foot (figure 1).
With native IEF or SDS-PAGE, mantle extracts produced too few bands for spe- cies identification, and with native IEF the bands for the foot were more nume- rous but not sufficiently characteristic. As mentioned by KITTSet al. (1997), the proteins present in adductor muscle give enough specific bands to be used as marker proteins to allow the identification of different shellfish species. There- fore, all subsequent analyses were performed on extracts from adductor muscle.
Figure 2
Native IEF profiles of bivalves obtained on PAG Plate 3.5/9.5 from adductor muscle of raw (lanes 1 to 4) and cooked samples (lanes 5 to 8).
Lanes Species
1, 5 grooved carpet shell pattern (Ruditapes decussatus) 2, 8 common edible cockle (Cerastoderma edule) 3, 4 and 6,7 Vietnamese “baby clams” (Meretrix lyrata) batches 1 and 2
Mk isoelectric point protein markers (pI) Species specific proteins are highlighted between brackets.
3.2 Comparison of the electrophoretic patterns of raw and processed bivalve meat (adductor muscle)
Some bands present in the IEF profiles of raw bivalve meat disappeared during processing (figure 2), whereas SDS-PAGE and urea-IEF profiles remai- ned unchanged (figure 3 and 4). These results were in accordance with ETIENNE et al. (1999) and REHBEINet al. (1999).
3.3 Analytical techniques for species identification of bivalve meat
3.3.1 IEF of water-soluble protein extract
For raw samples, native IEF on PAG plates 3.5/9.5 allowed to distinguish the shellfish species tested (figure 2), the discriminating zone was situated between pH 5.5 and pH 6.5. The pI of the main characteristic bands were, 6.11 – 5.59 –
Figure 3
PAGE-SDS profiles of bivalves obtained on ExcelGel SDS Homogeneous 15% from adductor muscle of raw (lanes 6 to 9) and cooked samples (lanes 1 to 5)
Lanes Species
2, 3 and 6,7 Vietnamese “baby clams” (Meretrix lyrata) batches 1 and 2 1, 5, 9 grooved carpet shell (Ruditapes decussatus)
4, 8 common edible cockle (Cerastoderma edule) Mk molecular weight protein markers (MW) Species specific proteins are highlighted between brackets.
5.41 for grooved carpet shell, 6.26 – 6.14 – 6.06 – 5.22 – 4.31 for common edible cockle and 6.10 – 5.88 – 5.79 – 5.60 – 5.25 for “baby clam”. Although numerous bands disappeared after cooking, the patterns still showed enough differences for discrimination, the “baby clam” pattern (Meretrix lyrata) was dif- ferent from that of the grooved carpet shell (Ruditapes decussatus) and the common edible cockle (Cerastoderma edule).
3.3.2 SDS-PAGE of SDS extracts
The SDS-PAGE profiles obtained from SDS extracts of adductor muscle were characteristic of each studied species (figure 3). The protein fraction under 50 kDa afforded discriminatory bands that permitted to distinguish between these species. The profiles from the raw and cooked batches of “baby clams”
were identical. Thus no intra-species difference and no differences due to the processing were observed, indicating that the profiles of the raw samples could be used as a reference.
Figure 4
Urea-IEF profiles of bivalves obtained on 8 M urea CleanGel from adductor muscle of raw (lanes 3, 4, 6, 8) and cooked samples (lanes 1, 2, 5, 7)
Lanes Species
1, 2 and 3, 4 Vietnamese “baby clams” (Meretrix lyrata) batches 1 and 2 5, 6 grooved carpet shell pattern (Ruditapes decussatus) 7, 8 common edible cockle (Cerastoderma edule)
Mk isoelectric point protein markers (pI) Species specific proteins are highlighted between brackets.
3.3.3 Urea CleanGel IEF of 8 M urea extracts
The urea IEF profiles obtained from urea extracts of adductor muscle were characteristic of each studied species (figure 4). The protein fraction showing an apparent pI under 6 afforded discriminatory bands that permitted to distinguish these species. The urea IEF pattern profiles obtained from raw and cooked samples were practically identical (figure 4), indicating that the raw samples could be used as a reference.
4 - CONCLUSION
Although all three methods tested could be used for bivalve identification, SDS-PAGE or urea-IEF are preferable for cooked samples to avoid the pro- blems in pattern reference experienced with native IEF.
This study shows that it is possible to differentiate bivalves belonging to dif- ferent genera, after cooking by analysing extracts of adductor muscle with SDS- PAGE or urea-IEF. Raw “authentic” samples can be used to establish the reference profiles essential for electrophoretic species identification.
Received 24 November 1999, accepted 6 March 2000.
REFERENCES
BOSSIER P., 1999. Authentication of Sea- food Products by DNA Patterns. J. Food Sci., 64, 189-193.
CODURI R.J., 1972. Vertical plate gel electro- phoresis for the differentiation of fish and shellfish species. J. Assoc. Off. Anal. Chem., 55, 465-466.
