'11... -——:—"—-:-1——« ‘I.—II__I...._........_..—-—I— *— -. , .r - . 3. ,.3 .. - #51 l. .1 13.»: 11111.1 3.219.111.1111 .1 111 1 3113;; ,. .111 ~ 1 '1 1.11 In“. 1‘ 13, 1.1311 331 1U! 11$: {1, I I ' I "é I‘f'“ "" ‘ 11-" q \1 3%. I '1 ' .-°‘, I” E1 I" '3‘: ”M?" '(IA 1I‘I \LiI . “3.10.1.1" I? 1 _vg'1: :. II ' ‘ r‘ I "I '13 ”1y’4l Jfiv ‘1 V ' 11“ ‘- R?” a ‘ 3 ‘. 1. .3. 333 31 ,.'-‘1111~" 1,1. I --“€3)1}¢-";.; - ,1 1.;%:.~,_I 331:2": 1711.; . ."fi' (1'- 11‘1I:11I3 "1“ 1"'- I , . - ‘u‘...°' . 9 “at"... A _ \1(\§? A. I 1" - “" """ . A"'1:'1"I;I?I:II : "‘ " ’ W ”“12"?" 1‘ ”AA-3 A 1a}; (:33 ‘ ’ “1.1g. 3 -::. I .I f." 3. . 5"; I r 1 11'1'.~ I ' ;.. ' H'L‘Ifiifu . . - - . . . " 1 «NH. ' II' 1). 1'1" "3’”. 111.: 1‘.I1I..;- ,I1‘11‘11111IIIEI3A 1 I 1 1. 1 '. :1 131‘1 ‘ . «"37: '3 " " A ' 1 ' I" ."1' 1 '11" " "":"'A'""'"-1;’{""|3 "‘"\ Nari: A: ._.'.'-::"K':‘::‘- 1 I1 1 . ,J IL‘1113113 . 1.111. 1 I 1.\1vI(I'115111'CILI'I‘:(\'\-ch .‘\\ .“ M 1 .2: .. " I '1‘ '11" '1‘ "'."1' I " 1 1 ‘ 1 11- 11‘»'9'-'1"1'1'-;' "‘::l' \U '1‘ VAL 7% ' ’1 ad. 11‘; In 1 "I . 4 U . 'I 11 1 1 '1 \‘W C 'h‘ I 'I'.‘1'&V'N'm'un\1"&‘-:1. I\ 311 ". "1"'1'I"' 3 1 1 ~ ._1y1j.313\I.13N:1 {It WI: “gt-112%“ 3?} '5er .‘ 1 11- t" n |.1, 1 .- 'HII" '1 IV‘I “I I - 1- .' in I 3 ‘ 1 I‘ ':'II1'1"1:"' III}. MI :11 GIN “35",; II1,III\1I:$II;: I931 . . \f' I. V \ ~‘ {'1 - 1. 5%‘ “a” T131 5va 1 0).. "1 1' "13;."11': ‘0 1‘-1..'.“_I;j'1 1" ’ ‘VIQII I.» 1111'? ‘._--r1-_,._—..v—’ :31"— — —4 \ 1‘ 1 I“ 1" 1 1 "'1‘I’:.'1‘)':'N"I'I'P\’1'fié't'vfl 1'3 .: 3 ‘1 '. .' 1'1I\'.I.I1 'I: l \‘11‘ '1I 1 13' 1 'T 2‘1T" 1‘) {$3.}. lI 33:33 '53; '1': 51:1 '. 3 ' .'.'- 1i Irv?” 33:31.3 ' II '1" III '1 11 '11.“ J. ‘11 .1 Y1" 1II1I I1IIItIII33 I“ 111'. .1 1.31 If. .I 11:". I. 11)“? és-Zi “:th ~ 1 ‘l ‘ :\ "".'1'1' 1 ‘1 ‘ ---‘ if . h." 11 .‘ I .I11II3 32 I‘ ‘ «1:1 1 J, I u 3 q ,1 1 W1 1..1~.. 1‘1... >' I, ' 1. --0 .1...‘ .3 “Kamxmaz‘fldjgfif‘ur 31‘}. Y:,.. '1‘" -~ "‘" "NR” “=1 ' -___ Duh.- I' "I‘F‘."\}%I 3 ' 1“. n I ~ 2'..'v" ,I 1'. i... (.1. ‘1’" ’ 1' ' 1 1.: 1":\-'1f: - 1 r 1 '1 ' 1 1 1. 1 3 ’13- 1 n 1'. . 1 . 1.11 _'..1 1 1 1.11.1.1. ~ 1 111.111 .111. 121:.» {13-1 ”311-33,... .3 _I '1 1 'II 11" ' 1 1 I 1' ‘ 1 1 ‘ II 1‘ 1 . . '3' 1 'u & 1 '0" .I L ' 1"1 I '5' .13 (“‘1 ~ ' ' 1 1 . 1 ‘ .1 I'c . I ._ '. 1 I I 1' I, IIC1:I:'\" . 1 11 . . ' . :1 - ' . - '11. f 11 1 1.11.1111”. " I 1 1 1' 1 1 ' ' .5. .3 . 3. ‘V'. .‘ 1 . 1 II 1 I '1' ' - 1. I. ' 353:- .1? . «'1‘ - 33_.. . . 3 1 1 - 1 9 1‘ ' ‘ 1 ’ " ' 1 "I:"I"'\:;1h 1' II.1:"'.' '. 1 1 ' I. _ " ' " ‘1 ‘11. '1...“.' 1 ' 1 ':1'11'h‘.|:"..")::;“ I. h 'l' '-- '1'- ' I'III'III'I."'.'A.'1' 1 b '33.“. . . '3k 3 1 ' 1: 1' 1 '11: - {"C'vu-IJ.‘ I1"1' III-T‘- '1' ‘3I(1JI1L'IY3' 3:???“ .1 'l "1" ' "I" 1 l l' 1 ‘I'I'I‘I. A ' 3"""1'v'1"'I'1I".1"' I" :.I . I I . I I‘ I. . . 1 I , I "111 ‘ . 1 .1”! I 1‘ "\ J‘“""““:’:’:\;"’ 1‘.- 3. 1T '1'" " " ' l I ,' 1‘ :1 ‘1' ' ."', ‘ '1 , 11 ' II: I. . 1'-' 1'- V 1’ I11!- "1 .I . 1 ,- . .. 1 I 3‘1"; I 1I-I‘ '. ' I "ml , II' 'I 1 III: 1‘ I ' " ' ' ' ' "J I :'I ' I 1"' ' “'1' ' I "'| 1 1.11 I}. ‘ 1 1 - 1 I II I 1 a. '1 V 1 ' mil: I :1 . 1 '. 1 1 ‘ 1"41‘11'.” ‘ 'I 1 ' ' . I r ' 1' <1IVII11quI)‘:I1 t I‘ 1 1 1 1 .1' ”r, 1 3. I 1 5 1 1 I I3 1 'I 3 1.1 1 ' 1 13“ 1,131. 1 1 ’1 ' -.‘- 13W. . ' 1: 1 t " ' IJ . . v. ' \ ‘ “ I \ II I. I 1 1 . , 11 II 1 ‘ .1 H' ' t 1. ':':; £1 .‘ 1; .. 1 1 - 11: ,I "~11“ I.3.1I 131. I 4"I1'1'III11I: . ,1‘I' (2.1.)5'1'1:I|5Il "“V)3H:., 4"!‘1'1'Qsa ATM-ha ”Amh‘hAl'fi'un .1)” 11 ”131.111.": 1‘ Lin}! ' M'L‘CL Lid‘ 'l} 30'111'” . lllllllllllllllllll lllllllllltl [— 3 1293 10706 8458 LIB RA R Y M m State Umversity This is to certify that the thesis entitled High Pressure Liquid Chromatographic Analysis of Methyl Ketones in Quick Ripened Blue Cheese presented by S‘hahram Dokhani has been accepted towards fulfillment of the requirements for Ph.D. degree in Food Science & Human Nutrition @043»); Major professor Date April 17, 1978 0-7 539 HIGH PRESSURE LIQUID CHROMATOGRAPHIC ANALYSIS OF METHYL KETONES IN QUICK RIPENED BLUE CHEESE BY Shahram Dokhani A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science and Human Nutrition 1978 €3/fgpéu5§> ABSTRACT HIGH PRESSURE LIQUID CHROMATOGRAPHIC ANALYSIS OF METHYL KETONES IN QUICK RIPENED BLUE CHEESE BY Shahram Dokhani The components of a high pressure liquid chromato- graphic (HPLC) system were assembled to separate and quan- titate C3 to C13 methyl ketones from Quick-ripened (QR) blue mold cheese and other cheese. The QR cheese were processed and cured during 7 to 15 days at 52° or 62°F and 95% RH. At the end of the ripening period portions of the cheese were packed and stored at 400, 480, or 620E for an addi- tional period of 4 to 15 days. The pH and moisture content of the samples were determined throughout the ripening and storage period. Standard methyl ketones with 99+% purity were deri- vatized with 2,4-dinitrophenylhydrazine dissolved in aceto- nitrile and separated on a u Bondapak C18 column using ace- tonitrilezwater (75:25 v/v) as the mobile phase. The sepa- rated components were detected by a U.V. monitor at 254 nm in the HPLC unit and were compared with their corresponding derivatives which had been isolated from Blue cheese. The Shahram Dokhani peak heights were used in quantitation. Percent recovery of individual methyl ketones and the slopes of calibration graphs were also used in quantitation. Hexane extracts of the cheese samples were passed through a DNPH reaction column and after fat removal following several clean-up steps, the monocarbonyl deriva- tives were injected into the HPLC unit and analyzed for methyl ketones. HPLC and mass spectral analysis of 2,4- DNPH derivative of 2-undecanone obtained from the reaction column showed the possibility of geometric isomerization of methyl ketones. Methyl ketone development in QR cheese followed a pattern similar to that observed in commercial Blue cheese ripened several months. The predominant methyl ketones were C and C . White mold QR cheese contained the C ketone 7 9 9 predominantly whereas the C ketone was most abundant in 7 blue mold QR cheese. A large variation in the concentration of methyl ketones was observed in the samples. White mold QR cheese contained a lower concentration of methyl ketones compared with conventional blue mold cheese. Brief storage at 480F after curing resulted in a greater formation of methyl ketones in the cheese. Cheese ripened at lower temperatures (52°F) and stored two weeks at 40°F was of superior flavor although total ketone content of such cheese was lower than in cheese ripened and stored at higher temperatures. Salting methods affected color and Shahram Dokhani methyl ketone formation in the samples. A darker color developed in cheese salted on days 1, 9, and 10 rather than 7, 8, and 9. Blue cheese salted on days 7, 8, 9 demon- strated a greater ketone concentration than cheese made by salting on days 1, 9, 10 and stored at 48oF. The cheese lost 9 - 13% moisture in the curing room, but negligible losses occurred in storage. All cheese showed the same changes in pH during ripening. Changes in pH were faster at higher temperature of curing. Freeze drying of Blue cheese samples resulted in a significant loss of methyl ketones. The HPLC method was advantageous in this study for rapid analysis and minimal sample clean up. Dedicated To My Father Dr. Morteza Dokhani ii ACKNOWLEDGMENTS This author extends sincere appreciation to his major professor, Dr. Charles M. Stine, for his helpful suggestions and counsel throughout the course of this study and during the preparation of this manuscript. Appreciation and thanks are extended to members of the guidance committee, Drs. P. Markakis, Department of Food Science and Human Nutrition, H. A. Lillevik, Department of Biochemistry, and D. R. Heldman, Department of Agricultural Engineering for their advice and effort in reading this manuscript. The author also wishes to acknowledge Mr. Bruce R. Harte and Mr. M. L. Richmond for their time in assisting in the manufacture of this cheese. In addition, the technical assistance of Dr. C. Sweeley for mass spectrometry analyses is gratefully appreciated. The author desires to acknowledge his parents for their constant support and stimulation and encouragement to pursue a higher degree of education. iii TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . viii LIST OF FIGURES . . . . . . . . . . . . . . . . . . xi INTRODUCTION . . . . . . . . . . . . . . . . . . . . 1 REVIEW OF LITERATURE . . . . . . . . . . . . . . . . 2 The History, Chemistry, and Classes of Flavor Compounds in Blue Cheese . . . . . . . . . 2 Free Fatty Acids . . . . . . . . . . . 2 Lipid Metabolism (lipolysis) . . . . . . 5 The Lipase Native to Milk . . 5 Lipases of P. rogueforti . . . . . 7 Lipases of Other Microorganisms Present During Ripening of Blue Cheese . . . . . . . . . . . . . 10 Methyl Ketones (2-Alkanones) . . . . . . 12 Fatty Acid Metabolism and Methyl Ketone. Formation . . . . . . . . . . . . . . . 15 Mechanism of Methyl Ketone Formation by Fungi . . . . . . . . . . . . 15 Secondary Alcohols (2- Alkanols) . . 24 Alcohols Other than 2-Alkanols . . . . 25 Aldehydes . . . . . . . . . . . . . . 26 Esters . . . . . . . . . . . . . . . . 27 Other Classes (Miscellaneous) Flavor Components . . . . . . . . . . . . . . . 28 Protein Metabolism (Proteolysis) . . . . . . 29 Blue Cheese Flavor Concentrates . . . . . . . 34 Accelerated (Quick) Ripening of Blue Cheese . . 36 Cheese Dehydration and Flavor Retention . . . . 41 EXPERIMENTAL PROCEDURES . . . . . . . . . . . . . . 47 Production of Quick Ripened (QR) Cheese . . . . 47 Blue Mold Powder . . . . . . . . . . . . . 47 Preparation of White Mold Powder . . . . . 47 Manufacturing Procedures Used for Quick Ripening (QR) Cheese . . . . . . . . . . 48 iv Commercial Blue Cheese . . . . . Freeze Drying of Commercial and Quick Ripened Blue Cheese . . . . . . . . . . . . . . ANALYTICAL PROCEDURES . . . . . . . . . . pH . . . . . . . . . . . . . . . . . . . . . Moisture . . . . . Liquid Chromatographic Analysis of Methyl Ketones . . . . . . . . . . Preparation and Purification of Solvents . . . . . . . . . . . . . . . Hexane . . . . . . . . . . . . . . . Benzene . . . . . . . . . . . . . . Chloroform . . . . . . . . . . . . . . Nitromethane . . . . . . . . . . . . . Acetonitrile . . . . . . . . . . . . Ethanol . . . . . . . . . . Ethylene Chloride . . . . . Butanol- -1 . . . . . . . . Purification of 2, 4- -dinitropheny1- hydrazine . . . . . . . . . . Preparation of 2, 4- DNPH Derivatives of Standard Methyl Ketones . . . Isolation of Monocarbonyls from Cheese Samples . . . . . . . . . . . . . . . . . Extraction of Fat . . . . . . . . . . . Formation of 2, 4-DNP-hydrazone Derivatives . . . . . . Removal of Fat from 2, 4- DNP- -hydra- zones . . . . . . . . . . . . . . Fractionation of 2, 4-DNP-hydrazones on Weak Alumina . . . . . . . . Determination of Methyl Ketones by High Pressure Liquid Chromatography (HPLC) HPLC Apparatus . . . . . . . . . . HPLC Specification . . . . . . Preparation of Standard Curves . . . Recovery of Methyl Ketones . . . . . Quantitation of Individual Ketones in Cheese Samples . . . . . . . . . Mass Spectral Analysis of 2-Undecanone- DNPH-Derivative . . . . . . . . . . . . . Preparation of the Sample . . . . . . . Mass Spectrometer Analysis . . . . . . Separation of Methyl Ketones from Other Classes of Monocarbonyls in Blue Cheese . . . . . . . . . . . . . . . . . Page 49 49 51 51 51 51 51 51 53 53 53 53 54 54 54 54 55 56 56 56 57 58 58 58 58 61 62 62 63 63 64 65 RESULTS AND DISCUSSIONS . . . . . . . . . . . . . Liquid Chromatographic Determination of Methyl Ketones as their 2,4-Dinitrophenylhydrazone (2,4-DNPH) Derivatives . . . . . . . . . . . . Extraction and Fractionation of Carbonyl Compounds in Cheese Fat . . . . . . . . High Pressure Liquid Chromatographic (HPLC) Separation of Methyl Ketones . . . . . . The Reliability of HPLC System . . Detection of "Pre-peaks" related to individual methyl ketones . . . . . . HPLC and Mass Spectrometric Analysis of 2, 4- DNPH Derivatives of C11 methyl ketone . . . . HPLC Separation of Methyl Ketones from Other Classes of Monocarbonyls . . Quantitation of Individual Methyl Ketones from C to C13 . . . . . . . . . Preparation of Standard Curves . . . . Recovery of Methyl Ketones . . . . . . Methyl Ketones in Commercial Imported or Domestic Blue Cheese . . . . . . . . . Sensory Evaluation of Commercial Blue Cheese . . . . . . . . . . . . . Methyl Ketones. in Quick Ripened (QR) Cheese Incorporated with White Mutant of P. Roqueforti . . . . . . . . . . . . . . . Sensory Evaluation of Quick Ripened White Cheese . . . . . . . . . b . Methyl Ketones in Blue Cheese QR at 52 F (11°C) . . . . . . . . . . . . . . Sensgry Evaluation of Blue Cheese QR at 52 F . . . . . . . . . . . Methyl Ketones in Blue Cheese QR at 626F (16. 70C) . . . . . . . . . . . . . . . . Sensory Evaluation of Blue Cheese QR at 62°F . . . . . . . . . . . . . . . . . . Changes in pH During Quick Ripening of Cheese . . . . . . . . . . . . . . . . . . . . Changes in Moisture During Ripening of QR Blue Cheese . . . . . . . . . . . . . . . . . . . . Retention of Methyl Ketones During Freeze Drying of Blue Cheese . . . . . . . . . . . . . . . . SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . APPENDICES Appendix A . . . . . . . . . . . . . . . . . . . Appendix B . . . . . . . . . . . . . . . . . . . Page 66 66 66 68 68 77 81 86 89 91 94 99 103 108 111 117 120 124 126 131 134 138 142 143 Page Appendix C . . . . . . . . . . . . . . . . . 144 Appendix D . . . . . . . . . . . . . . . . 145 BIBLIOGRAPHY . . . . . . . . . . 147 vii Table 10. LIST OF TABLES Page Free Fatty Acid Content of Quick Ripened and Commercial Blue Cheese . . . . . . . . . . . 4 Process Variables Used in Preparation of Quick Ripened Blue Cheese . . . . . . . . . . . . 49 The Retention Time Values for 2,4-DNP-hydra- zones of Standard Methyl Ketones and Ketones Present in Monocarbonyl Fraction of Blue Cheese Samples . . . . . . . . . . . . . . . 76 Intensity Comparison of C11 and C11* for Peaks Higher than 20% of Basepeak . . . . . 85 The Values for SIOpes, Correlation Coeffi— cient, and Intercepts of Standard Curves of 2,4-DNP-hydrazone Derivatives of C3 - Cl3 Methyl Ketones . . . . . . . . . . . . . . . 91 Percent Recoveries of Standard, C3 - C13, Methyl Ketones . . . . . . . . . . . . . . . 93 Methyl Ketone Concentration of Domestic and Imported Commercial Blue Cheese . . . . . . 95 Sensory Evaluation of Commercial Imported and Domestic Blue Cheese . . . . . . . . . . . . 102 Methyl Ketone Concentration of Quick Ripened Cheese, Using White Mutant of P. rogue- forti, at 52 i 1 F (11 C) and over 95% RH, Salted on Days 1, 6, and 9 During 15 Days in the Curing Room . . . . . . . . . . . . . 104 Methyl Ketone Concentration of Quick Ripened Cheese, Uging White Mutant of P. roqueforti, at 52 i 1 F (11 C) and over 95% RH, Salted on Days 6, 7, and 8 During 15 Days in the Curing Room . . . . . . . . . . . . . . . . 105 viii Table Page 11. Methyl Ketone Concentration of the 11th and 15th Days Quick Ripened Cheese, Using Whige Mutant of P. roqueforti, at 52 i 10F (11 C) and Over 95% RH, Salted on Days 1, 6, 9 05 6, 7,08 Plus Two Weeks Stored at 40 i 1 F (4.4 C) . . . . . . . . . . . . . . 107 12. Sensory Evaluation of QR Cheese, Using White Mutant of P. roqueforti at 52 i 1 F (11 C) and Over 95% RH, Salted on Days 1, 6, 9 or 6, 7, 8 During 15 Days Ripening Plus 2 Weeks Storage at 40 i 1 F (4.4 C) . . . . . 110 13. Methyl Ketone Concentratign of Blue Cheese Quick Ripened at 52 i 1 F and Over 95% RH Salted on Days 1, 9, 10 During 15 Days in the Curing Room . . . . . . . . . . . . . . 112 14. Methyl Ketone Concentratign of Blue Cheese Quick Ripened at 52 i 1 F and Over 95% RH, Salted on Days 7, 8, 9 During 15 Days in the Curing Room . . . . . . . . . . . . . . 113 15. Methyl Ketone Concentration of the 11th and 15th Dgys Blue Cheese Quick Ripened at 52 i 1 F and Over 95% RH, Salted on Days 1, 9, 10 or 7, 86 and 9 Plus Two Weeks Stored at 40 i 1 F . . . . . . . . . . . . . 115 16. Methyl Ketone Concentration of the llgh Day Blue Cheese Quick Ripened at 52 i 1 F and Over 95% RH, Salted on Days 1, 9, 10 or 7, 8, 8nd 9 Plus Two Weeks Stored at 48 i 1 F . . . . . . . . . . . . . . . . . . 116 17. Sensory Evaluation 8f Blue Cheese Quick Ripened at 52 i 1 F and Over 95% RH, Salted on Days 1, 9, 10 or 7, 8, and 9 During 150 Days Ripening Plus Two Weeks Stored at 40 or 48 F . . . . . . . . . . . . . . . . . . 119 18. Methyl Ketone Concentratign of Blue Cheese Quick Ripened at 62 i 1 F and Over 95% RH, Salted on Days 4, 5, and 6 During 7 Days in the Curing Room . . . . . . . . . . . . . . 121 19. Methyl Ketone Concentration of the 7th Day Blue Cheese Quick Ripened at 62 i 1 F and Oger 95% RH Plus 4 Days Storage at 62 i 1 F . . . . . . . . . . . . . . . . . . . . 123 ix Table Page 20. Methyl Ketone Concentration of the 7th Day Blue Cheese Quick Ripened at 62 i 1 F and Over 93% RH Plus Two Weeks Stored at 40 1 1 F or 48 1 1 F . . . . . . . . . . . 124 21. Sensory Evaluation 8f Blue Cheese Quick Ripened at 62 1 1 F and Over 95% RH During 70Days Plug 4 - 15 Days Storage at 40 1 1 , 48 1 1 , or 62 1 1 F . . . . . . . . . 126 22. Moisture Content of Blge Cheese Quick Ripened at 52 or 62 F and Over 95% RH, Salted on Days 1, 9, 10 and 7, 8, 9 or 4, 5, 6 During 7 or 15 Days Ripening . . . . 132 23. Moisture Content of Quick Ripened Cheese, During Storage Period and Commercial Blue Cheeses O O O O O O I O O O O O O O O O O 133 24. Percent Retention of Methyl Ketones in Freeze Dried Commercial or Quick Ripened Blue Cheese . . . . . . . . . . . . . . . 135 A. The Molecular Weight of Methyl Ketones or Their 2,4-Dinitr0pheny1hydrazone Derivatives . . . . . . . . . . . . . . . 142 C. Moisture Content of Fresh and Freeze Dried Blue Cheese I O O O O O O O O O I O O O O 1.44 pH Values of QR Cheese, Using Blue or White Mold Powdeg or P. roqueforti Cured at 52 F or 62 F and Over 95% RH . . . . . . . 145 pH Values of Commercial Blue Cheese or QRO Cheeses, Dgring Two Weeks Storage at 40 , 48 , or 62 F . . . . . . . . . . . . . . . 146 Figure 1. 2. 3. 4. 9. LIST OF FIGURES Schematic of High Pressure Liquid Chroma- tographic (HPLC) System . . . . . . . . . Typical HPLC Separation of 2,4-dinitro- phenylhydrazine Derivatives of Standard Methyl Ketones, 2-Pr0panone (C ), 2-Penta- none (C ), 2-Heptanone (C7), 2-Nonanone (C9), 2-Undecanone (C11), and 2-Tridecanone (C13) . . . . . . . HPLC Resolutions of 2,4—DNPH Derivatives of Standard Methyl Ketones, C3 - C13, at Dif- ferent Solvent Ratio of Acetonitrile: H o O O O O O I O O I O O I O O O O O O O O 2 Typical HPLC Separation of 2,4-DNPH Deriva- tives of Methyl Ketones, C3 - C13, Present in Monocarbonyl Fraction of Commercial Blue Cheese Crumbles . . . . . . . . . . . HPLC Separation of 2,4-DNPH Derivatives of Standard Methyl Ketones, C3 - C13, Prepared in Solution and "Aged" by Passing Through a Celite—magnesia and Alumina Column . . . HPLC Separation of 2,4-DNPH Derivatives of Standard Methyl Ketones, C3 - C13, Prepared Through a DNPH-Reaction Column . . . . . . HPLC Separation of 2,4-DNPH Derivatives of Standard 2-Undecanone . . . . . . . . . . . Mass Spectrometry Patterns of 2,4-DNPH Derivatives of Standard 2-Undecanone . . . Typical HPLC Separation of Methyl Ketone- DNPH-Derivatives Fractionated from Other Classes of Monocarbonyls, Present in Quick Ripened Cheese . . . . . . . . . . . . . . xi Page 59 70 72 73 78 79 82 84 88 Figure 10. ll. 12. l3. 14. 15. 16. 17. 18. Page Calibration Graphs, Concentration vs Peak Height Units, for 2,4-DNPH Derivatives of Methyl Ketones, C3 - C13, Based on a Detector Attentuation of 0.08 AUFS . . . . 90 Typical HPLC Separation of 2,4-DNPH Deri- vatives of Methyl Ketones, C3 - C13, in Roquefort and Wisconsin Blue Cheeses . . . 98 Typical HPLC Separation of 2,4-DNPH Deriva- tives of Methyl Ketones, C3 - C13, in Danish and Norwegian Blue Cheeses . . . . 101 Typical HPLC Separation of 2,4-DNPH Deriva- tives of Methyl Ketones, C - C13, in QR Cheese Incorporated with ngte Mutant of P. roqueforti . . . . . . . . . . . . . . 109 Typical HPLC Separation of 2,4-DNPH Deriva- tives of Methyl Ketones, C3 - C13, in QR Blue Cheese . . . . . . . . . . . . . . . 118 Typical Concentrations of 2-Heptanone and Z—gonanone in Blue Cheese Quick Ripgned at 62 F and stored at 40 , 48 , and 62 F . . 125 Commercial Blue Cheese Crumbles and Quick Ripened Cheeses Incorporated with White or Blue Mold Powder of P. roqueforti . . . . 127 Changes in pH During Quick Ripening of Cheese Incorporated with Blue or White Mutant of P. Roqueforti . . . . . . . . . . . . . . 129 Typical HPLC Separation of 2,4-DNPH Deriva- tives of Methyl Ketones, C3 - C13, in Commercial Blue Cheese and Freeze-dried Commercial Blue Cheese . . . . . . . . . . 137 xii INTRODUCTION The consumption of blue mold cheeses has increased significantly in popularity since 1960. Much of this cheese is consumed directly, but increasing quantities and greater demands are for the food products which are flavored with blue mold cheeses such as French Roquefort, Italian Gorgon- zola, English Stilton, Danish blue cheese, American blue cheese, etc., to provide salad dressings, mayonnaise, snack foods, bakery items and chip dips. Because of the growing market for blue mold cheeses and their related food products, attempts have been directed toward more economical and efficient manufacturing and "quick ripening" of such cheeses. Considerable progress has been achieved in the elucidation of the flavor chemistry of Blue cheese. Other related research has been directed toward methods for the production of a quality Blue cheese flavor concentrate; and for the qualitative and quantitative redundant. The purpose of this study was to develop a rapid technique for the quantitative analysis of methyl ketones by high pressure liquid chromatography. The improved pro- cedure was then applied to the routine, rapid determination of these compounds during curing of quick ripened cheese. 1 REVIEW OF LITERATURE The development of the unique peppery and sharp flavor of the cheese ripened with Penicillium roqueforti intrigued investigators for many years. More recently, the biosynthesis of the compounds responsible for Blue cheese flavor and the role of P. roqueforti in fermented cultures have been studied extensively. Different classes of flavor components in blue—veined cheese can arise from several sources. The important ones are deve10ped by the concerted action of numerous enzymes of P. roqueforti involved in lipid and protein metabolism. The biosynthesis of such compounds during cheese ripening was recently reviewed by Kinsella and Hwang (1977). The History, Chemistry, and Classes of Flavor Compounds in Blue Cheese Free Fatty Acids The "peppery and spicy" taste of Roquefort cheese, was initially studied by Currie (1914). He concluded that the short chain free fatty acids, caproic, caprylic, and capric and their readily hydrolyzable salts were responsible for the characteristic response of the tongue and taste receptors upon eating such cheese. Butyric, valeric, and 2 caprylic acids in Blue cheese were also reported by Thomasow (1947). Coffman et 31. (1960) analyzed dry Blue cheese and detected butyric, caproic, caprylic, capric, isovaleric, and heptanoic acids by using gas chromato- graphy. The presence of aromatic acids, p-hydroxyphenyl acetic, p-hydroxybenzoic, and benzoic acids was revealed in Roquefort cheese by Simonart and Mayaudon (1956b). Formic and/or acetic acids were identified by the same authors (1956a) from Roquefort cheese. Several investigators have obtained quantitative data for free fatty acids in Blue cheese. Morris 23 31. (1955) separated and quantitated butyric and caproic acids in Blue cheese using partition chromatography. The concen- tration of acetic, butyric, caproic, and higher molecular weight free fatty acids (average molecular weight of 200) in Blue cheese was reported by Sjostrom and Willart (1959). Using a combination of liquid-liquid column chromatography and gas-liquid column chromatography, Anderson and Day (1966) and Anderson (1966) analyzed domestic and imported blue-veined type cheeses and quantitatively measured the major free fatty acids from acetic to linolenic. Similar concentrations of free fatty acids were obtained by Blakely (1970) in quick ripened (QR) loose curd blue-veined cheese. The lipid and fatty acid content of Blue cheese was also studied by Fujishima gt 31. (1971). Harte and Stine (1977) reported the same free fatty acid (butyric-linolenic) patterns between QR and commercial Blue cheese. Their data are presented in TABLE 1. The data show the concentration of free fatty acids (FFA) in QR Blue Cheese after seven days ripening at 170C and following storage at 4°C for two weeks. During storage the concentration of certain free fatty acids (butyric, capric, lauric, and myristic) increased significantly. Table 1.--Free Fatty Acid Content of Quick Ripened and Commercial Blue Cheese. Free fatty acid (mg/kg) Quick ripened Initial, after After 2 wk curing 7 days at 40C Acid at 17°C Commercial Butyric 1,270 1,680 720 Caproic 490 490 60 Caprylic 420 360 70 Capric 820 2,240 870 Lauric 1,240 1,500 1,440 Myristic 4,090 4,660 5,700 Palmitic 12,500 12,500 12,700 Stearic 6,590 6,460 6,020 Oleic 13,900 13,000 13,800 Linoleic 790 750 610 Linolenic 710 650 630 Anderson (1966) showed a difference between the fatty acid content of Roquefort and imported Blue cheese. Roquefort lacked the strong pungent flavor and aroma of butyric acid and was more characteristic of caprylic and capric acid. These results were in accord with the compo- sition of sheep's milk which is used in the manufacture of Roquefort cheese. According to Hilditch (1956) sheep's milk is low in butyric and high in caprylic and capric acids compared to cow's milk. Similar results were obtained by Sadini (1963), Kuzdzal-Savoie and Kuzdzal (1963), and Benassi (1963). Lipid Metabolism (lipolysis) The processing of a quality Blue cheese is very much dependent upon the metabolism of the lipid substrate in the cheese. Triglycerides are progressively hydrolyzed to monoglycerides and free fatty acids. Triglycerides decrease from 96-98% of the lipids (about 35% of the cheese) in early stages of ripening to 75-80% of lipids (32% of cheese) at maturity (Kinsella and Hwang, 1976 and 1977). The extent of lipolysis is governed by the lipase activity. The lipase systems involved in the hydrolysis of triglycerides in cheese may be from 3 possible sources (Anderson, 1966). 1. The lipase native to milk. 2. The lipase from P. roqueforti 3. The lipase of other microorganisms The Lipase Native to Milk: Cause a significant amount of fat hydrolysis, par- ticularly, in the cheese from raw homogenized milk. Her- rington and Krukovsky (1939) postulated the presence of at least two lipases in milk, based on formaldehyde sensiti- vity. This postulation was confirmed by Roahen and Sommer (1940), and Peterson gt gt. (1943). Two general lipase systems, each composed of 2 lipase enzymes, were demon- strated by Albrecht and Janes (1955) in raw skim milk within the pH range of 5.0 and 6.6. Tarassuk and Frankel (1957) indicated two lipase systems native to milk, a plasma lipase associated with casein and a lipase absorbed on the fat globule membranes. Harper and Gaffney (1970) showed that lipase is an enzyme absorbed on various caseins and may form a multienzyme system in milk. A pure and homogenous milk lipase with a single optimum pH of 9.0 to 9.2 was isolated and characterized by Chandan and Shahani (1963a and 1963b). The optimum temper— ature for this enzyme was 370C. Schwartz gt gt. (1956) reported at least three pH optima: 6.5 to 7.0, 7.9, and 8.5 to 9.0 for characterization of milk lipase(s). Lipases with pH optima above 7.0 were reported to be less active in Blue cheese manufacturing (Bakalor, 1962). Gould (1942) showed that milk lipase will hydrolyze butter fat and most natural fats, oils and hydrogenated oils. Kelly (1944 and 1945) demonstrated the ability of milk lipase in hydrolyzing tributyrin, tricaproin, triace- tin, tripalmitin, ethyl oleate, and diacetin. Jenson (1964) reported that fats containing short chain acids in a mixture of triglycerides are attached preferentially by milk lipase. Lipases of P. roqueforti: Many researchers have examined the lipase of P. roqueforti and its importance in the ripening of cheese. Currie (1914) stated that "g. roqueforti produces a water soluble lipase which is the chief factor in the accomplish- ment of fat hydrolysis and free fatty acid accumulation in Roquefort cheese." Morris and Jezeski (1953) reported two lipase system in P. roqueforti, one with pH optima 7.0 to 7.8 on tributyrin and 6.5 to 6.8 on butter fat; the other lipase system, isolated from mycelial extract, had pH optima of 6.0 to 6.7 on tributyrin and 7.0 to 7.2 on butter fat. It was shown that the enzyme system was lipase and not an esterase. The effect of temperature, pH, type of sub- strate, and NaCl concentration on the activity of lipase system was demonstrated. Niki gt gt. (1966) studied an intra- and extra-cellular lipase system isolated from P. roqueforti in Blue cheese and reported two pH optima, 6.0 and 7.5, for both enzymes, indicating the possible presence of discrete enzymes in both preparations. Imamura and Kataoka (1963a and 1966) and Niki gt gt. (1966) showed that the extra-cellular lipase of P. roqueforti is more active than intra-cellular lipase. Higher lipase activity was shown (Eitenmiller gt gl., 1970) by growing 3. roqueforti in a medium containing "Casitone" broth rather than Czapek medium. Thibodeau and Macy (1942) and Morris and Jezeski (1953) had also obtained poor activity of the mold lipase on Czapek medium whose only nitrogen source was nitrate. Imamura and Kataoka (1963b) indicated that lipase production by P. roqueforti was inhibited by lactose, glucose, and galactose. However, the addition of butterfat increased the production of lipase. In contrast, Eitenmiller gt gt. (1970) observed a low lipase production by adding 1% (v/v) butter oil or corn oil to the growth medium. Niki gt gt. (1966) found the greatest lipase production at neutral or alkaline pH and temperature at 200C. In contrast, however, lipase pro- duction by g. roqueforti was reported by Eitenmiller gt gt. (1970) to be the greatest at lower pH (5.5) and a higher medium temperature (270C). The optimum temperature of lipase activity in g. roqgeforti was reported to be 30 to 35°C by Shipe (1951), and Morris and Jezeski (1953). Eitenmiller gt gt. (1970) presented an optimum temperature of 370C. Obviously, the variation in the optimum temperatures for the lipase acti- vity may be due to the difference in growth condition, fungal strain and analytical method used. The lipase sys- tem of g. rggueforti has been shown to be active at low temperatures of cheese ripening such as 10°C (Kinsella and Hwang, 1977). The ability of lipase to attack a mixture of tri- glycerides is of a practical importance from the standpoint of cheese ripening. Morris and Jezeski (1953) studied the lipase activity of g. rqqueforti on different triglycerides and found out that tributyrin and synthetic butterfat, composed of simple triglycerides with short chain fatty acids were hydrolyzed more rapidly by lipase than were natural milkfat or other substrate. Imamura and Kataoka (1963a) showed that extra-cellular lipase liberated caproic, capric, and caprylic acids selectively from butterfat. Shipe (1951) reported that the intra-cellular lipase of E. roqueforti hydrolyzed different triglycerides in the order of C4>C6>C8>C3. Eitenmiller gt gt. (1970) presented similar results. Salvadori and Salvadori (1967) reported that lipase from different strains of g. tgggg— tgttt exhibited different substrate specificity. This was confirmed by Stepaniak and Habaj (1974). Knight gt gt. (1950) produced a white mutant of g. rogueforti by irradi- ating the mold with ultraviolet light and showed that lipolytic activity of the mold was not associated with the green color of the parent mold. The effect of salt concentration on the lipolytic activity of g. roqgeforti was studied by Morris and Jezeski (1953) for both lipase systems isolated from mycelia or growth medium. As the sodium chloride concentration in the reaction mixture was increased to 4.0%, there was a rapid decline in lipolytic activity. This finding was in agree- ment with the results obtained by Thibodeau and Macy (1942) and Poznanski gt gt. (1967). Similar inhibitory effects of sodium chloride were observed by Gould (1941) in raw homo- genized milk and cream. Stadhouders (1956) reported that 10 the salt concentration in the range of 1-2% had more inhi- bitory effect on lipase than higher concentrations. Morris (1969) stated that a 6.0% salt concentration was inhibitory to g. roqueforti and no survival was noted at 10% salt con- centration. Eitenmiller gt gt. (1970) studied the effect of different salts on the lipolytic activity of g. roqueforti. These authors found stimulation of lipase activity by adding 10-2(M) MnCl or 10-3(M) MgCl to the medium. 2 2 Homogenization of milk prior to Blue cheese manu- facturing is a critical process in the hydrolysis of milk fat by lipase systems, particularly those from g. tgggg— tgttt. Morris gt gt. (1963) showed the importance of homo— genization with progressive accumulation of free fatty acids during Blue cheese ripening. The effect of homogeni- zation of milk on the manufacture of soft and white cheese was studied by Dozet gt gt. (1972). The triglycerides in non-homogenized raw milk are not readily available to lipase since an intact "membrane" surrounds the fat globule in discontinuous, multi-layers which inhibit attack and hydrolysis. On the other hand homogenization expands and alters the total surface of milk fat available for lipase attack (Kinsella and Hwang, 1976). Lipases of Other Microorganisms Present During Ripening of Blue Cheese: Although the hydrolysis of milk fat and accumulation of free fatty acids in Blue cheese is largely dependent upon 11 the lipase system of g. rgqueforti, some investigators have suggested that other microorganisms may be involved in the release of fatty acids from triglycerides during cheese ripening. Parmelee and Nelson (1947 and 1949) studied the lipolytic activity of organisms such as Candida ltpolytica, Alcaligenes lipotyticus, Achromobacter lipolyticum, Psuedo- monas fragi and Mycotorula lipolytica in the manufacture of Blue cheese. Only 9. lipolytica and M. lipolytica improved the Blue cheese flavor. Stadhouders and Mulder (1957) showed that the bacteria which survived in the cheese had little lipolytic activity and Willart and Sjostrom (1959) reported that the starter organisms did not exert lipolytic effects. Stadhouders and Mulder (1960) suggested that if the microorganisms principally represented in raw milk (such as Alcaligenes, Achromobacter, Pseudomonas, and Serratia) were added to pasteurized milk, considerable amount of lipolysis would occur in the cheese made with that milk. Fryer and Reiter (1967) indicated a weak lipolytic activity in several strains of Streptococcus lactis. Khan gt gt. (1966) studied the effect of microbial lipases on milk fat and suggested that the microbial lipases released larger amounts of unsaturated fatty acids and smaller amounts of short chain fatty acids compared to milk lipase. 12 I Methyl Ketones (2—A1kanones) The occurence of aliphatic methyl ketone was first established by Williams (1858) in essential oils of rue. Watts (1886) found the same methyl ketone, 2-undecanone, in the essential oil of lime leaves. Heptan-Z—one was shown to be present in Cinnamon oil by Walbaum and Huthing (1902). Haller and Lassieur (1910a and 1910b) identified several methyl ketones, 2-nonanone, 2-undecanone, and 2-tridecanone in the essence of Cocoanut. Undecan-Z-one was reported by Jasperson and Jones (1947) in palm kernel, palm, peanut, cotton seed and sunflower seed oil. Many reports have con- firmed a plant origin for methyl ketones (Forney and Marko- vetz, 1971). Milk and dairy products have been studied for methyl ketones by many researchers. Stérkel (1924) reported the occurence of methyl ketones in Roquefort cheese and organo- leptically showed that the characteristic flavor and aroma of this cheese was due to the presence of 2-heptanone and 2-nonanone produced by Penicillium roqueforti. Hammer and Bryant (1937) confirmed Starkel's findings and indicated that 2-heptanone was the major methyl ketone in Blue cheese. By means of ether extraction and fractional dis- tillation Patton (1950) isolated acetone, 2-pentanone, 2-heptanone, and 2-nonanone from Blue cheese, and stated that the distinct flavor of Blue cheese was due to the minute quantities of these ketones produced by the mold. 13 Morgan and Anderson (1956) identified odd-numbered methyl ketones, C3-Cll, in French Roquefort, Italian Gorgonzola, Danish and Domestic Blue cheeses by chromatographic analysis of 2,4-dinitrophenylhydrazone derivatives of methyl ketones. The same technique was used by Harper and Bassett (1959) to analyze neutral and acidic carbonyl compounds in Blue Cheese. Even-numbered methyl ketones are less important than odd-numbered ketones. Butanone-2 was isolated from Blue cheese by Morgan and Anderson (1956), and Harper and Bassett (1959). Bavisotto gt gt. (1960) reported the pre- sence of 2-octanone in Blue cheese. Several even-numbered ketones, 2-butanone, 2-hexanone, and 2-octanone were iden- tified by Nawar and Fagerson (1962). With combination of a gas chromatography and mass spectral analysis, the presence of methyl ketones: C3, C5, C6' C7, C8' C9, C10’ C11 and C13 was confirmed by Day and Anderson (1965) in blue vein cheeses. Using chromatographic methods Schwartz and Parks (1963) quantitated the odd numbered methyl ketones from C3 to C in domestic Blue cheese. Schwartz gt gt. (1963) made 15 similar measurements for Roquefort cheese. Heptanone-2 was the major ketone in all samples except one which contained more 2-nonanone. Anderson and Day (1966) found the concen- tration of individual methyl ketones, C3 to C11, in domestic Blue and Roquefort cheese and showed a significant variation 14 among samples. These authors showed that 2-heptanone was the most abundant ketone in all samples. With the exception of one cheese, acetone was detected in all samples of Roque- fort cheese. Several methyl ketones such as 2-pentanone, 2-heptanone, 2-nonanone, and 2-nonen (8,9)-one were reported by Svensen and Ottestad (1969) to be the major volatile com- pounds responsible for the flavor of Norwegian blue cheese (Normanna). Ney and Wirotama (1972) identified C3, C5, C7, C and C methyl ketones in a German blue cheese (Edel- 9 ll pilzkase). Groux and Moinas (1974) reported 2-heptanone to be the ketone in greatest concentration in Roquefort and 2-nonanone in Camambert cheese. Blakely (1970) analyzed quick ripened (7 days ripened) Blue cheese and found much lower concentration of methyl ketones, C3 to C13, compared with normal ripened Blue cheese. The same author reported that the major ketone in Danish and quick ripened Blue cheese was 2-nonanone with a range of 34.4-53.3 micromoles per 10 gram fat for Danish and 8.5 to 15.2 micromoles per 10 g. fat for QR Blue cheese. In the same study, he obtained a large concentration of 2-undecanone in quick ripened cheeses made from filled milk containing hydro- genated coconut oil. Albert (1974) found much higher con- centration of methyl ketones in several samples of quick ripened Blue cheese compared with Blakely's results. How- ever, a loWer content of methyl ketones was reported for commercial domestic Blue cheese by Albert. 15 Fatty Acid Metabolism and Methyl Ketone Formation Dakin (1908a and 1908b) theorized the biochemical conversion of free fatty acids to methyl ketones. The ability of several mold cultures, Penicillium glaucum, E. roqueforti, Aspergillus niger, and g. fumigatus, to convert free fatty acids to methyl ketones was initially presented by Stérkel (1924). This investigator isolated methyl 7, C9 and C11 from oxidative decomposition pro- ducts of cocoa fat and butter and compared them with ketones, C ketones obtained by oxidation of pure fatty acids with H202. Stbke (1928) and COppock gt gt. (1928) confirmed the conversion of individual fatty acids to their corresponding methyl ketones by pure cultures of a variety of fungi. Stake (1928) reported that Penicillium palitans and g. glaucum produced methyl ketones when they were grown on Coconut oil. The same author suggested that the noted Penicillium species, as in the higher animals and man, oxidize a chain compound primarily at the B-carbon atom with the formation of a keto acid and subsequent decompo- sition of the keto acid to methyl ketone and carbon dioxide. Mechanism of Methyl Ketone Formation by Fungi: It was initially demonstrated that only spores of g. roqueforti were capable of producing methyl ketones from fatty acids (Girolami and Knight, 1955; and Gehrig and Knight, 1958, 1961, and 1963). These authors studied l6 2-heptanone formation from octanoic acid by mold spores because of the importance of this compound in giving aroma and flavor to Blue cheese. The spores of various filamen- tous fungi such as Aspergillus and Penicillium genera were reported (Gehrig and Knight, 1961; and Franke gt gt., 1961) to be capable of producing methyl ketones from fatty acids. Gehrig and Knight (1963) concluded that the capacity of spores of g. roqueforti to form methyl ketones disappeared rapidly and progressively as the spores germinated. Lawrence (1965 and 1966) showed that the addition of those amino acids and sugars such as casamino acid, alanine, serine, proline, and glucose, xylose, galactose, and sucrose that stimulate germination of spores, markedly increased the slow rate of 2-heptanone formation from octanoic acid in a spore suspension of g. roqueforti. Hawke (1966) and Lawrence and Bailey (1969) suggested that the substances readily oxidizable by spores of g. roqueforti would stimu- late germination and ketone production. Dartey and Kin- sella (1973a and 1973b) showed that the addition of sub- stances such as D-glucose and L-proline supressed the cata- bolism of fatty acids to carbon dioxide and increased the methyl ketone formation. Hwang gt gt. (1976) showed an increase in the rate of 2-undecanone formation by adding glucose to the resting spores of g. roqueforti. The changes in morphology and biochemical properties of the spores of g. roqueforti during germination have been studied by Fan and Kinsella (1976) and Fan gt gt. (1976). 17 In contrast with the data reported by Gehrig and Knight (1958, 1961 and 1963), Lawrence and Hawke (1968) reported the formation of methyl ketones by mycelia of g. roqueforti from low concentrations of fatty acids with less than 14 carbon atoms over a wide range of pH. Dwivedi and Kinsella (1974a and 1974b) confirmed the findings of Lawrence and Hawke by carefully obtaining mycelia cultures and demonstrating the ability of mycelia to oxidize fatty acid and produce methyl ketones. Fan gt gt. (1976) showed the relative rates of methyl ketone formation from fatty acids by spores of g. roqueforti at progressive stages of germination. They concluded that under short incubation conditions, mycelia were much more active; however, the resting spores showed very high activity during a long period of incubation. Lawrence and Hawke (1968), and Dwivedi and Kinsella (1974a) demonstrated the mycelia that had been cultured over 48 hours showed a superior ability to metabolize fatty acids to methyl ketones as compared to fresh mycelia. Accumulated information of many investigators have revealed that methyl ketone production from fatty acids by spores and mycelia, follows the classical fatty acid B-oxidation cycle. Gehrig and Knight (1963) used radio- active labeled fatty acid, (1-14C) octanoate as substrate, and demonstrated that in small concentration (1 uM), sodium octanoate was oxidized completely by spores of 18 g. roqueforti; however, in larger concentrations (20 MM) it was partially oxidized to 2-heptanone and partly to carbon dioxide. The carbon dioxide was radioactive but 2-heptanone was not. Lawrence (1966) presented that 2-heptanone formed from (1-14C) octanoate was not radioactive, whereas that from (2-14C) octanoate was radioactive. On the other hand, by enzyme preparation from various fungi by Franke gt gt. (1961) had revealed the decarboxylation of B-keto acids to methyl ketones. Hawke (1966) outlined a schematic of the pathway of fatty acid oxidation for methyl ketone formation similar to the classical B-oxidation cycle of fatty acids as following: Pathway of Fatty Acid Oxidation and Methyl Ketone Formation +2H ‘__. RCHZCHZCOSCoA-—:§fir RCH—CHCOSCoA +H20 -H20 -2H RCOCH COSCoA -————w RCHOHCH COSCoA 2 +2H 2 Deacylase CoASH +H 0 / 2 RCOCHZCOOH + CoASH RCOSCoA + CH3COSCoA Decarboxylase RCOCH3 + CO2 Further formation C02 + H20 of CH COSCoA (via TCA cycle) 3 19 This mechanism was reported for: (1) only methyl ketones of one less carbon atom than the fatty acids used as substrates, and (2) a simultaneous process for B-oxidation of fatty acids. Kinsella and Hwang (1976 and 1977) presented a general outline of the fatty acid metabolism by the spores or mycelia of g. roqueforti as occurs in Blue cheese in the following: Lipids lipase(s) Fatty acids B-oxidation thiolase Keto Acyl-CoA ‘F’Acyl-CoA + Acetyl-CoA b / \ ./ Kreb Cycle Thiohydrolase \. // \\ \~ _— ,/ CO2 B-Keto Acid /;;Alkanol Decarboxylase Reductase 2-A1kalone These authors outlined the following steps similar to Hawke (1966): '1. Lipolysis of triglycerides to form free fatty acids. 20 2. Formation of the B-Keto Acyl-CoA by the dehydro- genation of the B-Hydroxy Acyl-CoA. 3. Deacylation of B-Keto Acyl-CoA and formation of B-Keto Acid and CoASH by a B-Keto Acyl-CoA deacy- lase (Thiohydrolase). 4. Rapid decarboxylation of B-Keto Acid to carbon dioxide and 2-Alkanones (methyl Ketones) by the B-Keto Acid decarboxylase. 5. Reduction of 2-Alkanones to their corresponding 2- Alkanols (secondary alcohols) by a reductase system. In above steps the acyl group is from 4 to 16 car- bons and the broken line denotes the possible inhibition of thiolase by acyl-CoA, thereby facilitating the deacylation- decarbdxylation step (Kinsella and Hwang, 1977). There have been many efforts to isolate the enzyme complex responsible for fatty acid oxidation from fungi, but they have failed due to difficulty in extracting and obtaining an active enzyme system from the mold. Franke gt gt. (1962) reported that B-decarboxylase was very active in several molds. Beta-keto Acyl decarboxylase enzyme was extracted from g. roqueforti by Hwang gt gt. (1976) and its activity was examined. They suggested the presence of 2 species of B-keto Acyl decarboxylase in the mold, one of which was heat stable and the other heat labile. It was demonstrated that this enzyme decarboxylates B-Keto-laurate 21 to 2-undecanone in the resting spores, germinating spores, and mycelia. The rate of methyl ketone formation in the mycelia was the greatest. The major intermediates of the fatty acid B-oxida- tion cycle, B-keto acids, were isolated from Blue cheese and studied by Bassett and Harper (1958). The esters of these keto acids were reported to be formed in butterfat in the presence of moisture (Wong gt gt., 1958). Van der Ven gt gt. (1963) demonstrated that the formation of pyra- zolones in butterfat during heating was due to the presence of nonvolatile B-keto acid esters. The same authors sug- gested that the six even number keto acids (C6 - C16) were the precursors of odd number methyl ketones (C5 - C15). The concentration of these keto esters as glycerides was only 0.03 percent. Schwartz and Parks (1963) reported that the existence of C13 or C15 ketones in Blue cheese was probably due to non-microbial breakdown of ketone pre- cursors. Studies have shown that B-Keto octanoyl-CoA is a preferential substrate for g. roqueforti in the deacylation step to produce 2-heptanone. However, methyl ketones of different chain length can be formed from longer chain fatty acids. Dartey and Kinsella (1973a and 1973b) used [U-MC] lauric acid and [U-14C] palmitate and showed the formation of a homologous series of odd number methyl ketones, C3 - C15, at successive cycles of B-oxidation. 