CRAIG A., RITCHIE A., MACKIE I., 1995.
Determining the authenticity of raw reformed breaded scampi (Nephrops norvegicus) by electrophoretic techniques. Food Chem., 52, 451-454.
DURAND P., LANDREIN A., QUERO J.C., 1985. Catalogue électrophorétique des pois- sons commerciaux. Institut français de recherche pour l’exploitation de la mer, Nantes, France.
ETIENNE M., JÉRÔME M., FLEURENCE J., REHBEIN H., KÜNDIGER R., MALMHEDEN- YMAN I., FERM M., CRAIG A., MACKIE I.,
JESSEN F., SMELT A., LUTEN J., 1999. A standardized method of identification of raw and heat-processed fish by urea isoelectric- focusing : A collaborative study. Electropho- resis, 20, 1923-1933.
KITTS D.D., LEUNG PAN SHUM M., SMITH D.S., 1997. Species identification of shellfish using SDS-PAGE electrophoresis. In: SHA- HIDI F., JONES Y., KITTS D.D. (ed.), Seafood safety, processing, and biotechnology, Tech- nomic Publishing Co., Lancaster, Pennsylva- nia, 235-241.
LUNDSTROM R., 1980. Fish species identifi- cation by thin-layer polyacrylamide gel isoe- lectric focusing: Collaborative study. J.
Assoc. Off. Anal. Chem., 63, 69-73.
MACKIE I.M., 1980. A review of some recent applications of electrophoresis and iso-elec- tric focusing in the identification of species of fish in fish and fish products. In: CONNELL
J.J. (ed.), Advances in Fish Science and Technology, Fishing News Books, Farnham, UK, 444-450.
MACKIE I.M., 1997. Authenticity of fish. In:
ASHURST P.R., DENNIS M.J. (ed.), Food Authentication, 140-170, Blackie Academic and Professional, London.
PIÑEIRO C., BARROS-VELÁSQUEZ J., PÉREZ-MARTÍN R., MARTlNEZ I., JACOB- SEN T., REHBEIN H., KÜNDIGER R., MENDES R., ETIENNE M., JÉRÔME M., CRAIG A., MACKIE I., JESSEN F., 1999.
Development of a sodium dodecyl sulfate- polyacrylamide gel electrophoresis reference method for the analysis and identification of fish species in raw and heat-processed samples. A collaborative study. Electrophore- sis, 20, 1425-1432.
REHBEIN H., 1990. Electrophoretic tech- niques for species identification of fishery products. Z. Lebensm. Unters. Forsch., 191, 1-10.
REHBEIN H., 1992. Parvalbumins as marker proteins for the fish species in fishery pro- ducts. In: HUSS H.H., JAKOBSEN M., LIS- TON J. (ed.), Quality assurance in the fish industry, Elsevier Science Publ., Amsterdam, 399-405.
REHBEIN H., ETIENNE M., JÉRÔME M., HATTULA T., KNÜDSEN B., JESSEN F., LUTEN J.B., BOUQUET W., MACKIE I.M., RITCHIE A.H., MARTIN R., MENDES R., 1995. Influence of variation in methodology on the reliability of the isoelectric focusing method of fish species identification. Food Chem., 52, 193-197.
REHBEIN H., KÜNDIGER R., MALMHEDEN- YMAN I., FERM M., ETIENNE M., JÉRÔME M., CRAIG A., MACKIE I., JESSEN F., MARTlNEZ I., MENDES R., SMELT A., LUTEN J., PIÑEIRO C., PÉREZ-MARTÍN R., 1999. Species identification of cooked fish by urea isoelectric focusing and sodium dode- cylsulfate polyacrylamide gel electrophoresis:
a collaborative study. Food Chem., 67, 333- 339.
REHBEIN H., 2000. Fish muscle parvalbu- mins as marker proteins for urea isoelectric focusing. Electrophoresis (in press).
SCOBBIE A.E., MACKIE I.M., 1988. The use of sodium dodecyl sulphate polyacrylamide gel electrophoresis in fish species identifica- tion – a procedure suitable for cooked and raw fish. J. Sci. Agric., 44, 343-351.
SECCHI C., SONCINI G., BERRINI A., RUSSO V., BIONDI P.A., 1982. Reliable mol- lusc intragenera and intraspecies identifica- tion by polyacrylamide gel isoelectric focusing. Arch. Veter. Ital., 33, 99-105.
SOTELO C., PIÑEIRO C., GALLARDO J.M., PÉREZ-MARTÍN R., 1993. Fish species iden- tification in seafood products. Trends Food Sci. Technol., 4, 395-401.
TORO J.E., 1998. Molecular identification of four species of mussels from southern Chile by PCR-based nuclear markers: the potential use in studies involving planktonic surveys. J.
Shellfish Res., 17, 1203-1205.
WIGGIN K., KRYNOWEK J., 1983. Identifica- tion of frozen, cooked shellfish species by agarose isoelectric focusing. J. Assoc. Off.
Anal. Chem., 66, 118-122.