22 Lawrence (1966) observed that the pH optimum for methyl ketone formation varied with the concentration and type of fatty acids. At a concentration of 1.0 mM octa- noate, the pH Optimum was between 5.5 and 6.0 and increased as the fatty acid concentration was increased. The yield of methyl ketone between pH 5.5 and 7.0 decreased pro- gressively from a maximum of 75% from octanoic acid to zero from myristic acid. Hwang gt gt. (1976) reported an optimum pH, 6.5—7.0, for 2—undecanone formation from B-ketolaurate which corresponded to the Optimum pH observed for methyl ketone production by Dartey and Kinsella (1973a). Gehrig and Knight (1963) suggested an optimum temperature of 250C for methyl ketone production. Lawrence (1966) indicated that the rate of methyl ketone formation reached a maximum at a temperature of 270C and then rapidly decreased with further rise in temperature. Dartey and Kinsella (1973a and 1973b) reported a 20-fold increase in methyl ketone formation as the temperature of incubation was dropped from 370C to 30°C. The oxygen tension and carbon dioxide evolved in the reaction medium of g. roqueforti are important in the rate of methyl ketone production. Gehrig and Knight (1963) showed no ketone formation when the spores of the mold were incubated in 100% nitrogen or carbon dioxide. Lawrence (1966) demonstrated an increase in the rate of formation of 2-heptanone from octanoic acid in the presence of small amounts of carbon dioxide evolved in the reaction medium. 23 Lawrence and Hawke (1968) indicated that the removal of carbon dioxide as it was formed during incubation of the mold decreased the yield of methyl ketones up to 50%. Kinsella and Hwang (1976 and 1977) suggested that carbon dioxide may actually enhance methyl ketone formation by inhibiting the acetyl CoA via Kreb cycle and, therefore, causing a "back up" of the fatty acid oxidation pathway. Langler and Day (1964) reported that there was no correlation between the relative amounts of ketones formed and their corresponding fatty acid precursors in milk fat. The results of Anderson and Day (1966) confirmed the above statement. It was found that the proportion of each methyl ketone produced in Blue cheese by g. roqgeforti did not depend directly on the amount of available fatty acid pre- cursor in the cheese. They showed that the proportion of acetone was relatively low compared with its precursor, butyric acid, and conversely the concentration of 2-hepta- none was high relative to its precursor, octanoic acid. High concentration of fatty acid precursors tend to decrease the complete fatty acid oxidation by g. roqueforti. This inhibitory effect of fatty acids was demonstrated by Lawrence (1966) on spores, and by Lawrence and Hawke (1968) on the mycelia of the mold. Spores are much more resistant to the inhibitory effects of fatty acids (Kinsella and Hwang, 1977). Optimum concentration of fatty acids for methyl ketone production is 1.0-5.0 mM for spores, and 1.0 24 mM for mycelia. The same authors suggested that the fatty acids may inhibit the thiolase enzyme of B-oxidation path- way, whereas they do not effect the ketoacyl-CoA deacylase (thiohydrolase) nor the B-keto acid decarboxylase. There- fore, in the presence of high concentration of fatty acids, the major products are methyl ketones. However at very low concentration of fatty acids, mycelia exhibit complete oxidation of fatty acids to carbon dioxide and water (Law- rence and Hawke, 1968). Thus, in Blue cheese manufacturing, strains of g. rggueforti with poor lipase activity produce much lower methyl ketones and strains with very active lipase system generate more ketones (Kinsella and Hwang, 1977). Secondary Alcohols (2-Alkanols) In addition to methyl ketones and free fatty acids, which appear to constitute the major flavor components of Blue cheese, secondary alcohols have also been identified. Stéke (1928) suggested the generation of secondary alcohols by P. roqueforti. It was reported that these alcohols are probably intermediate products of the mold oxidation pro- cess. Jackson and Hussong (1958) used paper and gas chro- matography and isolated 2-pentanol, 2-heptanol, and 2-nonanol from Blue cheese. They suggested that secondary alcohols are formed by reduction of methyl ketones of the same chain length. The concentration of secondary alcohols in Blue cheese was measured by Anderson and Day (1966). 25 Blue cheese contains minute amounts of secondary alcohols compared to their corresponding methyl ketones. Svensen and Ottestad (1969) reported that 2-pentanol and 2-nonanol were among major flavor compounds in Norwegian blue cheese (Normanna). Considerable variation in concentration of these alcohols were found, in agreement with the results of Anderson and Day (1966). Nawar and Fagerson (1962) also demonstrated the presence of 2-pentanol, and 2-heptanol in Roquefort cheese. Ney and Wirotama (1972) identified C5, C7 and C9 secondary alcohols in German blue cheese (Edel- pilzkése). Groux and Moinas (1974) observed significant amounts of secondary alcohols in Roquefort and Camembert cheese. Kinsella and Hwang (1977) suggested that when 2-alkanols are present in large concentrations in Blue cheese, they may impart a musty flavor in the cheese. Alcohols Other than 2-A1kanols Primary alcohols such as methanol, ethanol, 2-methy1 butanol, 3-methyl butanol, l-pentanol, and 2-phenyl ethanol were identified in Blue cheese by Anderson and Day (1966). It was suggested (Anderson, 1966; and Kinsella and Hwang, 1977) that these alcohols, and particularly methyl butanols, are probably formed by reduction of aldehydes derived from oxidative deamination of free amino acids present in Blue cheese. These alcohols may contribute to the fruity and nutty flavor of the cheese. Ney and Wiro- tama (1972) reported the presence of l-butanol and 26 iso-pentanol-l in German blue cheese (Edelpilzkase). Acety methyl carbinol and diacetyl were isolated from Roquefort cheese by Csiszar gt gt. (1956). They suggested (as is well known for most fermented milk products) that the pre- sence of diacetyl was due to the lactic starters metabolism. These compounds were also detected in domestic Blue cheese by Harper and Bassett (1959). Unsaturated alcohols such as l-octene-3-ol and 2-octene-l-ol are reported to be produced by penicillium species (kaminski gt gt., 1974). Groux and Monias (1974) identified l-octene-3-01 in Camembert and Roquefort cheese. Trace amounts were identified in Roquefort cheese. These alcohols, which are characterized by a musty, mushroomy flavor at high concentrations may be formed by the oxida- tion of linoleic acid (Kinsella and Hwang, 1977). Linoleic acid was found to be abundant in the mycelia and spores of g. roqueforti (Fan and Kinsella, 1976; Shimp and Kinsella, 1977). Aldehydes Acetaldehyde was detected by Morgan and Anderson (1956) and Harper and Bassett (1959) in Blue cheese. Ace- taldehyde, prepionaldehyde, and isobutyraldehyde were reported in Blue cheese volatiles (Nawar and Fagerson, 1962). Anderson (1966) identified acetaldehyde, 2-methyl propanal, 3-methyl butanal, and furfural in Blue cheese. It was suggested (Anderson, 1966; Kinsella and Hwang, 27 1977) that the presence of these aldehydes may be due to the action of enzymes involved in the oxidative deamination of amino acids. Kinsella and Hwang (1977) presumed valine and leucine to be the precursors of 2-methyl propanal and 3-methy1 butanal respectively. Ney and Wirotama (1972) reported the following aldehydes in German blue cheese (Edelpilzkase): acetaldehyde, propionaldehyde, butyralde- hyde, isobutyraldehyde, valeraldehyde, iso-valeraldehyde, and phenylacetaldehyde. Esters Anderson and Day (1966) initially reported the importance of methyl and ethyl esters in the overall com- posite flavor of Blue cheese. Prior to that, Bills and Day (1964) had suggested that excessive levels of ethyl butanoate and ethyl hexanoate were responsible for the fruity flavor defect of cheddar cheese. Ethyl esters of butanoate, hexanoate, octanoate, and decanoate were iden- tified in German blue cheese by Ney and Wirotama (1972). Anderson (1966) identified different analogs of methyl, ethyl, and isoprOpyl esters. This author also tentatively identified two cyclic internal esters, deltaoctalactone, and deltadecalactone. Jolly and Kosikowski (1975) identi- fied and quantitated five different lactones in Blue cheese containing added microbial lipase. The lactones are the result of internal esterification of hydroxy alkanoic acids in triglyceride molecules. 28 Other Classes (Miscellaneous) Flavor Components Numerous organic compounds have been identified in Blue cheese which may directly flavor or function as part of the "background" cheese flavor. Significant quantities of organic acids such as acetic, lactic, pyruvic, malic, and isobutyric have been reported in Blue and Roquefort cheese (Kinsella and Hwang, 1977). Acids such as succinic, ben- zoic, hydroxybenzoic, p-hydroxybenzoic, and p-hydroxyphenyl acetic have been found in cheese (Simonart and Mayaudon, 1956a and 1956b; and Schormfiller, 1968). Valerie (Thomasow, 1947), isovaleric and heptanoic (Coffman gt gt., 1960) acids have been reported in Blue cheese. Acidic carbonyl compo— nents such as beta-ketocaproic, beta keto-caprylic, beta- keto capric and alpha-acetolactic have been shown in Blue cheese (Harper and Bassett, 1959). Free amino acids and peptides undoubtedly play an important role in the overall flavor of Blue cheese. Amino acids, depending on their composition, may contribute to the sweetish or bitter aftertaste in Blue cheese (Kinsella and Hwang, 1977). Glycine, alanine, proline, and threonine have a sweetish taste. The aliphatic hydrophobic amino acids and cystine are bitter (Ney, 1974). Volatile amines such as methyl amine, ethyl amine, n-butyl amine, isobutyl amine, iso-pentyl amine, di-methyl amine, di-ethyl amine, di-n-propylamine, and di-n-butyl amine were identified in German blue cheese (Edelpilzkase) by Ney and Wirotama 29 (1972). Amines such as tyramine and tryptamine were detected in Roquefort, Stilton, and domestic Blue cheese (Voigt gt gt., 1974). The highest concentration, 1.1 mg/g, of tryptamine was reported in Blue cheese among different varieties of cheese tested. Sen (1969) found the concen- tration of tyramine in several varieties of blue-veined cheeses to be 27-520 micrograms per gram of cheese. The presence of tyramine in quick ripened Blue cheese was demonstrated by Kuehler (1974). This author reported 825 micrograms tyramine per gram of cheese after 39 days ripening. Commercial blue cheese, ripened for 90 days contained 40 micrograms tyramine per gram of cheese. The presence of benzene, toluene, and cresyl methyl ether, identified by Anderson (1966) were unexpected com- ponents found in Blue cheese. Protein Metabolism (Proteolysis) Proteolysis (degradation of proteins) is necessary and important in Blue cheese ripening. This biochemical process develops a proper teXture, and flavor background. Amino acids act as precursors of some flavor components, germination stimulants for spores of g. roqueforti, and enhancer for methyl ketone formation during Blue cheese ripening (Kinsella and Hwang, 1977). Active proteolytic enzymes in g. roqueforti which degrade casein of milk have been found. The differences in proteolytic activity of some strains of g..roqueforti were 30 observed by Funder (1949). The lipolytic activity of the same strains had been shown to be different (Macy and Thibodeau, 1942). Porks' g a_l_. (1959) showed that the strains of g. rggueforti, used singly or in mixtures, differed in their lipolytic and proteolytic activities. Proteolytic activity was demonstrated by formation of different amounts of albumin, peptones and amino acids, or by different amino acids which were liberated. The proteo- lytic activity of white mutant of g. rqqueforti produced by Knight gt gt} (1950) was shown to be independent of the green pigmentation of the original mold. Salvadori (1964) reported that proteolysis varied as much as five fold among 19 different strains of g. roqueforti. Niki gt gt. (1966) showed the strains with high protease activity, exhibited low lipolysis and vice versa. The same authors suggested that the strains of g. roqueforti with low proteolytic and high lipolytic activities produced a better quality Blue cheese in a shorter period of time. Modler gt gt. (1974a) observed different proteolytic activity among three strains of E. roqueforti. The bitterness of cheese has been attri- buted to the accumulation of peptones and peptides (Czulak, 1959). Niki gt gt. (1966) and Pelissier gt gt. (1974) related the bitterness with excessive protease activity. The presence of an intra- and an extra-cellular proteases with an optimum pH of 5.5 was demonstrated by Niki gt gt. (1966). The intra-cellular protease, isolated 31 from mycellium contributes tO the proteolysis throughout the ripening period, whereas the extra-cellular protease degrade casein during the early stages Of ripening. Modler gt gt. (1974b) isolated and examined a pure extra-cellular protease. Gripon and Debest (1976) studied the proteolytic activity of eight strains Of E. roqueforti. One or two exo—cellular proteases such as acid or alkaline carboxypep- tidase and alkaline amino—peptidase were detected electro- phoretically. The ability of different strains of g. roqgeforti tO hydrolyze proteins at acid, neutral, and alkaline pH was presented by the same authors. Gripon gt gt. (1977) showed the proteolytic activity Of an acid pro- tease from g. roqueforti and a neutral protease from g. caseicolum during curd ripening. Both enzymes exhibited high proteolytic activity and produced large amounts of pH 4.6-soluble nitrogen and non-protein nitrogen, but had little effect on production of free amino acids. The intra-cellular protease isolated from mycellium had a range Of pH optima, 5.5-7.0 (Niki gt gt., 1966). The extra-cellular protease exhibited a narrow range Of pH Optima, 5.0-6.0. Modler gt gt. (1974a) reported that pH Optima Of extracellular protease were different on different substrates. The Optimum pH with 1% BSA was 3, whereas, with casein an Optimum pH of 5.5 was Obtained. The same authors demonstrated a wide range of pH, 3.0-6.0, for the maximum stability of extracellular protease. Modler gt gt. E) 0E ti c1 CC PI S). gr (1t Rc ti fr bl We Th re 32 (1974a and 1974b) showed that.the Optimum temperature for extracellular protease activity was 45 to 460C. Two pH Optima, 3 and 6, for casein hydrolysis were found by Gripon and Bergere (1972) using a protease isolated from Pencil- lium. The Optimum temperature was found between 35 and 430C. Below 250C the activity Of protease falls Off. Therefore, ripening temperatures Of 10-150C minimize ex- cessive peptide and amino acid accumulation in Blue cheese. Addition of sodium chloride to the cheese also affects proteolysis. Gripon and Hermier (1974) reported the inhibition of extra-cellular protease by adding sodium chloride tO the medium. Protease activity of Penicillium species was progressively decreased by sodium chloride concentration above 0.5% (Kikuchi and Takafugi, 1971). Protein breakdown during ripening of Blue cheese was suggested to be 20 to 30%. The protein degradation of quick ripened Blue cheese was studied by Kuehler (1974) during 11 or 39 days ripening. Amino acids in several varieties Of Blue cheese, Roquefort, Gorgonzola, and domestic Blue cheese were quan- titated by Kosikowsky and Dahlberg (1954). The level of free amino acids in Camembert and Danish blue cheese (Dana- blue) was reported by Ismail and Hansen (1972). Danablue was characterized by a high level Of free amino acids. This was in agreement with the results of Hanni (1967) who reported the presence of free amino acids up to 38% of total 33 nitrogen. Kuehler (1974) found the concentration Of free amino acids in several quick ripened Blue cheese samples. The amino acid content increased to 28% Of the total pro- tein after ripening 39 days. This was compared with 16% after 11 days ripening. The rOle of other microorganisms in hydrolyzing proteins of Blue cheese has been of little importance. Morris gt gt. (1951) concluded that the surface microflora of Minnesota blue cheese accelerated the ripening, parti- cularly in the edged portions of the slimed cheeses. These microorganisms consisted primarily Of yeasts and some molds followed by the growth Of Cocci and rod forms Of bacteria. Hartley and Jezeski (1954) reported five groups Of bacteria in slimed Blue cheese. Bacterium linens and g. erythro- ggggg were predominant. Imamura (1960), and KnOOp and Peters (1971) indicated that Streptococcus and Lacto- bacillus species may initially produce free amino acids in the cheese to stimulate germination of the spores of E. rogueforti. The proteolytic activity Of Streptococcus lactis was studied by Gripon gt gt. (1977). Amino-pepti- dase activity was demonstrated in g. lactis, resulting in the accumulation of free amino acids. The same authors suggested a synergistic action between Penicillium pro- teases, resulting in the development Of peptides, and exopeptidase, in liberating free amino acids from the 34 available peptides. Thus, microorganisms other than B. roqueforti may be helpful in the production Of Blue cheese flavor components. Blue Cheese Flavor Concentrates The demand for the Blue cheese flavor in several foods such as salad dressings, processed cheese, snack foods, and baked goods has encouraged processors to pro- vide typical Blue cheese flavor concentrates produced by either aerobic fermentation methods of chemical formulation. The patent of Watts and Nelson (1963) described a fermentation method which consisted of incubating homo- genized-sterilized whole milk inoculated with the spores Of g. rggueforti. The mixture was shake—incubated at 21-250C for 96 hours to produce the desired flavor compo- nents of blue mold cheeses. Knight (1963) patented a fer- mentation process involving the inoculation spores of g. roqueforti in milk containing lipolyzed milk fat or free fatty acids. Litchfield (1969) and Nelson (1970) have described the production Of Blue cheese flavor by large scale submergion methods. The concentration of methyl ketones is ten fold larger in these cultures than is found in Blue cheese. Luksas (1973) patented a method Of inocu- lating P. roqueforti in a medium containing 10% sodium caseinate and 5% butter fat. Agitation, aeration, and incubation for 48 hours at 25°C produced the desired level Of flavor compounds. Dwivedi and Kinsella (1974a) studied 35 methyl ketone and total carbonyl production in lipolyzed milk fat by continuous production of g. roqueforti mycelia in a fermentation process. It was Observed that octanoic and decanoic acids were easily metabolized by the mold tO produce C7 and C9 the concentration of total carbonyls. Based on their methyl ketones. Sodium chloride lowered findings the same authors (1974b) designed a two stage, semi-continuous method of submerged fermentation by E. roqueforti. This system was suggested to be helpful in Obtaining a desirable flavor balance to suit an individual food product. More recently, Jolly and Kosikowski (1975) demonstrated a submerged fermentation method with micro- bial lipase and g. roqueforti, using acid and sweet whey with butter fat or coconut Oil. The freeze-dried powder of the liquid concentrate served as a good product with a ketone content almost five fold higher than Blue cheese. Several investigators have attempted to formulate a synthetic Blue cheese flavor. Anderson (1966) suggested that an acceptable Blue cheese flavor could be Obtained by blending selected fatty acids, methyl ketones, secondary alcohols, 2-pheny1 ethanol, and methyl and ethyl esters in a cottage cheese and butter fat mixture. Moinas gt gt. (1974) prepared a Blue cheese flavor concentrate con- taining 0.8% 2-heptanone, 18% 2-nonanone, 14% 2-heptanol, 0.2% phenol, 63% propionic acid, 27% l-Octene-3-Ol, and 2% methyl cinnamate. A combination of 38 flavor components 36 were used (Ney gt gt., 1975) tO Obtain a desirable Blue cheese flavor. Addition of 1-Octene-3-Ol tO the mixture was reported to improve the synthetic Blue cheese flavor. Accelerated (Quick) Ripening Of Blue Cheese A number of investigators have developed and described several methods for quick ripening or shortening the period during which desired texture, color, and flavor are Obtained in Blue cheese. According to Mocquot and Béjambes (1960) French Roquefort blue cheese needs a ripening period of 3 months in Roquefort caves with a temperature of 43-46°F. Peters and Nelson (1948), and Babel (1953) reported roughly a 3 month period for Blue cheese, ripened at 50° or 50° - 55°F and 90-95% relative humidity. Lawrence (1966) suggested that in a typical Blue cheese, a period of 90 days is required for g. tgggg- tgttt to develop methyl ketones from fatty acids to a desired concentration. Jolly and Kosikowski (1975) manu- factured Blue cheese cured 45-75 days at 5°C and demon- strated maximum methyl ketone and flavor development during this curing period. A method was developed by Kondrup and Hedrick (1963) and patented by Hedrick gt gt. (1968) to shorten the ripening period of Blue cheese from the initial 90 days to 10 days. The mold was incorporated into the milk at the time Of rennet coagulation and the Open granu- lar curd rather than the normal hooped loaf was ripened at 62°F and over 95% relative humidity with the addition of up 37 tO 4% salt. The curd was agitated daily until a desirable texture, color, and flavor was Obtained. Blakely (1970) analyzed a 7-day ripened Blue cheese made by the same pro— cedures and reported poor flavor development. A modified procedure Of Hedrick gt gt. (1968) was used by Kuehler (1974), Harte (1974) and Albert (1974) to produce quick ripened Blue cheese with more desirable quality. These authors also used a white mutant of g. rgqueforti (Knight gt gt., 1950) and obtained a satisfactory quick ripened cheese with a Blue cheese flavor. The manufacture Of a Blue cheese made from soya milk and cured for two weeks was patented by Lundstedt and Yau-Yee LO (1973). The soya milk was fortified with milkfat and nonfat milk solids and ino- culated with g. roqueforti and Streptococcus diacetilactis. The latter organism aids the growth of the mold. Homogenization reduces the size Of the fat globules in milk, thus increasing the area of the reactive surface exposed to lipolysis. Lane and Hammer (1937) Observed a uniform, faster ripening and flavor development in the cheese made with homogenized milk compared with unhomo- genized raw milk. Homogenized-pasteurized milk showed a rapid development of volatile acids and flavor characteri- stic of the cheese. A soft and brittle curd resulted in the cheese made with milk homogenized at 2000 Psi. Morris gt gt. (1955) reported a 9-11 months ripening period for Blue cheese made with unhomogenized raw milk. This period was reduced to 4-5 months by using raw or pasteurized 38 homogenized milk to Obtain the "best flavor." Raw homo- genized milk resulted in a higher accumulation of free fatty acids in Blue cheese compared with unhomogenized milk. However, the use Of raw homogenized milk resulted in a less mold growth and smaller change in pH. Peters and Nelson (1961a) Obtained similar trends, by using treatments such as 2000 Psi homogenization and 143°F/30 minutes pasteurization, while studying volatile free fatty acid changes during ripening. The quality of Blue cheese at different ripening temperatures was also studied by the same authors (1961b). An improved quality and faster ripening of Roquefort cheese was found by Shchedushnov (1961 and 1962) when a pasteurized mixture Of homogenized cream and skim milk was used compared with unhomogenized controls. It was shown (Shchedushnov, 1961) that the break- down Of protein and fat and the formation of volatile acids increased in homogenized samples during ripening. The homogenization Of milk had taken place at 50-100 atmospheres pressure. The highest quality cheese was Obtained from milk homogenized at 50 atmospheres pressure. Lane and Hammer (1938) reported a normal Blue cheese ripening and a satisfactory growth Of g. roqueforti when milk was homo- genized two stage up to 2500 or 3500 Psi at 95°F. In both cases 500 Psi was used for the second stage. With 3500 Psi two-stage homogenization of milk, a rapid flavor development and less color was Obtained. Babel (1953) suggested a mixture Of homogenized cream, at 2500 Psi, with skim milk 39 for manufacturing American blue cheese. Kondrup and Hedrick (1963) used two stage homogenization, 1400 Psi on the first and 700 Psi on the second stage, of whole milk at 130°F for manufacturing a 10 days - QR Blue cheese. Many investigators have tried tO accelerate the period of Blue cheese ripening and the flavor development by addition of enzymes, particularly lipases. Coulter and Combs (1939) tried tO shorten the ripening period Of Blue cheese by adding steapsin to the milk or curd. The same degree Of flavor development was Obtained after 5 months compared with 12 months ripened commercials. However, sen- sory evaluations indicated the cheese to be inferior to normal cheese due to a characteristic bitter flavor. WOjtkiewicz and InikkOff (1934) attempted to accelerate the ripening Of cheese by adding pepsin, trypsin, erepsin, and rennet individually or in various combinations. Faster ripening was Obtained but cheese containing trypsin had a bitter flavor. Thibodeau and Macy (1942) produced a fine quality Blue cheese by using the enzymes from g. tgggg- tgttt in the form of mycellium homogenate. The ripening period was decreased to 5 months compared with a 10 month commercial cheese. The addition of lipolytic organisms such as Candida lipolytica to the pasteurized milk enhanced ripening and improved flavor (Parmelee and Nelsen, 1949). The same authors (1947) had indicated that not all lipolytic organisms, when added to the milk, caused an improvement in the flavor quality of Blue cheese. 40 Achromobacter lipolyticum and Alcaligenes lipolyticus were among the microorganisms which did not affect fat hydrolysis or flavor development. Peters and Nelson (1948) demon- strated that the addition of cell-free lipase from Myco- torula lipolytica to pasteurized-homogenized milk increased the hydrolysis of fat and intensified the desired flavor in Blue cheese. Lipolytic agents such as pancreatin, erepsin, steapsin, mulberry juice lipase, and chicken stomack lipase were reported (Kosikowski and Mocquot, 1958) to cause ran- cidity and bitter flavor in cheese. Nilson and Willart (1960 and 1961) reported an acceleration in flavor develOp- ment of Blue cheese when lipases from Candida lipolytica and pancreatic substrate were added to milk, but bitterness resulted in the cheese. The rate Of flavor development, particularly methyl ketones, was enhanced by adding micro- bial or animal lipase to the pasteurized milk (Jolly and Kosikowski, 1975). These results confirmed their findings (1973), showing that Blue cheese prepared with that method was ripened significantly faster than controls. The same authors (1975) added microbial lipases tO a submerged fer- mentation Of g. roqueforti and Obtained a fast release Of fatty acids, making them available in higher concentration for the mold, for subsequent conversion to methyl ketones in a Blue cheese flavor product. 41 Cheese Dehydration and Flavor Retention Dehydrating natural cheese for preservation has been studied by many investigators. Cheese can be dried by techniques such as tray drying, spray drying, and freeze drying to produce cheese powder or products suitable for grating. The reduction in weight and bulk is an advantage of dehydration, as is storage stability at room temperature. Depending upon the mode of drying, dried cheese may contain between 3 and 18% moisture (Kosikowski, 1977). During dehydration or evaporation Of foods, particularly heat- sensitive foods such as dairy products, loss of volatile components and changes in chemical constituents of food take place. Dehydration of cheese was initially reported by Sanders (1943) tO produce a cheese powder of 3% moisture or less for use by the Armed Forces. Cheese drying tOOk place in two stages. First, partial drying of grated cheese at room temperature by blowing an air current above it to prevent lumping Of particles, exudation Of fat and loss of flavor. Second, the partially dried cheese was exposed to a temperature of 145°F until the desired cheese flakes were obtained. Yakhontov (1956) dried a Soviet cheese in thin layers with air current at 38-42°C during 6.0 hrs. The final moisture was 10% and the cheese required refrigeration at 10-15°C 70-75% relative humidity. Olsanski and Schmidt (1960) reported dehydration Of 42 Parmesan cheese by high quality by using dark infra-red lamps with filters for exclusion of visible light. Faster drying and a better keeping quality were Obtained compared with bright infra-red lamps or drying with warm air current. Spray drying of cheese slurry was reported for the first time by Bohéc (1957) and Irvine gt gt. (1957). The latter authors spray dried a "low fat" Cheddar cheese emul- sion which could be stored at 50°F for 6 months. According to Bullock et a1. (1963) the addition Of 5% Blue cheese to the cheese emulsion improved the flavor Of the powder. The Cheddar cheese used for spray drying contained 20% fat. Slurries of cheddar and Blue cheeses (40% solids) were spray dried by Bradley and Stine (1962) in a commercial type horizontal, cocurrent dryer. The samples were homo- genized at 2000 Psi before drying. The same authors (1963) reported a better flavor retention by drying larger parti- cle cheese powder. These authors (1964) demonstrated foam spray drying of cheddar cheese by incorporating nitrogen gas into the slurries prior to atomization of the product into the spray-dryer. The larger particle diameter of cheese powders resulted in an improved flavor quality and better retention of volatile compounds. The incorporation of nitrogen enhanced the shelf life Of the product. Hedrick (1968) reported that spray dried QR Blue cheese exhibited more intense flavor if an extended ripening period was employed. 43 Freeze drying was applied initially tO milk and dairy products by Nickerson gt gt. (1952). The same flavor quality and shelf life were reported for freeze dried as did for spray dried whole milk and it was concluded that freeze drying of whole milk was not superior tO spray dry- ing. The application for freeze drying Of liquid milk products was patented by Ogden (1967). Flosdorf and Hamilton (1957) claimed that a better freeze drying process for cottage cheese was achieved at a slower freezing rate. It was reported that better wettability and porosity were Obtained if the cheese was slowly frozen over a 30 minute interval. Several varieties of cheeses such as Blue, Cream, Cheddar, Brick and Munster were lyOphilized (freeze dried) by Meyer and Jokay (1959). The dried Blue or Cream cheese samples exhibited poor rehydration properties and had inferior shelf life. Jokay and Meyer (1959) developed a satisfactory processing technique for freeze drying of Cottage cheese. Schultz (1964) studied freeze drying Of process cheese and natural cheeses such as Quarg, double Cream cheese, Edam, Camembert, Gouda, Emmental, Romadur, Filsit, and Danish Blue. Cheese Of higher initial moisture content, such as quarg, had better reconstitutability. The loss of volatile flavor component in a pure state during an evaporation or drying process is directly proportional to its vapor pressure. However, in aqueous solutions the loss of a volatile compound is dependent upon its vapor pressure, solubility, presence Of salts and 44 sugars, and miscibility of the compound with other organic constituents. Volatility of a flavor compound results from high vapor pressure and low solubility in its solvent. Com- pletely soluble substances in aqueous solutions such as acetic acid have a pattern of water-liquid equilibria of two miscible compounds during evaporation (Saravacos, 1968; and Hawrysh, 1970). During spray drying of butter containing known amounts of volatile fatty acids it was observed (Boudreau gt gt., 1966) that the degree of retention of longer chain fatty acids was greater than short chain fatty acids. It is concluded (Boudreau gt gt., 1966; Saravacos, 1968; and Saravacos and Moyer, 1968) that generally, as the volatility of compound is increased, flavor retention in a dried pro- duct is decreased. Gas chromatographic studies by Bradley and Stine (1968) showed that the retention of volatile compounds in foam-spray dried Cheddar increased as the particle size of the powder increased. The retention of flavor compounds was greater in foam spray-dried than conventional spray dried samples. However, the retention of flavor compounds added to skim milk was found (Reinec- cius and Coulter, 1969) to be independent of the particle size of the resultant powders. Desai (1966) indicated that volatile acids and flavor components such as diacetyl were retained to a greater extent in freeze-dried sour cream than in foam spray powders. 45 Saravacos and Moyer (1968) observed that the reten- tion of flavor components in vacuum dried fruit juices was affected by the temperature of drying, the vapor pressure of compound, the relative volatility and the composition of the juice. These findings were confirmed by the results of Hawrysh (1970) who studied the effect of freeze drying process on the retention of volatile monocarbonyl flavor compounds such as methyl ketones, saturated aldehydes, and unsaturated aldehydes (2-enals) by incorporating them into cream with 30.5% milk fat. During lyophilization, short chain aldehydes were retained to a greater concentration than short chain methyl ketones. Retention of both classes of aldehydes was greater than methyl ketones in freeze- dried samples. It was also suggested that the recovery of high molecular weight monocarbonyls from freeze-dried samples may be determined by factors such as boiling point, solubility in the lipid phase of the cream, molecular size, and properties of absorption and adsorption. Blakely (1970) observed that the loss of volatile fatty acids during freeze-drying of Blue cheese was related to their boiling points. The lower the boiling point, the greater was the loss of volatile acids during dehydration. The re- tention of fatty acids in freeze-dried Blue cheese was greater than spray dried samples. The same author showed that the retention of methyl ketones with less than 10 carbons was not related to the boiling point of the 46 component. Lower retention of methyl ketones was found in freeze dried than spray dried Blue cheese. EXPERIMENTAL PROCEDURES Production of Quick Ripened (QR) Cheese Blue Mold Powder Blue mold powder of g. roqueforti, Code #65001 was Obtained from Midwest Mold Powder (Dairyland Food Labora- tories, Inc., Waukesha, Wisconsin 53186) and used directly in the cheese. Preparation of White Mold Powder Spores of the white mutant of E. roqueforti were prepared by the method of Hussong and Hammer (1935). The mold was isolated from a slant obtained from Midwest Mold Company and transferred several times on acidified potato dextrose agar slants. Whole wheat bread (no additive added) was cut into cubes, approximately 0.5 inches and sterilized in wide mouth jars at 15 lbs pressure for about 30 minutes. After cooling to room temperature a suspension of spores in water was prepared from the slants and inocu- lated into the sterilized bread. The inoculated samples were incubated at 21°C for 12 days for spore formation. During incubation the bottles were shaken occasionally. After spore formation the bread pieces were dried in an oven at 339-4o°c for 36-48 hours (Pulay, 1959) and 47 48 pulverized in a sterile mortar and pestle. The dried mold was sieved through a 45 mesh sieve (United States Standard Testing sieve) and stored at 4°C for later use. Manufacturing Procedure Used for Quick Ripened (QR) Cheese QR Cheese using 3. roqueforti or the white mutant of E. roqueforti was manufactured according to the method of Kondrup and Hedrick (1963), and Hedrick gt gt. (1968). Suggested modifications of Kuehler (1974), Harte (1974), and Albert (1974) were used with minor changes. Cheeses were made from 20 gallon lots of milk. Mixed herd raw whole milk was obtained from Michigan Milk Producers Association. The milk was preheated to 135°F and homogenized in a two stage homogenizer (2000 Psi first stage, and 500 Psi second stage). The milk was then pasteurized at 143°F for 30 minutes, cooled to 88°F and transferred to a fifty gallon cheese vat. A frozen concen- trated cheese starter (Hansen's Lab, Inc., Milwaukee, Wis- consin 53214) was added at the rate of 17 grams per batch. When the acidity increased 0.020-0.025%, 24 grams of mold powder and 10 ml single strength rennet were added to the milk with agitations. The coagulum was cut with 3/8" knives and allowed to idle for 10 minutes. The curd was stirred gently and heated to 104°F for approximately one hour. The drained curd was placed on a fine mesh stainless steel screen in the ripening room to a depth of 3 to 4 inches. The curing room was maintained at over 95 percent 49 relative humidity(RH) and temperatures of 52 i l°F or 62 1 1°F for a period of 7 to 15 days. The cheese curd was stirred twice the first day and once each day during the ripening period to prevent matting. Salt was added to the curd in the curing room at a rate of 1.0 lb per 22 lbs of cheese. The sequence for adding salt was: t, k, and 8 of total salt on a varying three day schedule during ripening. Five different lots of cheese were prepared and cured (see 0 Table 2), and then packed and stored at 400, 48 or 62°F for a period 4 to 15 days after the initial curing period. Table 2.--Process Variables Used in Preparation of Quick Ripened Blue Cheese. Cheese Curing Salting, Curing lot mold Tempegature, days period, F days 1 white 5211 1,6,9 11 or 15 2 white 52:1 6,7,8 11 or 15 3 blue 5211 1,9,10 11 or 15 4 blue 5211 7,8,9 11 or 15 5 blue 6211 4,5,6 7 Commercial Blue Cheese Commercial imported and domestic Blue cheeses were obtained from local markets in E. Lansing and were analyzed as obtained. Freeze Drying of Commercial and Quick Ripened Blue Cheese Quick ripened Blue cheese and commercial Blue cheese were freeze dried in a Virtis Repp Model 42 50 freeze-dryer. The freeze dryer was equipped with automatic condenser and platen temperature and absolute pressure controls. The samples were layered to a depth of 3/4" in 11 1/4 x 7 1/2 inches aluminum trays. Then the fresh cheese was frozen overnight to -10°F and dried in the freeze dryer at a platen temperature of 100°F. The con- denser temperature was set at -80°F. The vacuum of the 3 Torr (McLeod Gauge) during 24.0 hours chamber was 5 x 10- drying period. After conclusion of the drying, the vacuum was released with nitrogen and the dried cheese was packed in polyethylene bags and stored at 4.4°C. ANALYTICAL PROCEDURES RE The pH of the cheese samples was measured with a Chemtrix digital pH meter, Type 60A, to 10.01 pH unit. Moisture Moisture content of samples was determined according to the method of the Association of Official Analytical Chemists for cheese (AOAC, 1975). Liquid Chromatographic Analysis of Methyl Ketones Preparation and Purification of Solvents Hexane: Carbonyl-free hexane was prepared from redistilled hexane by the methods of Hornstein and Crow (1962), and Schwartz and Parks (1961). Forty five milliliters of con- centrated sulfuric acid were added to 75.0 grams of celite 545 (Johns Manville) and blended to a homogeneous mixture. The acid-celite was packed in a chromatographic column, 25 x 500 mm, equipped with a 500 ml reservoir at the tOp and a stopcock (teflon) to control the flow rate. A piece 51 52 of glass wool was placed in the bottom of the column. Each portion of the acid-celite mixture was tapped and firmly packed with a glass rod. Anhydrous granular sodium sulfate was used to cover the lower and upper levels of the mixture in the column to a depth of about 10 cm. Redistilled hexane was passed over this column at a rate of 3-5 ml per minute. The eluate from this column was passed over a 2,4 DNPH-reaction column (Schwartz and Parks, 1961). Five- tenths gram of recrystallized 2,4-dinitrophenylhydrazine (2,4-DNPH) was dissolved in 6.0 milliliters of 85% phospho- ric acid by grinding in a mortar until a clear yellow solution was obtained. Four milliliters of deionized distilled water was added and the precipitated DNPH was redissolved by grinding. Ten grams celite (Johns Manville Analytical grade) was ground with the solution until a homogeneous bright yellow blend was obtained. The mixture was packed in a 25 x 500 mm chromatographic column contain- ing 25 ml carbonyl-free hexane. Four equal portions of the mixture were added separately to the column and tapped firmly. A piece of glass wool was placed over the last portion to prevent disturbing the column materials at the upper level. The column impurities were flushed with 50 ml carbonyl-free benzene, followed by 50-100 ml carbonyl free hexane. The eluate from the reaction column was mixed with Sea Sorb 43 (Adsorptive magnesia, iodine no. 80, Johns 53 Manville) as suggested by Blakely (1970). The Sea Sorb- hexane mixture was filtered through a fluted filter paper to remove the carbonyl derivatives from hexane. Carbonyl free hexanes were stored in brown bottles. Benzene: Carbonyl-free benzene was prepared by refluxing reagent grade benzene over 2,4-dinitropheny1hydrazine for 2.0 hours. Two grams of 2,4-DNPH per liter of solvent was used (Hawrysh, 1970). The benzene was redistilled and was stored in brown bottles. Chloroform: Reagent grade chloroform was refluxed over KOH pellets for 30 minutes and then redistilled. Alternatively, distilled in glass-chloroform (obtained from Burdick and Jackson Lab, Inc., Muskegon, Michigan) was used directly. Nitromethane: Nitromethane (Fisher Scientific Co.), or spectro- photometric grade nitromethane (Mallinckrodt, Inc., St. Louis, Missouri) was redistilled over boric acid. Puri- fied solvent boiling between 100.7-101.5°C was collected in brown bottles. Acetonitrile: High purity acetonitrile, 81-82°C boiling range, was obtained from Burdick and Jackson Lab, Inc. 54 Ethanol: Carbonyl free ethanol was prepared by distilling 95% ethanol over 2,4-dinitrophenylhydrazine, using one gram DNP per 1000 ml ethanol. Ethylene Chloride: Reagent grade 1,2-dichloroethane was obtained from Mallinckrodt, Inc. and redistilled over anhydrous potassium carbonate and stored in a glass bottle (Anderson, 1966). ButanOl-l: Distilled in glass Butanol-l was obtained from Burdick and Jackson, Inc. Purification of 2,4-dinitrgphenylhydrazine The 2,4-dinitrophenylhydrazine reagent was obtained from Eastman Kodak, Inc., or Pharmaceuticals, Inc. and recrystallized (Vogel, 1962) to use in this study. The 2,4-DNPH was added to butanol-l, at a concen- tration of 1:30 and heated on a steam bath with occasional shaking until most of the reagent was dissolved. The hot solution was filtered through a fluted filter paper and held overnight for recrystallization. Crystals were separated from the solvent by filtration through a Buchner funnel and dried at room temperature in a desiccator. Dried crystals were stored in a brown bottle at 4.4°C. 55 Preparation of 2,4-DNPH Derivatives of Standard Methyl Ketones Standard methyl ketones, 99% purity, were obtained from Analab, Inc. (North Haven, Conn. 06473), or Aldrich Chemical Co., Inc. (Milwaukee, Wisconsin) and were used to form hydrazone derivatives according to Shriner and Fuson, and Curtin (1964). Four m1 concentrated sulfuric acid were added to 0.8 grams of recrystallized 2,4-dinitrophenylhydra- zine. Then 6.0 milliliters distilled water were added dropwise to the mixture with swirling until solution was complete. Twenty m1 carbonyl-free ethanol were added to the mixture and this was called solution (A). A solution of standard methyl ketone (B) was prepared by adding 1.0 gram standard methyl ketone to 40 milliliters of 95% car- bonyl free ethanol. Solution (A) and (B) were mixed in a beaker until precipitation occured. The 2,4-DNP-hydra- zones were separated by filtration and dissolved in 60 ml of carbonyl-free ethanol and the mixture was heated on a steam bath until a complete solution was obtained. A few drops of distilled water was added until a cloud point was reached. Then the hot solution was filtered rapidly and allowed to stand overnight for recrystallization. If hydra- zones were not dissolved in ethanol, ethyl acetate was added up to 2.0 milliliters to obtain complete solution. The recrystallized 2,4-DNP-hydrazones were filtered, using a Buchner funnel, dried at room temperature, and then stored in the freezer for later use. 56 Isolation of Monocarbonyls from Cheese Samples Extraction of Fat: Approximately 250 g of each sample of cheese was ground thoroughly using a mortar and pestle. Duplicate analysis were performed from these representative samples for methyl ketone determination. Ten grams of cheese sample were mixed with 15 grams celite 545, dried 24 hours at 150°C, and ground to a homogeneous blend. The mixture was packed into a 25 x 500 mm chromatographic column equipped with fritted glass filter and then tapped gently with a glass rod. The fat was extracted with 200 milli- liters carbonyl-free hexane. A similar column was packed and extracted with 200 milliliters redistilled hexane into a 300 milliliters pre-weighed soxhlet flask to determine the fat recovery. The same flask was connected to a rotary evaporator and after removing the solvent from the fat, the flask was dried in a 100°C oven for 20 minutes, cooled in a desiccator, and re-weighed to calculate recovery. Formation of 2,4-DNP-hydrazone Derivatives: A 2,4-DNPH reaction column was prepared according to Schwartz and Parks (1961) to convert the carbonyl com- pounds present in fat extract to their corresponding 2,4-DNP-hydrazones. The hexane extract of the cheese sample was passed through this column. When the 200 m1 hexane extract passed the column, 200-300 ml carbonyl free 57 hexane was added. The column was flushed using 1-3 psi nitrogen pressure until the eluate had the same spectral properties as before addition of the sample. The solvent was removed by rotary evaporator. Removal of Fat from 2,4-DNP- hydrazones: The fat containing hydrazones were dissolved in 5-10 ml carbonyl free hexane and the 2,4-DNP-hydrazones were isolated from fat as described by Schwartz gt gt. (1963) and Anderson (1966) with slight modification. Eighteen grams Sea Sorb 43 (adsorptive magnesia, iodine No. 80) were mixed with 36.0 grams celite 545, dried 24 hours at 150°C, in about 250 ml carbonyl free hexane. The slurry was poured into a 25 x 500 mm chromatographic column equipped with a coarse fritted glass disk and a connected 500 m1 reservoir. The column was packed using 3-5 psi nitrogen pressure. The fat containing hydrazones was care- fully applied to the column. When the 2,4-DNP-hydrazones were adsorbed on the column as a deep brown band, the fat was washed off the column by subsequent addition of 200 ml hexane, 100 ml hexane:benzene (1:1 v/v), and 200 m1 benzene using 1-3 Psi nitrogen pressure. The hydrazones were eluted from the column by adding 250 ml chloroform-nitro- methane (3:1 v/v). The eluate was evaporated in a rotary evaporator to remove the solvents from DNP-hydrazones. 58 Fractionation of 2,4-DNP-hydrazones on Weak Alumina: The monocarbonyl derivatives were separated from ketoglycerides as described by Schwartz gt gt. (1963) on a weak alumina column (Schwartz and Parks, 1961). Alumina, grade F-20 (Aluminum Co. of America) was activated by heating 24 hours at 150°C and then partially deactivated by the addition of 6% (w/w) distilled water, mixing, and storing in a bottle for 24 hours to equilibrate. Five grams of partially deactivated alumina were packed in a 12 x 300 mm chromatographic column in hexane. The DNP-hydrazones described previous step were dissolved in 5 m1 hexane and applied carefully to this column. The monocarbonyl deri- vatives were eluted by adding 50 m1 hexane-benzene (1:1 v/v). The eluate was evaporated in the rotary evaporator. Determination of Methyl Ketones bngigh Pressure Liquid Chroma- togrgphy (HPLC) HPLC Apparatus: A high pressure liquid chromatographic system was used to isolate and quantitate the methyl ketones present in the monocarbonyl fraction of the cheese samples. A schematic of the HPLC system is shown in Figure 1. HPLC Specification: The components of the assembled HPLC system (Figure l) were composed of the following: 59 .Emumxm ADAL:V oflcmmu00umE0uno oflsvfla musmmmum 56H: mo OHumEmnom 3mm; 0... ii w r [m o “wouoowm LODMOOE .>.D CEDHOU Moxumz uumno O. O whoom whom sofluomncH QEsm Houhflum Ofiuwcmmz uflo>ummmm ucm>aom HOmCOOGOU v-INMVLD .H musmflm E N L 0.0625 port, 60 One liter solvent reservoir equipped with a con- denser and a magnetic stirrer. The reservoir was placed about 5 inches above the pump to maintain a slightly positive pressure for the system. C903 liquid chromatograph mini pump with motor, type NSI-33R (Waters Associates, Inc., Milford, Massachusetts) equipped with a 4000 Psi gauge pressure capable of a maximum flow rate of 2.7 ml per minute. An injector port model U6K equipped with a chart marker (Waters Associates, Inc.). A reversed-phase uBondapak Cl8 column with 3.9 (I.D.) x 300 mm dimensions (Waters Associates, Inc.). A dual beam U.V. monitor Model 100 with 10.0 mm pathlength flow cell, 1.00 mm cell diameter, and 8 microliters cell volume (Pharmacia Fine Chemicals, Piscataway, New Jersey). The maximum sensitivity of the monitor was 0.01 Absorbance Units Full Scale (AUFS). The measurements were obtained at 254 nm. A two-pen recOrder Model 281 (Linear Instruments Corporations, Costa Mesa, California). Stainless steel tubings, 0.009 inches I.D., and inches 0.D. were used to connect the pump, injection column, and the U.V. monitor. With these dimensions, a minimal dead volume was assured for the system. Teflon tubing was used to connect the solvent reservoir to the 61 pump. A 25.0 ul Pressure-Lok syringe (Precision Sampling Corp., Baton Rouge, Louisiana) with a needle of 0.02 in 0.D. x 2.0 in. was used to make the injections into the HPLC system. The mobile phase, Acetonitrile:water, was degassed before applying to the system. Preparation of Standard Curves: A mixture of standard methyl ketone-2,4-DNP hydrazone in acetonitrile was prepared for HPLC analysis. Derivatives of 2-Propanone, 2-Pentanone, 2-hepanone, 2-nonanone, 2-Undecanone and 2-tridecanone were prepared at a concentration of 5000 pg ketone per 50 ml acetoni- trile. The hydrazones mixture in acetonitrile was injected into the HPLC system at 3, 6, 10, and 15 microliter in- jection volume. The following HPLC conditions were used: Temperature: ambient U.V. monitor: 254 nm; 0.08-0.32 (AUFS) sensitivity range Flow rate: 1.43 ml/min Mobile phase: Acetonitrile-water (75:25 v/v) Column: uBondapak C18 (3.9 x 300 mm) Peak heights, chart units, were calculated on the base of 0.08 (AUFS) monitor sensitivity and were drawn vs concentration. The full-scale chart was 100 units which were equal to 10 inches. The slopes of standard curves were used in quantitation of methyl ketones. Duplicate analyses were performed for preparation of calibration graphs. 62 Recovery of Methyl Ketones: The percent recovery of individual methyl ketones was determined as described in part by Anderson (1966), Blakely (1970), and this study. A mixture of accurately weighed standard methyl ketones: C C5, C7, C9, C and 3' 11 C13 were dissolved in 5 m1 hexane. The mixture was passed over a 2,4-DNPH reaction column (Schwartz and Parks, 1961) to convert these ketones to their hydrazone derivatives. The concentrations used were approximately those found in Blue cheese. After the formation of hydrazones, 3.0 grams of fresh butter oil were added to the mixture to eliminate any hydrazone formation from fat. The remainder of the analysis was as described earlier for the cheese samples. The eluate from the alumina column was evaporated and the hydrazones were dissolved in acetonitrile and injected into HPLC. Duplicate analyses were performed for recovery determination. Two injections were made for each recovered sample. Percent recoveries were calculated as micromoles methyl ketone recovered divided by micromoles methyl ketone initially added. Quantitation of Individual Ketones in Cheese Samples: The monocarbonyls separated on weak alumina column were dissolved in small amount of pure acetonitrile and quantitatively transferred to volumetric flasks and made to mark with solvent. The injections into the HPLC system were 63 made at the same chromatographic conditions as for standard methyl ketones. The concentrations were in the range for calibration curves. Each sample was clarified before injection with a sample clarification kit (Waters Asso- ciates, Inc.) equipped with micr0pore solvent filters, 0.45 um pore size. The retention times of methyl ketones were Compared to their corresponding standard methyl ketone to identify each ketone. The peak heights were used in calculations of individual methyl ketone concentration (APPENDIX B). Duplicate analyses were performed in all samples, except those:noted by asterisk (*) in results and discussion section. Mass Spectral Analysis of 2-Undecanone-DNPH- Derivative Preparation of the Sample: A small quantity about 10 mg of pure, 99.0% 2- undecanone (C ) was dissolved in 10 milliliters carbonyl 11 free hexane and the mixture was passed through a 2,4-DNPH reaction column (Schwartz and Parks, 1961). After for- mation of 2,4 DNP-hydrazone of C methyl ketone, the 11 solvent was removed and the derivative was dissolved in acetonitrile and the mixture was examined by injecting a 5 ul aliquot into the HPLC. Two fractions were present in the mixture. The mixture of 2 fractions in acetonitrile was dried in rotary evaporator and then dissolved in 64 1,2-dichloroethane. This mixture was separated and col- lected individually by using a column of Celite 545- adsorptive magnesia (Schwartz gt gt., 1960; and Anderson, 1966). Fifteen grams adsorptive magnesia, heated to 400°C for 48 hours, and 30 grams of celite 545, heated to 150°C for 24 hours were slurried in 1,2-dichloroethane and packed in a 25 x 500 mm column equipped with a coarse fritted glass disk, using 3 - 5 Psi nitrogen pressure. The hydrazone derivatives of C methyl ketone were carefully 11 applied to the column. The column was develOped with 1,2- dichloroethane and the two separated fractions were col- lected in two flasks. The solvent was removed from each fraction and dissolved in acetonitrile. An aliquot of each fraction was injected into the HPLC and examined for single peak to confirm the presence of only one compound in each fraction. Finally each fraction was analyzed by mass spectrometry. Mass Spectrometer Analysis: The mass spectral analysis was accomplished in the Department of Biochemistry at Michigan State University. The measurements were done on a Varian MATCHS/DF which was connected to a PDP ll/or from Digital Equipment Corporation. The output routine was done with a PDP 11/40 interfaced to a Tektronic hard COpy 4610 and a Tektronic scope 4010-1. The mass spectrometer parameters for both samples were: Electron Energy, 70 ev; Source Temperature, 190°C; direct 65 probe temperature linear heating scan from 20°C - 270°C best sample temperature, 210°C. Separation of Methyl Ketones from Other Classes of Mono-Carbonyls in Blue Cheese A sample of monocarbonyl from QR Blue cheese was prepared and dissolved in 1,2-dichloroethane. The adsorp- tive magnesia-celite 545 column (Schwartz gt gt., 1960; and Anderson, 1966) was used to separate methyl ketones from other classes of monocarbonyls such as aldehydes, if present in the cheese. The column was developed with 1,2- dichloroethane and separation was followed visually since different classes formed different colors on the adsorbent. The first eluted fraction with a gray color belonged to methyl ketones. This fraction was separated. The solvent was evaporated and methyl ketone hydrazones were dissolved in acetonitrile and analyzed in the HPLC unit. RESULTS AND DISCUSSIONS Liquid Chromatographic Determination of Methyl Ketones as their 2,4-Dinitrophenyl- hydrazone (2,4-DNPH) Derivatives Extraction and Fractionation of Carbopyl Compounds in Cheese Fat Generally, carbonyl compounds play a major role in the flavor of oxidized fats and oils. The majority of these constituents, over 95%, are non-volatile when the usual methods of distillation are used for their deter- mination (Lillard and Day, 1961; and Lea and Swoboda, 1962). Therefore, extraction procedures with carbonyl-free solvents (Schwartz and Parks, 1961) were used to isolate the monocarbonyls responsible for Blue cheese flavor from cheese fat. Schwartz, Haller and Keeney (1963) extracted the carbonyl compounds of fats and oils, cheese fat and whole milk powder. The monocarbonyl fraction was separated into saturated aldehydes, methyl ketones, 2-enals and 2,4-dienals. In this experiment the homogeneous mixture of cheese sample and dried celite were extracted with carbonyl-free hexane to obtain a near theoretical yield of fat. A quanti- tative yield of fat was not considered critical (Schwartz and Parks, 1963). The hexane-fat mixture was applied to 66 67 a 2,4-DNPH reaction column to convert all of the compounds with carbonyl groups to their 2,4-DNP-hydrazones. The deri- vatives prepared in such a way were reported to be acid free (Schwartz, Haller and Keeney, 1963), thereby elimi- nating the need for extraction or other manipulations. Also, extremely low concentrations of carbonyls can be derivatized into hydrazones in this column. On the other hand the same authors (1963) demonstrated a linear rela- tionship between fat concentration in hexane and the yield of 2,4-DNP-hydrazones and reported that the reaction of car- bonyls with 2,4-DNPH in this column was quantitative over a wide range of fat concentration. The hexane effluent from the 2,4-DNPH reaction column contains all of the ori- ginal lipids and 2,4-DNP-hydrazone derivatives of carbonyl compounds which are soluble in the fat-hexane solution. Remaining on the column are those derivatives which are insoluble in the fat-hexane solution. The fat-hydrazone mixture in hexane was applied to a Celite 545-adsorptive magnesia column and the fat was eluted from the hydrazones by the addition of a sequence of solvents: hexane, hexane-benzene and benzene. Schwartz, Haller and Keeney (1963) used a ratio of 1:1 (w/w) Celite 545 and SeaSorb (adsorptive-magnesia) and reported the use Of 1.0 gram SeaSorb 43 per 5 to 10 uM hydrazone. However, Anderson (1966) and Blakely (1970) used a ratio of 2:1, 28.0 grams Celite 545 and 14.0 grams SeaSorb 43 and obtained 68 better results. The same ratio was used in this experi- ment, but because of overloading the column with hydrazones during extraction procedures of several samples, the amount of the adsorbent was increased to 36:18 grams. The adsorbed hydrazones on this column appeared as a deep brown color. After fat removal and complete elution of the hydrazones with chloroform-nitromethane mixture from the column, several blue, grey, and violet bands remained adsorbed on the column. Schwartz, Haller and Keeney (1963) reported that these bands which were retained on magnesia belong to all dicarbonyl bis (2,4-DNP-hydrazones). Therefore, the eluate from this column contained all monocarbonyls and ketoglycerides. The small amount of fat which was probably present in the eluate was removed from the samples in later steps. The mixture of monocarbonyls and ketoglycerides in hexane was fractionated on a weak alumina column (Schwartz and Parks, 1961) to remove the ketoglycerides from the samples (Schwartz, Haller and Keeney, 1963). Several samples were re-chromatographed on this column to obtain a sharper fractionation and fractions of greater purity. High Pressure Liquid Chromatographic (HPLC) Separation of Methyl Ketones The Reliability of HPLC System: The typical HPLC separations of 2,4-DNP-hydrazone derivatives of standard methyl ketones, C3 - C13, are shown 69 in Figures 2 and 3. These derivatives, which were prepared according to Shriner, Fuson and Curtin (1964) and stored in the freezer, were dissolved in acetonitrile at the time of analysis and injected into the HPLC. The data in Figure 2 represent the Optimum separation conditions of standard methyl ketone derivatives: 2-propanone (3), 2-Pentanone ) (C5), 2-heptanone (C7), 2-nonanone (C9), 2-undecanone (C11 and 2-tridecanone. The time for separation was approxi- mately 33 minutes. This time was sufficient to resolve all of the components present in the monocarbonyl fraction of Blue cheese samples (see Figure 4). Therefore, the re- solving power of the HPLC system was determined by chroma- tographying the monocarbonyl fractions of Blue cheese samples (Figure 4) and comparing the separated peaks with standard peaks in Figure 2 which were obtained by injecting methyl ketone'hydrazones into the HPLC system. The chromatograms in Figure 3 demonstrate the rapid resolution by HPLC of six DNP-hydrazone derivatives of stan- dard methyl ketones. The analysis time was approximately 6.0 minutes if 100% acetonitrile was used as the mobile phase (Figure 3-A). The separation time was longer when a mixture of acetonitrile and water was used: 12.0 minutes for acetonitrile-water (90:10 v/v); and 26.0 minutes for a ratio of 80:20 v/v (see Figure 3-B and C). The resolutions of methyl ketones in Figure 3 were obtained faster than that in Figure 2. However, a poorer separation and overlapping of peaks occurred for ketones present in the monocarbonyl 70 s.OOOMs yo: man :OAHOHOm :H ooummonm oumz mw>wum>fluwo one .Amaov wsocmomowualm com .AHHUV moosmomccsum .Amuv woo: lococum .Ahuv mcocmumosnm .Amov mcocmusmmnm .Amov mcosmmoumlm .mOCOqu ahsuoe osmosmum mo mw>wum>flumo weanmno>nawcmnmouuAOAOIv.N mo :oflumummwm 0AA: HMOfimwe mohazidzfi «m cm or 1 — q —) d — Jx < pro .saex.ca mm.o "oooom shore HE\m: o.ooa .Ocoo “H: o.oa "OEOHO> sofluommcH mms< mm.o “moose abe>eoemcmm E: «mm .uouOE0uonm .>.D "HOuowumo pcwflne< "wusumummEmB ces\Hs mo.H “oboe 30am N 1>\> mmumse o :HOHMMDHCOCODOO< "omens mambo: EE ocm x m.m uwHU xmmmocom __ucEOHOU j d . d .N musmflm new! BSNOdSS8 83080038 Figure 3. 71 HPLC resolutions of 2,4-DNPH derivatives of methyl ketones, C - C 3 13, at different solvent ratio of acetonitrile:H 0. 2 HPLC conditions A) C) Column: u Bondapak C18; 3.9 x 300 mm. Mobile Phase: Pure acetonitrile Flow Rate: 1.08 ml/min Detector: U.V. photometer, 254 nm Sensitivity Range: 0.32 AUFS Injection Volume: 5 ul Concentration: 100 ug/ml per peak Chart Speed: 0.25 in./min Column: u Bondapak C18: 3.9 x 300 mm Mobile Phase: Acetonitrile:H20 (90:10 v/v) Flow Rate: 1.10 ml/min Detector: U.V. photometer, 254 nm Sensitivity Range: 0.32 AUFS Injection Volume: 5 ul Concentration: 100 ug/ml per peak Chart Speed: 0.25 in./min Column: u Bondapak C18, 3.9 x 300 mm Mobile Phase: Acetonitrile:H20 (80:20 v/v) Flow Rate: 1.20 ml/min Detector: U.V. photometer, 254 nm Sensitivity Range: 0.32 AUFS Injection Volume: 5 pl Concentration: 100 ug/ml per peak Chart Speed: 0.25 in./min. RECORDER RESPONSE RECORDER RESPONSE Figure 3. 72 c3 c5 c3 c, A hes 3 °9 9: C11 c9 8 c“ g '5 l “3:? l c." .3 3‘3 .1 the L1 . o ‘ a o l a L TIMEJMNUTES c:3 C c5 .6 c, % co C11 913 L l l J/\ o l a A 15 J 24 TIMEJMNUTES HPLC resolutions of 2,4-DNPH derivatives of standard methyl ketones, C3 - C1 , at different solvent ratio of Acetonitrile:H2 . A) 100:0; B) 90:10; and C) 80:20 (v/v). The derivatives were prepared in solution and not "aged." 73 .menEsuo mmmmno wan Hmwoumaeoo mo noduomum ahconumoocoa :fl ucmmmum .mHU mu .mwcoumx H>£umE mo mm>flum>wumo mmzolv.m mo newumnmmom Damn Hmofimae .v wusmflm . mmh:z_2 mg... «0 ca up a 0 q q q 1(1) - q q < d a J .llI/(\ll/(\lr up 0 u a :0 M. 8 0 q: u o w so a no 3 35.5 25 "oooom photo 8 mHmEmm HE mm Mom H1 m.m "OEOHO> :OwuowncH 50 m“ mn5< mm.o 1 mod ”wmcmm >ua>flufimcwm 6 Es «mm .uwumEODOLQ .>.D "Menomuwa MW 3 ucwflnE¢ “wusumuwmswfi :- cflE\HE mv.H "wumm 30Hh 1>\> mmumse omrhosahuesosoo< "moose assoc: 0 EE com x ¢.m u U Rommocom 2 "casaou 0 ma 74 fraction of Blue cheese samples when faster analysis such as conditions in Figure 3 were used. The 33.0 minute sepa- ration Of methyl ketones (Figure 2) was selected as optimal. Snyder and Kirkland (1974) showed that pure water exhibited the greatest polarity among solvents and if mixed with other polar solvents such as acetonitrile, an intermediate polar- ity would result. Therefore, water was used to increase the polarity of the solvent. Since a reversed-phase column (u Bondapak C18) was used in this study, the less polar solvent resulted in a faster elution of components from the column, particularly more polar compounds such as C3 and C5 methyl ketones, and some peak overlap and poorer resolution were obtained (Figure 3-A). Peak height rather than peak area were used in this study for peak measurements. Peak height was measured as a distance from the baseline to the peak maximum. Snyder and Kirkland (1974) pointed out that the peak height method is almost always used for trace analysis. Peak height measurements are more accurate than peak area measurements because of lesser interference by neighboring and over- lapping peaks. Snyder (1969) described the advantages of using peak heights rather than peak areas in measurements: a) relative insensitivity to small variations in solvent flow rate; b) insensitivity to marginal resolution: and c) easy calculations for quantitation. The disadvantage of sensitivity to changing sample retention volumes and band widths was reported by the same author. 75 Snyder and Kirkland (1974) state that "it is not necessary to work at the maximum absorption of a peak when using the U.V. photometric detector." The wavelength, 254 nm, is often satisfactory for compounds having UV absorption. The absorption spectra of several 2,4-DNPH derivatives of carbonyl compounds have been studied by Roberts and Green (1946). An absorption maximum of 360 nm was reported for the 2,4-DNPH derivative of acetone. The absorption maximum of 363 nm was presented (Day, 1965) for the whole class of 2,4-DNPH derivatives of methyl ketones in chloroform. In this study the absorbance of 2,4-DNPH derivatives of standard methyl ketones and ketones in mono- carbonyl fraction of Blue cheese samples were made at 254 nm. Papa and Turner (1972) and Selim (1977) used 254 nm and 336 nm respectively for the determination of several standard 2,4-DNPH derivatives of carbonyl compounds in their HPLC systems. These authors obtained excellent analyses of components, even at trace levels. The typical retention times of separated methyl ketone derivatives by HPLC are presented in Table 3. These values were determined from double injection of samples into the HPLC system. The retention time of each peak was measured from the time of sample injection to the time the band maximum left the column (Snyder and Kirkland, 1974). Examples of typical chromatograms which were used in determining the retention times are shown in Figure 2 for 76 DNP-hydrazone of standard methyl ketones, and in Figure 4 for the hydrazones separated from the monocarbonyl fraction of Blue cheese. The 33 minutes HPLC analysis time was ended for each injection after the last component, Cl3 methyl ketone derivative, was eluted from the column. No ketone, nor any carbonyl compound was detected in the samples after C ketone was eluted from the column. 13 Table 3.--The retention time values for 2,4-DNP-hydrazones of standard methyl ketones and ketones present in monocarbonyl fraction of Blue cheese samples. Retention time, minutes* methyl ketone chain length standards Blue cheese C3 3.60 3.55 C5 4.90 4.86 C7 6.95 7.00 C9 11.00 10.96 Cll 18.60 18.57 C13 32.72 32.65 *Average values The retention times obtained from standard methyl ketones were used to identify individual ketones in the Blue cheese samples (compare the values in Table 3). The slight difference in the retention times of the ketone of identical chain length in the standard mixture or monocarbonyl frac- tion may be explained as the variation in ambient 77 temperature of purity of samples. This small variation in retention times did not affect peak heights and therefore the quantitation of individual methyl ketones. Detection of "Pre-Peaks" related to individual methyl ketones: During HPLC analysis, it was observed that 2,4-DNP- hydrazone derivatives of methyl ketones such as C7, C9, C and C ll 13 before the major component was eluted from the column, and demonstrated "Pre-peaks" which appeared just were detected at longer retention time than 7.0 minutes (see Figures 5 and 6). Comparing these Figures with Figure 2, it was found that this phenomenon happened if the stan- dard methyl ketones were "aged." A mixture of standard methyl ketones, C3 - C13, was passed through the DNPH- reaction column and after formation of hydrazones the same procedures were followed as did for the Blue cheese sample. The HPLC separation of the last eluate, derivatives mixture in acetonitrile, is shown in Figure 6. The pre-peaks of C C C and C methyl ketone derivatives were detected 7' 9' ll 13 after 6.5, 10.5, 17.0 and 30.0 minutes respectively. How- ever, the retention times of the major peaks were comparable with the corresponding values in Table 3. Both major and related pre-peak of individual standard methyl ketones were similar to those obtained from Blue cheese samples (compare Figures 4 and 6). In another trial a mixture of 2,4-DNPH derivatives of standard methyl ketones, C3 - C13, was dissolved in 5 ml 78 .sfisaoo mCHEsHo can .mwmmcmmelwuflaoo m cmsounu mcwmmmm an :oomm: ocm coHusHom cw venomoum OHO3 mw>wum>wuoo one .mau 1 mo .mwcoumx Hwnume osmocmum mo mo>wum>fluoo mAZOnv.~ mo cowumummwm qum $523.3: an em or m o q 4 1 J a - q u q A 90 < m. pro : Mm. .o .o .. ces\mm.o "oomom porno meEmm HE om\: OH "OESHONV COwuowncH m0 meoc mm.o . oH.o "mocha suhsauamcmm E: vmm .umumEouonm .>.D "HOuomumO ucmflnE< "ousumuwmeoe :flE\HE m¢.H ”mama 30am 1>\> mm mes omruoeahoasoooo< "omens oesoo: EE com x o.m umHU xomwocom : ":EOHOU .m musmflm 3SNOdS38 83080038 79 .QESHOO pageaao can .mHmosmoEIOuflHoo m canons» wmmmmm cwnu can CEOHOO cofluommu rezonv.~ M GA omnmmmum mum3 mm>wum>wumo one .m U 1 mo .mmcouox Hanuwa unaccoum mo mm>fium>flumo mmzo v.m mo sowumummmm Dam: .m musmfim $523.3: No ea or o o «1 u d q q d 4 A J l- 9 o a 3‘ o '0 U 0. a .551: 3.0 "oooom 398 co m 3 :ofiuomum MW HE mm you as ea "OEOHO> cofluOOnsH uu mesa mm.o u ea.o ”mmoom sue>euamcom nu a E: vmm .umumfiouonm .>.D "Houomumo ua ucmwnfid "musumquEmB ua CHI—{HE mvé ”oumm 30am NO a N Es 2an N o muwafluuflsoumo< “ommnm mambo: m M" 0 EE com on m.m “ U xmmmocom : “GEOHOU S ma 5: so 80 acetonitrile and the same above procedures, excluding the DNPH-reaction column, were applied to "age" the standards. The resulting HPLC separation is presented in Figure 5. The same pre-peaks were formed, but at slower rates. The smaller pre-peaks in Figure 5 (as compared to Figure 6) demonstrate the possibility of a significant conversion of original methyl ketone derivatives to their corresponding prepeaks components during derivatization in the DNPH- reaction column. In a further examination of pre-peaks, the HPLC mobile phase concentration was changed to acetonitrile: water (70:30 v/v), and a 60 cm u Bondapak C18 column was used instead of 30 cm. The pre-peak for C5 methyl ketone derivative was also detected as well as longer chain methyl ketones; however, the total HPLC analysis time was con- siderably longer; more than 60.0 minutes. The retention time of C methyl ketone was over 60 minutes and hardly l3 eluted from the column. The same results were obtained when the mobile phase concentration was changed to 65:35 (v/v) acetonitrile-water and the column was reduced in length to 30 cm. In this case the retention time for the first eluted peak, C3 methyl ketone DNP-hydrazone, was 4.5 minutes and for Cl3 methyl ketone was over 60 minutes. These observations revealed the possibility of pre-peak detection for C methyl ketone DNP-hydrazone, if the 3 polarity of the mobile phase was changed to a lower con- centration of acetonitrile and higher concentration of 81 water compared to 65:35 (v/v) acetonitrile-water. In con- trast, when 100% acetonitrile was used as the mobile phase the pre-peaks were combined with their corresponding major peaks as shown in the HPLC chromatogram in Figure 3-A. This suggested similar characteristics between a pre-peak and its corresponding major peak of methyl ketones. HPLC and mass spectrometric analysis of 2,4-DNPH derivative of C11 methyl ketone (2-undecanone) Pre-peaks were further analyzed by preparing an individual methyl ketone DNPH-derivative and examining by both HPLC unit and then mass spectrometry. The derivative was obtained by applying 2-undecanone (99+% pure) to the DNPH-reaction column. The C methyl ketone-DNP-hydrazone 11 was then injected into the HPLC (Figure 7 - injectl). Two peaks, C11 and its corresponding Pre-peak (Cll*) obtained which were qualitatively similar to the peaks for C11 methyl ketone in Figures 4, 5, and 6, shown with arrows. The slight difference in the retention time of peaks in Figure 7 with the other chromatograms was due to the ' difference in flow rates. This may also be related to a higher concentration of the samples in Figure 7 (Snyder and Kirkland, 1974). The mixture of C11 + Cll* was carefully fractionated on a Celite-adsorptive magnesia column which was develOped by dichloroethane as described in the analy- tical procedures. The appearance of the two bands frac- tionated on the column was very similar. The color of the 82 Column: 11 Bondapak C18: 3.9 x 300 mm C 11 Mobile Phase: Acetonitrile:water n (75:25 v/v) Flow Rate: 1.37 ml/min (:11 Temperature: Ambient Detector: U.V. photometer, 254 nm Ill . . . (D Sens1t1v1ty Range: 0.16 AUFS :: Ch - 25 ° ' C) art Speed. 0. 1n./m1n a Injectl: 5 111/5 ml fraction ‘ U) C |.l.l Inject2: 3 111/5 m1 fraction 11 ‘3 l Inject3: 1.5 ul/S m1 fraction I NJ 0 r m C11 0 .. O 3 N “1.9;. *5 O- _ 0 ,5 o i J 3? l 1 L- 1 l J o I; l 8 L n J 1 0 l6 8 TIME,MINUTES Figure 7. Injection 1: HPLC separation of 2,4-DNPH derivatives of standard 2-undecanone (C11). The derivative was prepared in 2,4-DNPH reaction column. Injection 2 and 3: HPLC detection of Pre-peak (C11*) and original (C11) peaks. C11* or C11 were prepared by fractionating the mixture (C11 + C11*) through a celite- magnesia column. 83 2 bands was reddish brown. The first fraction eluted from this column was examined by HPLC and the chromatogram Cll (Figure 7 - inject3) was obtained. This was identical to C in the mixture (compare inject and inject in Figure 11 l 3 7). When the second fraction from Celite-adsorptive mag- nesia column (C *) was separated and injected into the 11 HPLC unit, a faster elution was observed (Figure 7 - injectz). This opposite elution pattern for the 2 peaks, C11 and Cll*’ in conventional or HPLC fractionation demon- strated the difference between a normal and a reversed- phase u Bondapak C18 column. The two separated components, Cll or C11*, were also analyzed on the mass spectrometer and the results are shown in Figure 8. Fragment ions of m/e above 350 were of very low intensity. The basepeak of 100% intensity in both spectra were shown to be at mass 41.0. According to Klei- pool and Heins (1964) and Hawrysh (1970) the appearance of ion fragments at mass 178 and 238 are indicative of a methyl ketone 2,4-DNPH derivative. Kleipool and Heins (1964) studied the mass spectrometric identification of several DNPH derivatives of carbonyl compounds and con- cluded that aldehydes give strong peaks at mass 224, methyl ketones at mass 238, and ethyl ketones at mass 252. Both spectra in Figure 8 were similar and no peaks were observed at m/e 224 or 252. However, peaks are present at mass 178 and 238. A comparison of intensity of C11 and C11* for peaks higher than 20% of basepeak is shown in Table 4. 84 100.0 T .4 2"; 50.0 a 4;.) -a ‘8 m 0 ‘5 .3 100.0 8 . .H " ' 2 4.) .‘E’. a) 50.0 '1 m o - 1L1. . W 75.16%? 248 2 e 2 0 2 e 30 318 3 a See m/e (C11) 100.0 33 .1 5‘ r .H 30.0 - U) c a) a: 2 .H 0 J f j ' m .3 100.0 9 e e 19 La La 19 18 13 H J 3 w m 50.0 J 0 9:31:41 7 m/e (C11*) Figure 8. Mass Spectrometry Patterns of 2,4-DNPH derivative of standard 2-undecanone. The derivatives were obtained by treating the methyl ketone with the DNPH-reaction column. Top scan: C11, M.W. = 350. Bottom scan, C11*, M.W. = 350. 85 Table 4.--Intensity comparison of C11 and C11* for peaks higher than 20% of basepeak. % intensity of basepeak, (m/e 41.0), A m/e C11* C11 ATa 41.0 100 100 0 55.0 72 74 2 83.0 49 57 8 97.0 80 87 7 111.0 20 26 6 178.0 23 32 9 207.0 16 20 4 220.0 11 18 4 238.0 58 50 8 315.0 24 23 1 350.0 75 77 2 aDifferences in intensities of C11 and C11* at different fragment ions. Since the instrumentation error, AT, was less than 10%, the identity of the 2 spectra can be confirmed. The total molecular weight of either DNPH—C or DNPH-C * was pre- 11 ll sented to be at mass 350 (Figure 8) which is comparable with the theoretical molecular wt. of 2-undecanone-DNPH- derivative, 350.44 (see Appendix A). From the above findings, it can be presumed that pre-peak C *-DNPH derivative may be a geometric cis or 11 86 trans isomer of 2-undecanone-DNPH derivative. Similar to this, the pre-peaks shown in other methyl ketones such as C7, C9, and C (Figure 6) may exhibit the same molecular 13 weights as their major peaks and thus cis or trans isomers. Titlar (1960) studied the isomerization of the 2,4-DNPH derivatives of acetophenone (phenyl methyl ketone) and prepared the two cis and trans isomers of this compound. This author reported that the rate of isomerization of such compounds in aqueous acidic solutions was eight times greater than their bromination rate. Therefore, it is likely that during reaction of methyl ketones, C3 - C13, with 2,4-dinitrophenylhydrazine in the reaction column and presence of phosphoric acid, geometric isomerization would take place and pre-peaks result which are detectable by HPLC (see Figure 6). This did not happen when derivization was performed in solution by the method of Shriner, Fuson and Curtin (1965), and the mixture of derivatives in aceto- nitrile was injected without "aging" (see Figure 2). How- ever, if this mixture was "aged," slight formation of pre- peaks were demonstrated (see Figure 5). The same deri- vatives did not exhibit any pre-peaks as long as they were in the crystalline form and stored in the freezer. HPLC separation of methyl ketones from other classes of monocarbonyls To ascertain that the pre-peaks, described in the previous section, were not of aldehyde or any other classes 87 of carbonyl compounds, methyl ketones were separated from other classes of monocarbonyls in QR Blue cheese (Schwartz and Parks, 1961). The typical HPLC separation of these ketones is presented in Figure 9. Similar patterns of methyl ketones were observed and the retention time of each resolved peak was the same as was demonstrated for the separated ketones in the monocarbonyl fraction of Blue cheese or the standard ketones (compare Figure 9 with Figures 4 and 6). Several narrow and small aldehyde bands remained in the Celite-adsorptive magnesia column during fractionation of ketones from other classes of monocar- bonyls (Schwartz and Parks, 1961). Therefore it can be stated that classes of monocarbonyls such as saturated aldehydes, 2-enals, and 2,4-dienals, if present in QR Blue cheese are in small concentration and are retained on the columns during extraction of other monocarbonyl compounds from Blue cheese. Kinsella and Hwang (1977) suggested that very small quantities of alkanals, alkenals, and alkdienals occur in Blue cheese. Blakely (1970) used several long procedures in addition to extraction of monocarbonyls to separate the trace amounts of these aldehydes from ketones in Blue cheese. HPLC analysis of monocarbonyl fraction in this study was much faster and made possible the analysis of cheese samples every day as they were obtained from the curing room. Separation of several standard aliphatic and aro- matic carbonyl compounds derivatized with 88 .wmmmno pmcmmwu xoflsw cw ucmmmum .mahconumoocoe mo mommmao umnuo Eoum omumcofluomum mm>wum>flumplmmzo n mcouwx Hanumfi mo GOwumummmm qum Hmoflmhs .m musmflm 352.282.» Na" .QN 10w mu Au NH. _ . 14 . _ . a x. 4 90 3 .9 w .cfis\.cfl mm.o "owmam uumso m. cofluofiom HE m umm H: om "wander £08.00th 0 mh3< ma.c I mo.c "mmcmm muw>fluwmcww Ec vmm .uwquOuonm .>.D "neuomumo BSNOdSS8 83080038 ucmflnE< "wusumummamfi :flE\HE mv.H "mamm 30Hm m kn. A>\> mmnmev o znmfifluuecoumoa Homage mambo: EE com x m.m uxmnmpcom : ”CEDHOU 89 2,4-dinitrophenylhydrazine was also demonstrated by Papa and Turner (1972) and Selim (1977), utilizing HPLC systems. Selim (1977), at the same time with this study, used a reversed-phase u Bondapak C18 column and acetonitrilezwater as the mobile phase in a HPLC unit and separated the traces of low molecular weight aldehydes and ketones. The same author (1977) suggested that the reversed-phase liquid chromatography was more advantageous than adsorption chro- matography (Papa and Turner, 1972) in separating non-polar derivatives of carbonyl compounds. Quantitation of individual methyl ketones from C3 to C13 Preparation of standard curve: The linear-calibration graphs for six standard methyl ketone-2,4-DNPH derivatives are shown in Figure 10. These curves were presented as micromoles of hydrazones vs. the peak height units based on 0.08 AUFS detector sen- sitivity. The peak height units were expressed as the chart units. A full scale chart unit (100 divisions) were equal to 10.0 inches. The standard curves were obtained by injecting different volumes of known concentration of hydrazone mixture in pure acetonitrile into the HPLC. Duplicate analysis were performed and the $10pes of these curves were calculated (Table 5) utilizing a programmed calculator. The curves were drawn from 5 points including (0.0). Correlation coefficients and intercepts were 90 400 ’ C3 0 :5 4.) w-i C. 9 300 J J.) H (U 5 p. J: S? "U x 200 » it < - 07 “I O. 100 > 09 C311 C13 00 0.3 0.6 0.9 1.2 1.5 2,4 DNPH l MICROGRAMS I Figure 10. Calibration graphs, concentration vs peak height units, for 2,4-DNPH derivatives of methyl ketones, C3 - C13, based on a detector attenuation of 0.08 AUFS. Peak height: 100 chart units equal to 10 inches. 91 reported. The linearity of the standard curves were excellent in using the corresponding lepes in the quanti- tation of methyl ketones (Appendix B). The concentration of the methyl ketone in the samples was kept within the range found in the calibration curves. Table 5.--The values for slopes, correlation coefficient, and intercepts of standard curves of 2,4-DNP- hydrazone derivatives of C3 - Cl3 methyl ketones. Methyl ketone * Correlation chain length Slope coefficient Intercepts C3 254.02 0.9985 9.1282 C5 162.85 0.9993 4.4972 C7 97.22 0.9993 2.5481 C9 60.10 0.9993 1.8504 C11 33.07 0.9994 0.8939 Cl3 15.82 0.9998 0.2034 *Average of duplicates Recovery of methyl ketones, C3 - C13: The percent recoveries of standard methyl ketones, C3, C5, C7, C9, C11 and C13 carefully weighed mixture of standard methyl ketones (99+%) are presented in Table 6. A in carbonyl-free hexane was passed through a reaction column (Schwartz and Parks, 1961), and after formation of derivatives fresh butter oil was added to the mix to duplicate the extraction conditions when cheese was 92 analyzed. The fat added in this manner did not contribute any hydrazones to the samples. Recovered methyl ketones were analyzed as previously described. The high pressure liquid chromatogram obtained from recovered components was similar to the pattern in Figure 6. For the determination of peak heights of individual methyl ketones, the corre- sponding pre- and major peak of each methyl ketone were combined and percent recovery was calculated from the following equation: umoles methyl ketone recovered(l) (2)x100 umoles methyl ketone initially added % Recovery = and, (1) umoles methyl ketone recovered = Peak Height (chart units) injection volume (01) x Detector Attenuation (AUFS) 0.08 (AUFS) x x Dilution Factor (ml) x 1000 Slope of Std. curve (chart unit) x M.W. of hydrazone U9 (2) umoles methyl ketone initially added = pg methyl ketone initially added M.W. of corresponding methyl ketone The molecular weights of methyl ketones and their hydrazone derivatives are presented in Appendix A. It was shown (Table 6) that the percent recoveries of methyl ketones are decreased as the chain length of the ketones are lowered. Similar relationships between ketone 93 Table 6.--Percent recoveries of standard, C3 - C13, methyl ketones. Methyl ketone % Recovery chain length A B Averages* C3 77.97 77.01 77.49 C5 81.05 79.33 80.19 C7 90.71 88.57 89.64 C9 94.92 92.88 93.90 Cll 96.09 95.80 95.94 Cl3 98.50 99.03 98.76 *The average recoveries were used in quantitation procedures of methyl ketones in the cheese samples. chain length and recoveries were observed by Anderson (1966), Allen and Parks (1969), Arnold and Lindsay (1969), Blakely (1970). The recovery values of Anderson (1966) were quite similar to the values in Table 6. However, this author reported 64.9% recovery for acetone. This may be explained as a greater retention of acetone in several additional chromatographic steps that this author used. Blakely (1970) used similar procedures as Anderson and found a 63.8% recovery for acetone. However, 20-30% lower recoveries for longer chain methyl ketones were obtained by Blakely. 94 Methyl ketones in commercial imported or domestic Blue cheese The concentration of individual methyl ketones in Blue cheese samples were calculated (Appendix B) as micro- moles (uM) methyl ketone per 10 grams fat in cheese. Several investigators, Schwartz and Parks (1963), Anderson (1966), Blakely (1970), and Albert (1974) reported their results in the same manner. Table 7 represents the methyl ketone concentration in several commercial Blue cheese samples. It was demonstrated that 2-heptanone and 2-non- anone were the major ketones in the samples. However, in one brand of Blue cheese crumbles (domestic C1), domestic C and GorgonzolaI larger concentration of 2-propanone 2! (C3) was shown. The minimum ketone content was observed in GorgonzolaI with a total of 3.20 0M and the largest con- I centration was found in Blue cheese crumbles (domestic C2) with a total of 180.1 0M per 10 grams fat. The methyl ketone, C13, was not detected in the majority of Blue cheese samples. Significant variation of methyl ketone concen- trations which were found among individual ketones in the same brand or among different varieties of Blue cheese samples (Table 7), confirmed the results of Anderson (1966) and Blakely (1970). These authors reported a large vari- ation in the concentration of methyl ketones in Blue cheese samples. The analysis of Roquefort cheese for methyl Table 7.—-Methyl ketone concentration of domestic and 95 imported commercial Blue cheese. Blue cheese sample (a) 0M methyl ketones/10 g. fat C3 C5 C7 C9 C11 C13 Total Domestic A 2.83 11.13 19.73 6.53 1.93 - 42.15 Domestic B 2.85 4.70 11.08 5.20 1.70 - 25.53 Domestic C1 39.23 12.24 20.88 9.85 1.94 - 84.14 Domestic C2 59.69 24.24 34.07 40.01 15.08 6.99 180.1 Danish 4.49 7.84 13.25 14.17 3.53 0.85 44.13 Norwegian 0.38 1.31 20.71 34.12 7.52 0.77 64.91 GorgonzolaI 16.24 13.22 12.29 10.03 1.93 - 53.71 GorgonzolaI 0.28 0.47 1.81 0.64 Trace - 3.2 Stilton 1.54 0.83 1.89 0.93 0.21 - 5.4 RoquefortI 0.29 0.99 6.22 6.06 0.37 - 13.93 RoquefortII 0.43 0.21 6.06 7.59 0.67 - 15.01 (a) Domestics A and B: Different brands. Domestic C1 and C2: Blue cheese crumbles, same brand, purchased at different periods. Samples I and II: Purchased at different local markets. 96 ketones (see Table 7 and Figure ll-B) showed that C7 and C9 methyl ketones are the dominant and the most responsible ketone flavor in this cheese. Ketones other than C7 and C9 were found at concentrations less than 1.0 uM per 10 grams fat. Starkel (1924) initially demonstrated that C and C 7 9 methyl ketone were the major flavor components in Roquefort cheese. Many other investigators confirmed Starkel's findings. In Figure ll-A, the high pressure liquid chroma- togram of Wisconsin blue cheese demonstrates several other peak components, at approximately 3.0 minutes retention time which can be related to acetaldehyde. The first peak resolved after 2.5 minutes of injection in the same chroma- togram was free hydrazine present in this sample. The peaks which were detected between C3 and C5 or C5 and C7 ketones could be representative of 2-butanone (C4) and 2-hexanone (C ). The retention times of these peaks were at 3.8 and 6 5.5 minutes after the injection. The presence of these ketones in the cheese samples were confirmed by injecting a mixture of corresponding standard carbonyl-DNPH-deriva- tives (99+%) into the HPLC and comparing the retention times. The detection of trace even-numbered methyl ketones such as 2—butanone, 2-hexanone, and 2-octanone in Blue cheese has been reported by Bavisotto gt a1. (1960), Newar and Fagerson (1962), Dolézélek and Brabcova (1964), and Anderson (1966). The peak related to a DNPH-derivative of a carbonyl compound (Figure 11), eluted after 8.0 minutes, was present in all chromatograms. This compound could not Figure 11. 97 Typical HPLC separation of 2,4-DNPH-derivatives of methyl ketones, C3 - C11, in commercial: A) Wisconsin Blue; and B) French Roquefort cheeses. HPLC conditions A) Column: 0 Bondapak C18; 3.9 x 300 mm Mobile Phase: Acetonitrile:H20 (75:25 v/v) Flow Rate: 1.43 ml/min Temperature: Ambient Detector: U.V. photometer, 254 nm Sensitivity Range: 0.16 AUFS Injection Volume: 15 01 per 20 m1 sample Chart Speed: 0.25 in./min B) Column: 0 Bondapak C18, 3.9 x 300 mm Mobile Phase: Acetonitrile:H20 (75:25 v/v) Flow Rate: 1.43 ml/min Temperature: Ambient Detector: U.V. photometer, 254 nm Sensitivity Range: 0.16 AUFS Injection Volume: 20 01 per 25 m1 sample Chart Speed: 0.25 in./min Figure 11. 98 m 3 c c 7 2 5 A a: w m a: u: c3 :3 :r o .0 8 39. c9 1:: .5 i M M °" L l _. {a TIME,MINUTES O Q RECORDER RESPONSE L111 1 #6 TIME,MINUTES Typical HPLC separation of 2,4-DNPH derivatives of methyl ketones, C3 - C11, in commercial: A) Wisconsin blue; and B) French Roquefort cheeses. 99 be identified. The high pressure liquid chromatograms of Danish and Norwegian blue cheeses are shown in Figure 12. Both samples contained small concentrations of C methyl l3 ketone (see also Table 7). Like Roquefort (Figure 11), the Norwegian blue cheese exhibited a very low concen- tration of acetone, 0.28 uM per 10 grams fat (Table 7 and Figure 12): however, the concentration of C7 and C methyl 9 ketones were several fold greater in Roquefort cheese. Both Roquefort samples, obtained at different local mar- kets, showed the same patterns of methyl ketones, but the samples of GorgonzalI and (Table 7) resulted in a 15 II fold difference in concentration. This may be explained due to a different handling and storage conditions of such cheeses at the time of shipping. In this study, it seemed from the majority of high pressure liquid chromatograms such as those in Figure 4, 11, and 12 that the monocarbonyl fraction used for injec- tion contained methyl ketones as the largest proportion. Very similar patterns were observed between these chroma— tograms and the one in Figure 9. One may conclude that if monocarbonyls other than ketones were present, their con- centrations were so small that they were retained during the conventional column chromatographic procedures. Sensory evaluation of commercial Blue cheese The scores for sensory evaluation of several commercial cheese samples are presented in Table 8. Each "I Figure 12. 100 Typical HPLC separation of 2,4 DNPH deriva- tives of methyl ketones, C in: 3 ' C13' A) Danish; and B) Norwegian commercial Blue cheeses HPLC conditions for both A and B Column: 0 Bondapak C18; 3.9 x 300 mm Mobile Phase: Acetonitrile:H20 (75:25 v/v) Flow Rate: 1.43 ml/min Temperature: Ambient Detector: U.V. photometer, 254 nm Sensitivity Range: 0.08 - 0.16 AUFS Injection Volume: 15 pl per 20 ml sample Chart Speed: 0.25 in./min RECORDER RESPONSE o 1"; Inject L C L]J M “ °8 101 . 1 1 1 . 1 . 1j-k 8 16 24 32 TIME,MINUTES u: (D 2: 2 a) B u: a: C9 11: c, u: CD a: 8 111 8 a: E C11 c5 °3 K A/\ 1 is L 1 l 1 l 1 J 1 l O 8 16 24 32 TIME,MINUTES Figure 12. Typical HPLC separation of 2,4 DNPH deri- vatives of methyl ketones, C3 - C13, in A) Danish; and B) Norwegian commercial Blue cheeses. 102 Table 8.--Sensory evaluation of commercial imported and domestic Blue cheese. Average scores Cheese sample (a) Color (15) Flavor (20) Texture (10) Domestic A 13.0 17.5 8.5 Domestic Cl 12.0 18.5 8.5 Danish 13.5 18.0 9.0 Norwegian 13.5 18.5 8.5 GorgonzolaI 15.0 19.5 10.0 Stilton 11.0 15.0 9.0 RoquefortI 12.5 19.0 9.5 (a) Description of samples were the same as in Table 7. sample was evaluated for color, flavor, and texture by a panel of two experienced judges. The scores are the average values reported by judges and were based on a maximum score of 20 for flavor, 15 for color, and 10 for texture. The flavor was considered to be the most important quality fac- tor for incorporating the cheese samples into the other food products. Among the commercial cheeses, GorgonzolaI received the maximum score for color and texture, and a score of 19.5 for the flavor. The lowest score was given to Stilton for color and flavor. The very low methyl ketone concentration, particularly C7 and C9 ketones (Table 7) substantiate this poor flavor evaluation. An "ammonia" flavor was judged to be predominant in the Stilton cheese. 103 Kuehler (1974), Harte (1974), and Albert (1974) used the same method of sensory evaluation and reported a cheese was judged unacceptable if a total score of 30 was given. Based on this scoring all of the commercial samples in Table 8 were acceptable. Methyl ketones in quick ripened (QR) cheese incorporated with white mutant of P. roqueforti Utilizing a white mutant of P. roqueforti, two batches of cheese were prepared using this mold and salted on days 1, 6, 9 or 6, 7, and 8 during 15 days quick ripening. The white mutant of P. roqueforti was developed by Knight gt 31. (1950) by the ultraviolet irradiation of the blue mold spores. The purpose of using a white mutant in this study was to produce a white cheese with Blue cheese flavor. Kuehler (1974), Harte (1974), and Albert (1974) used the same white mutant of P. roqueforti to pro- duce quick ripened cheese. They obtained a satisfactory white cheese with Blue cheese flavor. Tables 9 and 10 represent the methyl ketone forma- tion during the initial ripening of cheese in the curing room, maintained at 52°F (11°C) and over 95% RH. The every other day analysis of this cheese showed that during 15 days ripening of cheese only small concentrations of methyl ketones were formed. On the 15th day, a total of 10.18 uM per 10 grams fat was produced in the cheese salted on days 1, 6, and 9 (Table 9) and a total of 7.84 uM was presented 104 Table 9.--Methy1 ketone concentration of quick ripened cheese6 using white mutant of P. roqueforti, at 52 i 1 F (11 C) and over 95% RH, salted on days 1, 6, and 9 during 15 days in the curing room. UM methyl ketone/10 g fat Time (daYS) c3 c5 c7 c9 €11 cl3 Total 0 0.82 Tr.(a) Tr. Tr. - - 0.82 1 0.26 0.31 1.93 0.91 Tr. - 3.41 3 2.34 1.21 5.15 5.64 0.57 - 14.91 5 1.08 4.51 0.80 3.03 0.64 - 10.06 7 1.43 1.47 0.60 4.80 1.96 0.21 10.47 9 0.92 0.82 0.40 4.44 1.45 Tr. 8.03 11 0.66 0.66 0.70 5.43 1.61 Tr. 9.06 13 0.56 1.62 0.65 5.43 1.84 Tr. 10.10 15 0.50 1.33 0.75 5.64 1.96 Tr. 10.18 (a) Tr. = Trace 105 Table 10.--Methyl ketone concentration of quick ripened cheese, using white mutant of P. roqueforti, at 52 i 10F (110C) and over 95% RH, salted on days 6, 7, and 8 during 15 days in the curing room. Time 0M methyl ketones/10 g fat (days) C3 C5 C7 C9 C11 C13 Total 0 0.83 Tr.(a) Tr. Tr. - - 0.82 1 0.26 0.31 1.93 0.91 Tr. - 3.41 3 1.87 1.02 5.72 7.15 0.68 - 16.44 5 0.79 14.98 0.80 2.67 0.66 - 19.86 7 1.25 1.27 0.57 4.59 1.31 Tr. 8.99 9 1.09 1.62 0.67 5.40 1.98 0.31 11.07 11 0.87 0.99 0.67 4.86 1.98 0.41 9.78 13 0.52 1.41 0.62 5.74 2.30 ,0.62 11.21 15 0.47 0.71 0.28 4.62 1.76 Tr. 7.84 (a) Tr. = Trace 106 in the sample salted on days 6, 7, and 8 (Table 10). The major ketone in both lots (Tables 9 and 10) was 2-nonanone. However, on the fifth day of ripening, C methyl ketone was 5 present in greatest concentration, particular in the cheese which had not yet been salted (Table 10). The concentra- tion of C5 methyl ketone decreased after the fifth day of ripening. The initial analysis of the curd revealed a very low concentration, 0.82 uM/lO g fat, of 2-propanone and C trace amounts of C and C9 methyl ketone. No C and 5' 7’ 11 C13 ketone were detected. The Cl3 methyl ketone was not detected until the 7th day of ripening. A large variation in the concentration of total methyl ketones was observed and almost similar in both lots during the initial ripening period (Tables 9 and 10). On the 11th and 15th days of ripening period the cheese samples were packed in beakers and covered with plastic wraps and transferred from the curing room to a refrigerator at 40°F (4.4OC) for a period of two weeks. The weekly analysis of methyl ketones are shown in Table 11. Single analysis were performed because of solvent deficiency at the time of analysis. The major and dominant ketone during this storage period was C9 methyl ketone. The total ketone concentration of 11th day cheese was increased from 9.06 to 19.95 UM/lo g fat during two weeks refrigeration when salted on days 1, 6, and 9 of initial ripening (compare Tables 9 and 11). The same increase in concentration was from 9.78 to 17.49 when the white cheese 107 Table ll.--Methy1 ketone concentration of the 11th and 15th days quick ripened cheese, usigg white mutant of P. roqueforti, at 52 i 1 F (11 C) and over 95% RH, salted on days 1, g, 9 oro6, 7, 8, plus two weeks stored at 40 t 1 F (4.4 C). 0M methyl ketone/10 g fat* Methyl ketone 11th day chain length lst week 2nd week 1,6,9 6,7,8 1,6,9 6,7,8 C3 2.31 1.30 3.17 2.02 CS 1.02 0.35 2.72 0.87 C7 2.84 2.04 3.97 2.28 C9 4 50 4.06 8.05 8.25 C11 1.44 2.23 2.04 4.07 Cl3 Tr. Tr. Tr. Tr. Total 12.11 9.98 19.95 17.49 15th day C3 1.26 0.29 1.30 0.73 C5 0.95 0.78 0.44 1.45 C7 2.23 1.86 2.04 2 91 C9 5.91 5.72 8.63 7.85 C11 2 77 3.30 3.34 4 08 C13 Tr. Tr. Tr. Tr. Total 13.12 11.95 15.75 17.02 *Single analysis 108 was salted on days 6, 7, and 9 (compare Tables 10 and 11). Likewise, the concentration of methyl ketones of the cheese of 15th day of ripening increased during refrigeration. The typical HPLC separation of methyl ketones in the white cheese is shown in Figure 13. The slight decrease in the retention times compared to standard ketone (see Figure 2 and Table 3) was due to an increase in flow rate of mobile phase from 1.43 to 1.48 ml/min. The other chromatographic conditions were the same as standards. However, the quan- titation of ketones was carried out at the standard separation conditions. The total methyl ketone concentration at the end of second week storage (Table 11) was lower than the values reported by Albert (1974) for the QR cheese with white mutant of P. roqueforti. This author reported that the white mutant of the mold produced a desirable QR cheese, although low methyl ketone concentration was obtained com- pared with commercial Blue cheeses. Nelson (1970) obtained ketone concentrations 7 to 15 times in a submerged culture of white mutant of P. roqueforti grown in homogenized milk plus lipolyzed milk fat. Sensory evaluation of quick ripened white cheese The average scores which were given by the judges for the color, flavor, and texture of this type of cheese are presented in Table 12. The scores for color and flavor were low, representing a poor quality cheese. However, the 109 .HuH0mmmvou .9 mo ucmuoe muflsz suds pmumuomuoocfl .mmmmno me cw .mHU I mu .mwcoumx Hmnuma mo mm>flum>flum© mmzolv.m mo coflumummwm 04mm Hangmme .ma musmflm 33258.2: «0 cu 9. m m 4 _ q q a _ 1.1.. 1 ‘ .cas\.ca mm.o "ommdm pumro ) 1 :1 __1 new! —--1- ° mHmEmm HE om umm an om "mEsHo> :ofiuomncH U a 8 mo 0 9.34 3.0 "magma 383855 a H E: vmm .kumeouonm .>.D ”Houomuma h ucmflg ”musumummfime 0 fl .cwE\HE mwé "oumm 30am 00 s t<> 363 m0 .0 m . AV 0 mumawuuwcoumo< "wmmnm mambo: . n" E: com x a.m .. U xmmmccom : "55500 8 ma 5: 110 total score for the three quality factors were acceptable. Both batches of cheese samples, 1, 6, 9 and 6, 7, 8 days salted, turned to a yellowish color on the 5th day of ripening at 52°F (see Figure 16) and this color remained in the cheese throughout the initial ripening or refriger- ation periods. Kuehler (1974), Harte (1974), and Albert (1974) obtained higher scores for color of the same cheese. However, they had incorporated a decolorizing agent in the milk at the time of rennet coagulation to produce a whiter color. In this experiment no decolorizer was used during cheese processing, and therefore, the develOpment of the yellow color was probably more evident in the cheese. The QR cheese samples were criticized for possessing a mild Blue cheese flavor, even after a brief storage at 40°F. Table 12.--Sensory evaluation of QR cheese, gsing white mutant of P. roqueforti at 52 i 1 F (11 C) and over 95% RH, salted on days 1, 6, 9 or 6, 7, 8, during lSOdays ripening plus two weeks storage at 40 t 1 F (4.4 C). Average scores Cheese sample(a) Color (15) Flavor (20) Texture (10) 1,6,9 6,7,8 1,6,9 6,7,8 1,6,9 6,7,8 A 9.0 9.0 13.0 12.0 B . . 15.0 14.5 9.0 9.0 C 9.0 9.0 14.5 14.0 8.5 (a) A: 15th day ripened cheese H- 1 F C: 15th day ripened cheese, plus 2 weeks at 40 i 1 F B: 11th day ripened cheese, plus 2 weeks at 40 111 Methyl ketones in Blue cheese quick ripened at 52°F (11°C) The development of methyl ketones in 2 batches of Blue cheese, salted on days 1, 9, 10 or 7, 8, and 9 is presented in Tables 13 and 14. During 15 days in the curing room small concentration of ketones was resulted. Like white cheese (see Tables 9 and 10) the total ketones in Blue cheese samples (Tables 13 and 14) exhibited a maximum concentration between the 3rd and 5th days of ripening. A maximum C9 methyl ketone development of 14.26 uM per 10 g fat (Table 13) was demonstrated in the cheese salted on days 1, 9, and 10 on the 5th day of ripening. This value for 7, 8, and 9 days salted cheese (Table 14) was 12.62 0M per 10 g fat. The rapid increase in C methyl ketone concentration 5 occurred on the 7th day of ripening. The ketone concen- trations decreased rapidly after 24-48 hrs following the noted ripening days. The decrease in methyl ketone concen- tration may be explained by the ready reduction of ketones to secondary alcohols (Jackson and Hussong, 1958; Kinsella and Hwang, 1977). This may happen, particularly, when the mold is still in the lag phase (Fan gt gt., 1976; and Fan and Kinsella, 1976) of its activity and thus not enough ketone has been produced from free fatty acids. Comparing Tables 13 and 14, it was shown that the sequence of salting days had none or very slight effect on methyl ketone pro- duction during which the cheese samples were in the curing room. The significant effect of salt treatment of methyl 112 Table 13.--Methyl ketone concentration of Blue cheese quick ripened at 52 i 10F and over 95% RH, salted on days 1, 9, 10, during 15 days in the curing room. uM methyl ketones/10 g fat Time (daYS) c3 c5 c7 C9 C11 C13 Total 0 0.75 Tr. Tr. Tr. — - 0.75 1 1.68 0 21 1.90 2.23 - - 6.02 3 1.20 0.19 3 11 6.88 Tr. - 11.38 5 2.50 0.90 2.74 14.26 Tr. - 20.40 7 1.09 6.31 0 21 2.66 Tr. - 10.27 9 0.60 6.15 0.16 0.52 0.19 - 7.62 11 0.54 0.94 0.31 1.47 0.49 Tr. 3.75 13 0.45 0.53 0.33 1.91 0.49 Tr. 3.71 15 0.29 0.29 0.33 2.31 0.73 Tr. 3.95 113 Table 14.--Methy1 ketone concentration of Blue cheese quick ripened at 52 i 10F and over 95% RH, salted on days 7, 8, 9 during 15 days in the curing room. 0M methyl ketones/10 g fat Time (days) C3 C5 C7 C9 C11 C13 Total 0 0.75 Tr. Tr. Tr. - - 0.75 1 1.68 0.21 1.90 2.23 - - 6.02 3 0.74 0.14 2.97 5.96 Tr. - 9.77 5 2.21 2.39 2.60 12.62 Tr. — 19.82 7 0.90 6.65 0.15 1 87 Tr — 9 57 9 0.38 2.04 0 30 0 88 0 ll — 3 71 11 0.59 0.84 0.39 1.90 0.41 - 4.13 13 0.32 0.45 0.43 3.08 0.83 0.26 5.37 15 0.29 1.08 0.31 2.89 0.99 0.39 5.95 114 ketone accumulation was observed during storage of the same cheese at lower temperatures, particularly 480F (Tables 15 and 16). Tables 15 and 16 represent the methyl ketone con- centration of the 11th or 15th day ripened cheese during two weeks storage at lower temperatures, 400F (4.4OC) or 480F (8.9OC), than the ripening temperature. It seemed from Table 16 that higher temperature of storage, 480F, favored methyl ketone production in the cheese, salted on days 7, 8, and 9. This batch of cheese showed a rapid increase in methyl ketones, particularly, C3, C5, C7 and C9 ketones during the first week of storage. The degree of methyl ketone formation in cheese salted on days 1, 9, 10 was very slow at this temperature of storage. However, the same cheese stored at 40°F produced more ketones than cheese salted on days 7, 8, and 9 (see Table 15). The 15th day cheese, salted on days 7, 8, 9 exhibited more ketones than the same cheese salted on days 1, 9, and 10. Total ketone concentration of (7, 8, 9) cheese at 48oF after 2 weeks storage was 149.1 uM per 10 g fat (Table 16). This value for the same cheese at 400F was 35.89 uM per 10 g fat (Table 15). Comparing the 11th and 15th day QR cheese stored two weeks at 40°F in the same table, similar values for total ketone concentration was obtained for 7, 8, and 9 days salted cheese. However, the 11th day cheese, salted on days 1, 9, 10 exhibited a higher concentration than the 115 Table 15.--Methy1 ketone concentration of the 11th 3nd 15th days Blue cheese quick ripened at 52 i 1 F and over 95% RH, salted on days 1, 9, 10 or 7, 8, and 9 plus two weeks stored at 40 i 10F. UM methyl ketones/10 g fat* methyl ketone 11th day chain length lst week 2nd week 1,9,10 7,8,9 1,9,10 7,8,9 C3 6.13 0.56 6.21 10.81 C5 3.53 1.20 7.44 7.88 C7 4.35 0.95 16.49 12.34 C9 4.58 4.67 6.59 3.80 C11 0.97 0.83 0.71 1.06 C13 0.19 Tr - - Total 19.75 8.21 37.44 35.89 15th day C3 1.99 0.97 3.07 5.32 C5 2.21 1.44 3.37 9.29 C7 4.26 2.75 4.95 13.71 C9 2.77 2.77 3.30 5.61 C11 0.66 0.66 0.79 0.71 C13 Tr. - Tr. Tr. Total 11.89 8.59 15.48 34.64 *Single analysis. 116 Table 16.--Methy1 ketone concentration of the 11th day Blue cheese quick ripened at 52 t 10F and over 95% RH, salted on days 1, 9, 10 or 7, 8, and 9, plus two weeks stored at 48 i 10F. uM methyl ketones/10 g fat* methyl ketone lst week 2nd week chain length 1,9,10 7,8,9 1,9,10 7,8,9 C3 0.68 41.91 0.98 37.63 C5 1.50 36.08 3.33 40.11 C7 1.48 44.29 6.69 47.62 C9 4.25 23.69 4.18 23.08 C11 2.35 1.56 2.04 0.68 C13 0.33 Tr. - - Total 10.59 147.5 17.22 149.1 *Single analysis. 15th day cheese when stored for two weeks at 40°F. The 15th day cheese samples stored for two weeks at 48°F were not analyzed, because they developed off-flavors during the first week of storage at this temperature. Therefore, it seemed that 11 days ripening period for the cheese QR at 52°F was preferred to 15 days if a brief storage at 48°F was necessary to develop more ketones in the cheese. The values obtained for methyl ketone in QR Blue cheese stored at 48°F were comparable with findings of Albert (1974). However, more acetone was found in the cheese analyzed in this study (Tables 15 and 16). The 117 relatively low concentration of total methyl ketones in cheese stored at 40°F (Table 15) was similar to the data obtained by Blakely (1970) in 7 days QR Blue cheeses. Blakely and Albert reported that C methyl ketone was the 9 major ketone in their sample, however, despite of white cheese in this study (Tables 9, 10 and 11), almost all quick ripened Blue cheeses exhibited a dominant C methyl 7 ketone after the storage period. The dominancy of C7 methyl ketone in Blue cheese has been confirmed by several investigators such as Hammer and Bryant (1937), Schwartz and Parks (1963), and Anderson (1966). Dartey and Kinsella (1971) reported that C7 and C9 methyl ketones were the major ketones present in the cheese at all stages of ripening. These authors showed that generally, both C7 and C accounted for more than 50% of the total methyl ketones. 9 Figure 14 represents the typical methyl ketone separation in QR Blue cheese. The chromatogram showed a clean separ- ation of methyl ketones among other monocarbonyls which were probably in trace amounts. Sensory evaluation of Blue cheeseyguick ripened at 52 F The subjective analyses of several samples of Blue cheese quick ripened at 52°F are presented in Table 17. Generally, most of these samples were judged to have a "mild" and "quite clean" flavor. Such cheeses scored between 15 and 18 for flavor. These cheese samples con- tained a total ketone concentration of 35 to 45 0M per m 118 .mmmmno mDHm m0 :H .MHU 1 U .mmcouwx Hwnume wo mo>wum>wump mmzouv.~ mo coflumnmmmm 04mm Hmowmhe .va musmflm . mmhazzz 52.... «0 ea or m 0 fl 1 q d A a fi - d 1.1/\11 1 ()3) 3 a. = J i o o 8 z m. a w 3 mflv _ 1. AV cae\.ca m~.o ”600mm pumro nu h a 075mm HE m Mom T. om "mEsHo> :ofluommcH U H mma< «v.0 I mo.o "mocmm >uw>fluflmcmm ua E: vmm .umumEODOLQ .>.D "nouomumo a ucmflnEd "musumumewB ”W 52:5 2.; "warm 303 no m 3? mmumt 00063635800004 .0923 330: S 3 EE com x m.m “mam xmmmocom : ucEDHOU 119 Table l7.--Sensory egaluation of Blue cheese quick ripened at 52 i 1 F and over 95% RH, salted on days 1, 9, 10 or 7, 8, and 9 durin9015 days ripening, plus two weeks stored at 40 or 48°F. Average scores Cheese(a) Sample Color (15) Flavor (20) Texture (10) 1,9,10 7,8,9 1,9,10 7,8,9 1,9,10 7,8,9 A 11.0 14.0 12.0 12.0 7.0 8.0 B 15.0 13.0 13.5 15.0 8.0 9.0 C 12.0 13.0 17.5 15.0 8.5 8.5 D 13.0 13.5 16.5 19.5 9.0 9.0 E 11.5 13.5 18.0 16.5 8.5 9.0 F 12.5 14.0 16.0 18.0 8.5 9.0 G 13.5 12.5 18.0 16.0 9.0 8.5 H 13.5 13.0 18.0 18.5 9.0 9.0 (a) A: 11th day ripened cheese B: 15th day ripened cheese C: 11 days ripened plus one week stored at 40°F D: 11 days ripened plus one week stored at 48°F E: 11 days ripened plus two weeks stored at 40°F F: 11 days ripened plus two weeks stored at 48°F G: 15 days ripened plus one week stored at 40°F : 15 days ripened plus two weeks stored at 40°F 120 10 g fat (Table 15). The highest score was given to the 11th day ripened cheese, salted on days 7, 8, and 9, plus one week storage at 48°F (see Table 17, Sample D). This cheese exhibited an excellent ketone flavor. An intense blue-green color was demonstrated in cheese A, salted on days 1, 9, and 10. This cheese received the lowest score for color. The color of this cheese improved during two weeks storage at lower temperatures. The method of salting caused a difference in color of the cheese at the end of initial curing day, 11th or 15th day (see Table 17 and also Figure 16). Higher scores were given to the samples which were salted on days 7, 8, and 9. An increase in mold growth was noted by judges for those cheeses salted on days 1, 9, and 10. Several samples with high flavor scores such as cheese F in Table 17 were criticized for an atypi- cal aftertaste. Texture was improved after 11 days ripening and, therefore, higher scores were obtained. Methyl ketones in Blue cheese quick ripened at 620F (16.7UC) Table 18 presents the methyl ketones of QR Blue cheese during 7 days in the curing room. Since the rate of mold growth and methyl ketone formation at this ripening temperature seemed greater than other lots ripened at lower temperature, daily analysis of methyl ketones was performed to observe the changes in ketone concentration. Hedrick gt gt. (1968) reported that 7 days were sufficient to ripen a Blue cheese at 62°F using their method, which was similar 121 Table 18.--Methy1 ketone concentration of Blue cheese quick ripened at 62 i 10F and over 95% RH, salted on days 4, 5, and 6 during 7 days in the curing room. Time, 0M methyl ketone/10 g fat (daYS) c3 C5 c7 C9 C11 €13 Total 0 0.87 Tr. Tr. Tr. - - 0.87 l 1.31 0.43 3.14 1.58 0.12 - 6.58 2 0.96 0.29 3.15 3.49 0.10 - 7.99 3 1.31 5.05 1.67 6.65 0.26 - 14.95 4 1.13 8.37 0.31 0.51 0.39 - 10.71 5 0.67 4.11 0.26 0.64 0.26 - 5.94 6 0.38 0.91 0.45 2.25 0.99 0.21 5.19 7 0.81 1.46 1.33 5.67 2.30 0.29 11.85 to the method used to manufacture the cheese in this study. Blakely (1970) confirmed this ripening period and quanti- tated methyl ketones in a 7-days quick ripened Blue cheese. This author reported low methyl ketone concentration in The predominant ketone several batches of QR Blue cheese. in the sample was C methyl ketone. The total methyl 9 ketone concentration that this author reported in 3 samples of QR Blue cheese was 19.0 - 27.0 uM per 10 grams fat. However, Albert (1974) reported a greater concentration of methyl ketones in Blue cheese quick ripened by this method. Methyl ketone accumulation in Blue cheese during seven days ripening was neither uniform nor cumulative 122 (Table 18). This was also shown for other lots of cheese, quick ripened at 52°F (see Tables 9, 10, 13, and 14) during 15 days ripening. The largest concentration of total ketones which exhibited in Blue cheeses ripened at 52°F on the 5th day of ripening (Tables 13 and 14), occured on the 3rd day of ripening for this lot of cheese (Table 18). Like other lots of cheese, large concentrations of C5 and C9 methyl ketones were demonstrated during 7 days ripening at 62°F. The 7th day ripened cheese (Table 18) was packed and stored at different temperatures, 40°, 48°, or 62°F. Tables 19 and 20 represent the methyl ketone content of the cheese samples at those storage conditions. In all samples, C7 methyl ketone was dominant, however, C methyl 9 ketone was also considered to be the major ketone. Only, trace amount of C13 methyl ketone was found in the cheese ripened at 62°F. The large concentrations of methyl ketones, 117.3 0M per 10 g fat, on the third day of storage at 62°F (Table 19) started to decline afterwards. It was necessary to refrigerate this cheese sample on the 4th day of storage to lower temperatures for longer shelf life. The sudden increase in the concentration of methyl ketones from 11.85 uM/lO g fat on the 7th day of ripening (Table 18) to 65.57-117.3 0M per 10 g fat during storage at 62°F (Table 19) demonstrates that the enzymes respon— sible for methyl ketone production are not necessarily 123 dependent on mold growth. And, when they are produced in such levels by the mold, they may generate methyl ketones at a large concentration in a short period of storing Blue cheese at 62°F (compare Tables 18 and 19). Table 19.--Methy1 ketone concentration of the 7th day Blue cheese quick ripened at 62 i l Foand over 95% RH, plus 4 day storage at 62 i 1 F. uM methyl ketone/10 g fat Time (days) C3 C5 C7 C9 C11 C13 Total 1 7.37 15.23 20.78 18.68 3.51 Tr. 65.57 2 13.54 25.98 32.45 24.53 3.90 Tr. 98.40 3 13.00 24.97 37.52 35.55 6.24 Tr. 117.3 4 8.29 20.48 25.35 19.16 3.51 Tr. 76.79 The large accumulation of methyl ketones in quick ripened cheese, stored two weeks at lower temperatures, 40° or 48°F is also shown in Table 20. Cheese stored for two weeks at 48°F exhibited the maximum methyl ketone content, 184.8 uM per 10 g fat. This value was similar to the obtained for commercial Blue cheese crumbles, 180.1 0M per 10 g fat (see Table 7). Overall, C7 and C9 methyl ketones were predominant during the storage period. Figure 15 represents a profile for C7 and C9 ketone production. The development of these two important flavor components was revealed significantly during the brief storage after 7 days initial curing of the cheese (B, C, and D). 124 Table 20.--Methy1 ketone concentration of the 7th day Blue cheese quick ripened at 62 i 1 F and over 95% 0 RH, plus two weeks stored at 40 i l0 or 48 t 1 F. uM methyl ketones/10 g fat methyl :gtg:e lst week 2nd week length 400 48° 40° 48° C3 19.96 23.52 19.82 29.28 C5 29.04 32.78 29.35 45.91 C7 40.99 44.79 54.08 64.23 C9 38.92 44.32 37.73 39.52 C11 7.80 7.32 5.86 5.86 Cl3 Tr Tr. Tr. Tr. Total 136.7 152.7 146.8 184.8 Sensory evaluation of Blue cheese quick ripened at GZUF The scores for color, flavor, and texture of the cheese ripened at 62°F are listed in Table 21. The scores for color were low, particularly those samples which were stored at higher temperature, 62°F (see Samples A, B, C, D, and E in Table 21). The color was slightly improved during two weeks storage at lower temperatures, 400 or 48°F and higher scores were given. The color was judged to be brownish green and less pleasant than the color of samples ripened at 52°F (see Figure 16). The scores for flavor (Table 21) were high, especially for the cheese stored at lower temperatures (Samples F, G, H, and I in Table 21). An 125 .mcmm new .cmv .oov um cwuoum com home an cwcmawu xofisc mmwmno moam :fi TCOCMCOZIN pcm wcocmumwmlm mo mcoflumuucwocoo Hmoflmxe .ma musmflb o O m < u . a F c n a F A o o v o a p o a o W. o. N .o. \\ .ou .oo mama um oououm .mromz u o Mona um otuoum .mxows . o .ov 1V8 SWVHOOl/831OW080IW howm um omuoum .m>mo I m mm wmmA .om tom momu um pmcwmwu .mhmn 1 a a ol .8 N.0! 126 Table 21. --Sensory evaluation of Blue cheese quick ripened at 62 + 10 F and over 95% RHOduring 7 Odays plus 4- 15 days storage at 40 i 1 , 48 + 1°, or 62 + 10F. Cheese (a) Average scores Samples Color (15) Flavor (20) Texture (10) A 10.5 15.0 7,5 B 10.0 15.5 7,5 C 10.0 16.5 3.0 D 10.0 17.5 3,0 E 10.0 15.5 3,0 F 11.0 19.0 9,0 G 11.0 18.0 3.0 H 11.5 19.0 9,0 I 11.5 18.0 8,5 (a) A: 7th day Blue cheese B: C, D, E: cheese samples stored at 62°F for l, 2, 3, and 4 days respectively F, G: first week stored at 40° or 48°F respectively H, I: second week stored at 40° or 48°F respectively atypical aftertaste was noted for cheese E. The texture of the cheese was improved during the two weeks of storage, particularly at 40°F. Changes inypH During Quick Ripening of Cheese The pH of the cheese samples were measured in dupli- cate every day of quick ripening and during the storage .Huuomwsmou .M mo HTGSOQ UHOE moan no muwsz sufl3 pwumuomuoocfi wwwmmno vmcmmwn xowsv new meQEDHO ommmzo mDHm HmwoquEou .Eobesow . .?8 wwwuwutum 127 wNthm» ” m.w.~uu~uu=8m . . u 0 u 0 6 m . «0.05“ 2.1m wt ,w ~ .oH munmwh 128 period. In each analysis a homogeneous sample was used. Morris and Jezeski (1953) showed the effect of pH on the enzyme activity of g. roqueforti, and therefore, the impor- tance of pH during ripening of Blue cheese. The change in pH during ripening of 3 batches of cheese is presented in Figure 17. Almost similar patterns were shown for all batches. Blakely (1970), Kuehler (1974), and Harte (1974) showed the same pH changes during quick ripening of Blue cheese. The range of pH maxima was 6.70 — 7.00. Each batch showed a slightly different pH maximum: 6.75 for the cheese ripened at 62°F; 7.0 for white cheese QR at 52°F; and 6.70 for the Blue cheese QR at 52°. The maximum acidity in QR cheese was not obtained until appro- ximately 24.0 to 72.0 hrs after the cheeses were manu- factured. The pH values at this stage of ripening were 4.7 - 4.8. Both cheese samples QR at 52°F remained at this low pH for a period of 48.0 hrs, lst to 3rd days of ripening period. For the white cheese (C), and 2nd to 4th days of this period for the Blue cheese (A). Remaining at the minimum pH was during a shorter period, 24 hrs, for the cheese ripened at higher temperature (B). The initial pH in Sample A, 6.25, which was higher than other batches may be explained as an insufficient lactic acid production at the time of manufacturing by the streptococcus starter. However, this cheese showed the same pH pattern, decrease and increase in pH, at a longer period of time, approximately 24.0 hrs, than other batches. 129 .mon um pocmmwu wmowno Open: .AQS..:6V U can “homo um pmcmmflu wmwmso moam .Aolllllov o “momm um pmcmmfiu wmwwno moam .Arllltv d .Huu0mmsm0u .a mo ucmuse opens no msam nufl3 pmumasoosfl mwmmno mo maficmmwu RUHDU mafiuop mm CH mmocmnv .nH muoowm w>onm on» ma U can .m .< mo uasmmu may .mu0mmumne mmmmso m ca Eoum pmuw>ooou pow mEmum one mamum oa x Ame mooumx axouos :5 n umw @ oa\w:oumx Hanuwe z: "ADV A< Raocmddcv rezone.~ .e.z Auo>oomm x A v mmon aafieooa x lass pwum>oowu umm macho x m>flum>fluw© mo.o x Aanv .HO> pmuomncw souomm coausaao_ x coaumsomuoa muacs uroaor room u moonwx assume mo mazouv.~ om Aop mw3 :ofipmsqm m>on¢ m: was: uumso .3.2 x >um>oomm x A V m>uso ppm mo wmon OH x oooa x “Hay 005660 coausaao x mo.o x Aanv mEsHo> pwuomflcw poms c0wumscmuu¢ uouomuwo Amuflcs pumsov unmflw: xmmm umm omuomuuxm madam OH you wcoumx axnumfi Aznv mmHOEouofiz "CORumuucwocoo mcoumx axcuwfi Hoopfl>w©cw mo coHumHsono m xHoZmam< APPENDIX C APPENDIX C Reconstitution of dried cheese to original moisture content: Table C.--Moisture content of fresh and freeze dried Blue cheese. Sample(a) % m01sture Fresh Freeze dried A 49.00 4.50 B 46.34 3.75 C 44.10 3.81 (a) The same samples as in Table 24. Pearson square method was used to calculate the amount of water which was added to the dried cheese for rehydration. Example for cheese A. M.C. Water: 100.0 .5 For every 51 parts dried cheese, need ///’44 .0 \\\\ 44.5 parts water 51. 0 \\\\ 49 ///’ F. D. Cheese 4.5 Therefore, for 10 g F. D. Cheese: 44.5 _ .5170 x 10 - gilt grams H20 should be added to F.D. cheese to obtain 49% moisture cheese. Proof: 10 9 Freeze dried cheese + 8.72 9 H20 = 18.72 rehydrated cheese 10 9 Freeze dried cheese x 0.045 = 0.45 g H 0 in Freeze dried cheese 18.72 x 0.49 9.17 g H 0 in rehydrated cheese 2 9.17 - 0.45 8.72 g H 0 added a: 2 144 APPENDIX D APPEND IX D Table Dl.--pH values of QR cheese, using blue or white mold powder of P. roqueforti, cured at 520 or 62°F and over 95% RH, salted on 3 days schedule during 7 to 15 days ripening. (a) pH values Days Blue mold, 52°F Blue mold, 62°F White mold, 52°F 1,9,10 7,8,9 4,5,6 1,6,9 6,7,8 0 6.25 6.25 5.38 5.43 5.43 1 5.75 5.75 4.86 4.82 4.82 2 4.99 5.01 4.90 4.80 4.78 3 4.87 4.91 5.37 4.78 4.82 4 4.86 4.69 6.65 5.12 5.58 5 4.90 4.87 6.75 6.58 6.52 6 5.27 5.51 6.73 6.76 6.75 7 6.43 6.28 6.60 7.05 6.95 8 6.47 6.56 - 6.80 6.59 9 6.34 6.34 - 6.76 6.66 10 6.55 6.45 - 6.93 6.82 11 6.30 6.47 - 6.70 6.69 12 6.60 6.56 - 6.69 6.70 13 6.46 6.56 - 6.70 6.77 14 6.66 6.70 - 6.73 6.73 15 6.70 6.72 - 6.70 6.70 (a) Duplicate analysis 145 146 Appendix D Continued. Table D2.--pH values of commercial Blue cheese or QR cheese during two weeks storage at 40°, 48°, or 62°F. Sample pH Gorgonzola 5.34 Stilton 6.47 Roquefort 6.39 Domestic (crumbles) 5.94 QRl (4 days at 62°F) 6.01 QR2 (2 weeks at 40°F) 6.22 QR3 (2 weeks at 48°F) 6.20 QR4 (1 week at 48°F) 5.99 0R5 (1 week at 48°F) 6.47 QR6 (2 weeks at 40°F) 5.77 QR7 (2 weeks at 40°F) 5.64 QR8 (2 weeks at 40°F) 6.15 QR9 (2 weeks at 40°F) 6.00 QRlo (1 week at 40°F) 6.48 QRl' 2, 3: Blue cheese QR at 62°F and stored at 62°, 40°, and 48°F respectively. Blue cheese QR at 52°F for 11 days, salted on days 7, 8, 9; l, 9, 10; 7, 8, 9; or 1, 9, 10 respectively. QR4' 5' 6’ 7‘ QRB' 9: Blue cheese QR at 52°F for 15 days, salted on days 7, 8, 9 or 1, 9, 10 respectively. QRlo: QR cheese using white mutant of P. roqueforti salted on days 6, 7, 8. BIBLIOGRAPHY BIBLIOGRAPHY Albert, S. A. 1974. The effect of process parameters on the flavor and methyl ketone content of quick ripened Blue cheese. M.S. Thesis. Michigan State University. 63 pp. Albrecht, T. W. and H. O. Jaynes. 1955. Milk lipase. J. Dairy Sci. t§:137. Allen, C. and O. W. Parks. 1969. Method for determining contributing methyl ketones to flavors of sterile concentrated milks. J. Dairy Sci. 2g:1547. Anderson, D. F. 1966. Flavor chemistry of Blue cheese. Ph.D. Thesis. Oregon State University. 95 pp. Anderson, D. F. and E. A. Day. 1966. Quantitation, evalu- ation and effect of certain microorganisms on flavor components of Blue cheese. J. Agric. Food Chem. tt: 241. AOAC. 1975. "Official Methods of Analysis," 12th ed. Association of Official Analytical Chemists, Washington, D.C. Arnold, R. G. and R. C. Lindsay. 1969. Quantitative deter- mination of n-methyl ketones and o-aminoacetophenone in sterilized concentrated milk. J. Dairy Sci. gt: 1097. Babel, F. J. 1953. The role of fungi in cheese ripening. Economic Botany 1:27. Bakalor, S. 1962. Research related to the manufacture of Blue-veined cheese. Dairy Sci. Abstr. 33:529 and 583. Bassett, E. W. and W. J. Harper. 1958. Isolation and iden- tification of acidic and neutral carbonyl compounds in different varieties of cheese. J. Dairy Sci. gt:1206. 147 148 Bavisotto, V. 8., Rock, L. A. and R. S. Lesniewski. 1960. Gas chromatography of volatiles of fermented milk products. J. Dairy Sci. gt:849. Benassi, R. 1963. The composition of fat of ewes milk. Latte. t1:468. in Dairy Sci. Abstr. t§:3593, 1963. Bills, D. D. and E. A. Day. 1964. Determination of the major free fatty acids of cheddar cheese. J. Dairy Sci. 51:733. Blakely, L. E. 1970. A study of the formation of methyl ketones and free fatty acids in quick ripened Blue cheese and subsequent losses during dehydration. Ph.D. Thesis. Michigan State University. 180 pp. BohaE, V. 1957. Dried cheese. (In Czech, English trans- lation) Zpravy vyzkumneko Ustavu pro melko a vejce. Praha no. 314:134. Boudreau, A., Richardson, T. and C. H. Amundson. 1966. Spray-dried butter and loss of volatile fatty acids during spray drying. Food Technol. t9:668. Bullock, D. H., Hamilton, M. 0L and D. M. Irvine. 1963. Manufacture of spray dried Cheddar cheese. Food in Canada tt(3):26. Bradley, R. L. Jr. and C. M. Stine. 1962. Spray drying Cheddar cheese and Blue cheese. J. Dairy Sci. (Abstr.). g§:649. Bradley, R. L. Jr. and C. M. Stine. 1963. Spray drying of natural cheese. Mfd. Milk Prod. J. §t(ll):8. Bradley, R. L. Jr. and C. M. Stine. 1964. Foam spray drying of natural cheese. Mfd. Milk Prod. J. §§(6): Bradley, R. L. Jr. and C. M. Stine. 1968. A gas chromato- graphic study of the volatile flavors in natural and spray dried Cheddar cheese. J. Gas Chromatog. 6:344. Chandan, R. C. and K. M. Shahani. 1963a. Purification and identification of milk lipase. I. Purification. J. Dairy Sci. g§:275. Chandan, R. C. and K. M. Shahani. 1963b. Purification and characterization of milk lipase. II. Characteri- zation of the purified enzyme. J. Dairy Sci. it: 503. 149 Coffman, J. R., Smith, D. E. and J. S. Andrews. 1960. Analysis of volatile food flavors by gas-liquid chromatography. I. The volatile components from dry Blue cheese and dry Romano cheese. Food Res. 25:663. Coppock, P. D., Subramaniam, V. and T. K. Walker. 1928. The mechanism of the degradation of fatty acids by mold fungi. J. Chem. Soc. London. 1298:1422. Coulter, S. T. and W. B. Combs. 1939. The use of steapsin in the manufacture of Blue cheese. J. Dairy Sci. 22:521. Coulter, S. T., Combs, W. B. and J. C. George. 1938. The pH of Blue or American Roquefort cheese. J. Dairy Sci. tt:273. Csiszér, J., Tomka, G. and A. Rohmlehner-Bakos. 1956. Development of flavoring substances (acetoin + di- acetyl) in cheese and their influence on cheese quality. XIV. Int. Dairy Congr. 2:125. in Dairy Sci. Abstr. t2:257, 1957. Currie, J. N. 1914. Flavor of Roquefort cheese. J. Agri. Res. 2:1. Czulak, J. 1959. Bitter flavour in cheese. Austr. J. Dairy Technol. tt:l77. Dakin, H. D. 1908a. The oxidation of butyric acid by means of hydrogen peroxide with formation of acetone, al- dehydes and other products. J. Biol. Chem. 3:77. Dakin, H. D. 1908b. A synthesis of certain naturally occur- ing aliphatic ketones, with a suggestion of a poss- ible mode of formation of these substances in the organism. J. Biol. Chem. 1:221. Dartey, C. K. and J. E. Kinsella. 1971. Rate of formation of methyl ketones during Blue cheese ripening. J. Agri. Food Chem. t2:771. Dartey, C. K. and J. E. Kinsella. 1973a. Oxidation of sodium [U-14C] Palmitate into carbonyl compounds by Penicillium roqueforti spores. J. Agric. Food Chem. 21:721. Dartey, C. K. and J. E. Kinsella. 1973b. Metabolism of U-14C] lauric acid to methyl ketones by the spores of Penicillium roqueforti. J. Agric. Food Chem. tt: 933. 150 Day, E. A. 1965. Some properties of carbonyl compounds encountered in the flavor isolates of dairy pro- ducts, a review. Food Technol. t2:1585. Day, E. A. and D. F. Anderson. 1965. Gas chromatographic and mass spectral identification of natural compo- nents of the aroma fraction of Blue cheese. J. Agri. Food Chem. tt:2. Desai, J. Rohini. 1966. A study of some physical and chemical properties of dehydrated sour cream. M.S. Thesis. Michigan State University. 63 pp. Dolezélek, J. and J. Brabcové. 1964. Methyl ketones in mold ripened cheese during ripening. Sb. Vys. Sk. Chem. Technol. Praze Protravin, Technol. §:l97, in Dairy Sci. Abstr. t§:3058. 1966. Dozet, N. and M. Stanisic. 1973. Effects of milk homo- genization in manufacture of soft white cheese. Food Sci. Tech. Abstr. §:2P, 231. Dwivedi, B. K. and J. E. Kinsella. 1974a. Carbonyl pro- duction from lipolyzed milkfat by the continuous mycelial culture of Penicillium roqueforti, J. Food Sci. tgz93. Dwivedi, B. K. and J. E. Kinsella. 1974b. Continuous pro- duction of Blue-type cheese flavor by submerged fermentation of Penicillium roqueforti. J. Food Sci. 22:620. Eitenmiller, R. R., Vakil, J. R. and K. M. Shahani. 1970. Production and properties of a Penicillium roque- forti lipase. J. Food Sci. t§:l30. Fan, Y. S., Hwang, D. and J. E. Kinsella. 1976. Changes in biochemical components during germination of spores of g. roqueforti. J. Sci. Food Agric. tl:745. Fan, Y. S., Hwang, D. and J. E. Kinsella. 1976. Methyl ketone formation during germination of g. roque- forti. J. Agric. Food Chem. 23:443. Flosdorf, E. W. and H. W. Hamilton. 1957. Process for freeze drying soft cheese. U.S. Pat. 2,789,909. In Dairy Sci. Abstr. tg:813, 1957. Forney, F. W. and A. J. Markovetz. 1971. The biology of methyl ketones. J. Lipid Res. tt:383. 151 Foster. E. M., Nelson, F. E., Speck, M. L., Doetsch, R.N. and J. C. Olson. 1961. Dairy Microbiology. Prentice Hall, Inc., Englewood Cliffs, New Jersey. 492 pp. Fujishima, T., Negishi, T., and F. Fujino. 1971. Composi- tion of lipids and fatty acids in Blue cheese. Dairy Sci. Abstr. tt:4329. Funder, S. 1949. Physiological variation in proteolytic properties of mold, in Proc. 12th Int. Dairy Congr. Vol. 2, Ivar Haeggstrém, Stockholm, 463. Franke, W., Platzeck, A. and G. Eichhorn. 1961. Zur Kenntnis des Fettsavreabbaus durch Schimmelpilze, Arch. Mikrobiol. 52:73. Franke, W., Platzeck, A. and G. Eichhorn. 1962. Zur Kennt- nis des Fettsavreabbaus durch Schimmelpilze. Ver- suche Zum Weiteren Umsatz der methyl ketone. Arch. Mikrobiol. tt:154. Fryer, T. F. and B. Reiter. 1967. Lipolytic activity of lactic acid bacteria. J. Dairy Sci. g9:388. Gehrig, R. F. and S. G. Knight. 1958. Formation of ketones from fatty acids by spores of Penicillium roque- forti. Nature 182:1237. Gehrig, R. F. and S. G. Knight. 1961. Formation of 2-hep- tanone from caprylic acid by spores of various filamentous fungi. Nature 192:1185. Gehrig, R. F. and S. G. Knight. 1963. Fatty acid oxidation by spores of Penicillium roqueforti. Applied Microbiol. tt:166. Girolami, R. L. and S. G. Knight. 1955. Fatty acid oxi- dation by Penicillium roqueforti. Applied Micro- biol. t:264. Gould, I. A. 1941. Effect of certain factors upon lipoly- sis in homogenized raw milk and cream. J. Dairy Sci. tt:779. Gould, I. A. 1942. Influence of pH, type of fat, and pan- creatic extract upon lipolysis in homogenized milk. J. Dairy Sci. t§:869. Gripon, J. C. and J. Bergere. 1972. The proteolytic sys- tem of P. roqueforti cultural conditions and nature of the proteolytic system. Lait. gg:497. 152 Gripon, J. C. and B. Debest. 1976. Etude electrophore- tique du Systéme protéolytique exocellulaire de Penicillium roqueforti. Lait. §§:423. Gripon, J. C., Desmazeaud, M. J., LeBars, D. and J. L. Bergere. 1977. Role of proteolytic enzymes of Sttgptococcus lactis, Penicillium roqueforti, and Penicillium caseicolum during cheese ripening. J. Dairy Sci. 60:1532. Gripon, J. C. and J. Hermier. 1974. Le systeme proteoly- tique de Penicillium roqueforti. Biochimie. §§:1323. Groux, M. and M. Moinas. 1974. Le Flaveur des fromages. II. Etude Comparative de la fraction volatile neutre de divers fromages. Lait. gg:44. Haller, A. and A. Lassieur. 1910a. Etude des échappées du beurre de coco. composition de I' essence de coco. Compt. Rend. Acad. Sci. 150:1013. Haller, A. and A. Lassieur. 1910b. Sur deux alcools actifs et une troisieme cetone contenus dans I' essence de coco. Compt. Rend. Acad. Scit. 151: 697. Hammer, B. W. and H. W. Bryant. 1937. A flavor constituent of Blue cheese (Roquefort type). Iowa State College J. Sci. tt:281. Hanni, H. 1967. Free amino acids released during cheese ripening. Schweizerische Milchzeitung. 23:442. Harper, W. J. and E. W. Bassett. 1959. A survey of acidic and neutral carbonyl compounds in various cheese varieties. The Milk Products J. 14. Harper, W. J. and P. Gaffney. 1970. Lipase. I. charac- terization of the milk lipase system. Centennial Res. Ser. Dept. Dairy Tech. Ohio State University 6:1. . Harte, B. R. 1974. The effects of process parameters on the formation of volatile acids and free fatty acids in quick ripened Blue cheese. M.S. Thesis. Michigan State University. 103 pp. Harte, B. R. and C. M. Stine. 1977. Effects of process parameters on formation of volatile acids and free fatty acids in quick-ripened Blue cheese. J. Dairy Sci. g9:1266. 153 Hartley, C. B. and J. J. Jezeski. 1954. The microflora of Blue cheese slime. J. Dairy Sci. 22:436. Hawke, J. C. 1966. Reviews of the progress of dairy science. Section D. Dairy Chemistry. The forma- tion and metabolism of methyl ketones and related compounds. J. Dairy Res. 22:225. Hawrysh, Z. L. 1970. The retention of volatile mono- carbonyl compounds in natural model systems. Ph.D. Thesis. Michigan State University. 123 pp. Hedrick, T. I. 1968. Spray and freeze dried Blue cheese. Milk Dealer 22(10):10. Hedrick, T. I., Kondrup, E. and W. T. Williamson. 1968. Preparation of Blue cheese by adding to the cheese a Blue cheese mold and ripening the curd in a divided condition. U.S. Patent. 3,365,303. Herrington, B. L. and V. N. Krukovsky. 1939. Studies of lipase action. I. Lipase action in normal milk. J. Dairy Sci. 22:127. Hilditch, T. P. 1967. The chemical constitution of natural‘fats. John Wiley & Sons, Inc., New York, 664 pp. Hornstein, I. and P. F. Crowe. 1962. Carbonyl removal from non-oxygenated solvents. Anal. Chem. 21:1037. Hussong, R. V. and B. W. Hammer. 1935. The preparation of mold powder for Blue-veined cheeses. J. Dairy Sci. t2:599. Hwang, D. H., Lee, Y. J., and J. E. Kinsella. 1976. B-keto-acyl decarboxylase activity in spores and mycellium of 2. roqueforti. Int. J. Biochem. 7:165. Imamura, T. 1960. Studies on the changes of milk proteins by Penicillium roqueforti. J. Agric. Chem. Soc. 34:892. Imamura, T. and K. Kataoka. 1963a. Biochemical studies on the manufacturing of Roquefort type cheese. I. Lipase producing ability of Penicillium roque- forti. Jap. J. Zootech. Sci. 23;344. in Dairy Sci. Abstr. 22:1695, 1964. 154 Imamura, T. and K. Kataoka. 1963a. Biochemical studies on the manufacturing of Roquefort type cheese. II. Characteristics of lipases produced by Penicillium roqueforti. Jap. J. Zootech. Sci. 33:349. in Dairy Sci. Abstr. 23:1696, 1964. Imamura, T. and K. Kataoka. 1966. Biochemical studies on the manufacturing of Roquefort type cheese. III. Isolation of lipases from mold-culture. Jap. J. Dairy Sci. t3:Al38. in Dairy Sci. Abstr. 23:1525, 1967. Irvine, O. R., Bullock, D. H., Pearson, A. M. and W. H. Sproule. 1957. Developing manufacturing methods for cheese of reduced milk fat. Can. Dairy and Ice Cr. J. 33(3):64. Ismail, A. and K. Hansen. 1972. Accumulation of free amino acids during cheese ripening of some Danish cheese. Milchwissenschaft 22:556. Jackson, H. W. and R. V. Hussong. 1958. Secondary alcohols in Blue cheese and their relation to methyl ketones. J. Dairy Sci. 3t:920. Jasperson, H. and R. Jones. 1947. Some unsaponifiable constituents of deodorization distillate of vege- table oils. J. Soc. Chem. Ind. 33:13. Jensen, R. G. 1964. Lipolysis. J. Dairy Sci. 32:210. Jolly, R. C. and F. V. Kosikowski. 1973. Flavor and chemi- cal changes in Blue cheese by microbial lipases. J. Dairy Sci. 33:624. Jolly, R. C. and F. V. Kosikowski. 1975. Flavor develop- ment in pasteurized milk Blue cheese by animal and microbial lipase preparation. J. Dairy Sci. 33:846. Jokay, L. and R. I. Meyer. 1959. Application of lyophili- zation to cheese products. II. The development of freeze-dehydrated creamed Cottage cheese. Abstr. J. Dairy Sci. 32:908. Kaminski, E., Stawicki, S. and E. Wasowicz. 1974. Vola- tile flavor compounds produced by molds Asper- gillus, Penicillium. Appl. Mikrobiol. 21:1001. Kelly, P. L. 1944. The determination of lipase from milk extracted with acetone and ether. J. Dairy Sci. 27:675. 155 Kelly, P. L. 1945. Milk lipase activity. A method for its determination and its relation to the estrual cycle. J. Dairy Sci. 33:803. Khan, I. H., Chandan, R. C., Dill, C. W. and K. M. Shahani. 1966. Hydrolysis of milk fat by microbial lipases. Abstr. J. Dairy Sci. 33:700. Kikuchi, T. and S. Takafugi. 1971. Proteases of Camembert cheese molds. J. Jap. Zootech. Sci. 33:205, 268. Kinsella, J. E. and D. Hwang. 1976. Biosynthesis of flavors by Penicillium roqueforti. Biotech. and Bioengineering. 33:927. Kinsella, J. E. and D. H. Hwang. 1977. Enzymes of Peni- cillium rogueforti involved in the biosynthesis of cheese flavor. Crit. Rev. Food Sci. Nutr. 3:191. Kleipool, R. J. C. and Heins, J. Th. 1964. Mass spectro- metric identification of organic derivatives. Nature 203:1280. Knight, 8. 1963. Process preparing flavoring compositions. U.S. Patent 3,100,153. Knight, S. G., Mohr, W. H. and W. C. Frazier. 1950. White mutants of Penicillium roqueforti. J. Dairy Sci. 33:929. Knoop, A. M. and K. H. Peters. 1971. Submikroscopiche structurver-anderungen im camambert-kase wahrend der reifung. Milchwissenschaft. 33:193. Kondrup, E. and T. I. Hedrick. 1963. Short time curing of Blue cheese. Michigan State Agri. Expt. Sta. Quart. Bull. 33:156. Kosikowski, F. V. 1977. Cheese and Fermented Milk Foods. 2nd Ed. Edwards Brothers. Inc. Ann Arbor, Michigan. p. 374. Kosikowski, F. V. and A. C. Dahlberg. 1954. Quantitative appraisal of the free amino acids in foreign type cheese. J. Dairy Sci. 31:167. Kosikowski, F. V. and G. Mocquot. 1958. Advances in cheese technology. FAO Agri. Studies No. 38. Columbia University Press, New York. 236 pp. 156 Kuehler, C. A. 1974. The effect of process parameters on the flavor and protein degradation in quick ripened Blue cheese. Ph.D. Thesis. Michigan State Univer- sity. 154 pp. Kuzdzal-Savoie, S. and W. Kuzdzal. 1963. Fatty acids in the milk of different animal species. Lait. 33:369. in Dairy Sci. Abstr. 33:3587, 1963. Lane, C. B. and B. W. Hammer. 1937. The homogenization of milk for Blue cheese. J. Dairy Sci. 33:468. Lane, C. B. and B. W. Hammer. 1938. Bacteriology of cheese. III. Some factors affecting the ripening of Blue (Roquefort type) cheese. Iowa Agri. Expt. Sta. Res. Bull. 333:197. Langer, J. E. and E. A. Day. 1964. Development and flavor properties of methyl ketones in milk fat. J. Dairy Sci. 33:1291. Lawrence, R. C. 1965. Activation of spores of Penicillium roqueforti. Nature. 208:801. Lawrence, R. C. 1966. The oxidation of fatty acids by spores of Penicillium roqueforti. J. Gen. Micro- biol. 33:393. Lawrence, R. C. and R. W. Bailey. 1969. Evidence for the role of citric acid cycle in the activation of spores of Penicillium roqueforti, Biochim. Biophys. Acta. 333:77. Lawrence, R. C. and J. C. Hawke. 1968. The oxidation of fatty acids by mycellium of Penicillium roqueforti. J. Gen. Mikrobiol. 33:289. Lea, C. H. and P. A. T. Swoboda. 1962. Simple vacuum distillation procedure for determination of the volatile carbonyl content of autoxidized edible fats. J. Sci. Food Agric. 33:148. Lillard, D. A. and E. A. Day. 1961. Autoxidation of milk lipids. II. The relationship of sensory to chemical methods for measuring the oxidized flavor of milk fats. J. Dairy Sci. 33:623. Litchfield, J. H. 1969. Factors affecting the production of microbial food flavors by submerged culture methods. Dev. Ind. Microbiol. 33:341. 157 Luksas, A. J. 1973. Preparation of Blue cheese flavored product. U.S. Patent 3,720,520. Lundstedt, E. and F. Yau Yee Lo. 1973. Preparation of Blue cheese from soybean milk. U.S. Pat. 3,743,515. Meyer, R. I. and L. Jokay. 1959. Application of lyophili- zation to cheese products. I. Some observations on the characteristics and stability on the freeze- dehydrated Cheddar, Brick, Muenster, Blue, Cream and Cottage cheese. Abstr. J. Dairy Sci. 33:908. Mocquot, G. and M. Béjambes. 1960. Technology of ewes' milk and goats' milk products. Dairy Sci. Abstr. 22:1. Modler, H. W., Brunner, J. R. and C. M. Stine. 1974a. Extracellular protease of Penicillium roqueforti. I. Production and characteristics of crude enzyme preparation. J. Dairy Sci. 33:523. Modler, H. W., Burnner, J. R. and C. M. Stine. 1974b. II. Characterization of a purified enzyme pre- paration. J. Dairy Sci. 33:530. Moinas, M., Groux, M. J. and I. Hormann. 1974. Aroma composition for imparting a Camembert or Blue cheese flavor to foods. Swiss Patent, 552,949. Morgan, M. E. and E. 0. Anderson. 1956. The neutral carbonyl compounds in Blue-mold type cheese. J. Dairy Sci. 33:253. Morris, H. A. 1969. The manufacture of Blue and Gorgon- zola cheeses. Paper presented 6th Ann. Marschall International Italian Cheese Seminar. Madison, Wisconsin. Morris, H. A., Combs, W. B. and S. T. Coulter. 1951. The relation of surface growth to the ripening of Minnesota Blue cheese. J. Dairy Sci. 33:209. Morris, H. A., Jezeski, J. J., Combs, W. B. and S. Kura- moto. 1963. Free fatty acid, tyrosine, and pH changes during ripening of Blue cheese made from variously treated milks. J. Dairy Sci. 33:1. Morris, H. A., Jezeski, J. J., Combs, W. B., and S. Kura- moto. 1955. Free fatty acids produced during Blue cheese ripening. Abstr. J. Dairy Sci. 33:590. 158 Morris, H. A. and J. J. Jezeski. 1953. The action of microorganisms on fats. II. Some characteristics of the lipase system of Penicillium roqueforti. J. Dairy Sci. 33:1285. Morris, T. A. 1964. The manufacture of Blue-veined cheese in Queensland. Austr. J. Dairy Technol. 33:9. Morris, T. A. 1969. Effect of ripening temperature on the properties of Blue vein cheese. Australian J. Dairy Tech. 33:69. Nawar, W. W. and I. S. Fagerson. 1962. Direct chromato- graphic analysis as an objective method of flavor measurement. Food Technol. 33:107. Nelson, J. H. 1970. Production of Blue cheese flavor via submerged fermentation by Penicillium roqueforti. J. Agric. Food Chem. 33:567. Ney, K. H. 1974. Contribution of amino acids and peptides toward food flavor, in Contribution of Chemistry to Food Supplies. Morton and Rhodes, D., Eds., Butterworths, London, p. 411. Ney, K. H. and I. P. Wirotama. 1972. Investigation of the aroma of Edelpilzkase, a German Blue mold cheese. Z. Lebensm. Unters. Forsch. 149:275. Ney, K. H., Wirotama, I. P. and W. G. Freytag. 1975. Blue cheese flavor. British Pat. 1,381,737. Nickerson, T. A., Coulter, S. T. and R. Jenness. 1952. Some prOperties of freeze-dried milk. J. Dairy Sci. 35:77. Niki, T., Yoshioka, Y. and K. Ahiko. 1966. Proteolytic and lipolytic activities of Penicillium rogueforti iso- lated from Blue cheese. XVII. Int. Dairy Congr. Nilson, R. and S. Willart. 1960. The influence of homo- genization on the fat splitting in milk. Rep. Milk and Dairy Res. Alnarp, Sweden 60. Nilson, R. and S. Willart. 1961. The heat inactivation of the fat splitting in milk. Rep. Milk and Dairy Res. Alnarp. Sweden 64, 24. Ogden, R. P. 1967. Method of freeze-drying liquid milk products. U.S. Pat. 3,297,455. 159 Olsansky, C. and M. Schmidt. 1960. Study on the manu- facture of dried natural cheese. Prumysl Potravin 33:407. in Dairy Sci. Abstr. 22:3373. 1960. Papa, L. J. and L. P. Turner. 1972. Chromatographic determination of carbonyl compounds as their 2,4 dinitrOphenyl hydrazones. II. High pressure liquid chromatography. J. Chromatographic Sci. 33:747. Parks, 0. W., Keeney, M., Katz, I. and D. P. Schwartz. 1964. Isolation and characterization of the methyl ketone precursor in butterfat. J. Lipid Res. 3:232. Parmelee, C. E. and F. E. Nelson. 1947. Special cultures for manufacture of Blue cheese from pasteurized milk. J. Dairy Sci. 33:524. Parmelee, C. E. and F. E. Nelson. 1949. The use of mold enzyme preparation in the manufacture of Blue cheese from pasteurized homogenized milk. J. Dairy Sci. 33:993. Patton, S. 1950. The methyl ketones of Blue cheese and their relations to its flavor. J. Dairy Sci. 33: 680. Pelissier, J. P., Mercier, J. C. and B. Ribadeau-Dumas. 1974. Problem of bitter flavor in cheese. Rev. Lait. Fr. 325:820. Peters, I. I. and F. E. Nelson. 1948. The influence of Mycotorula lipolytica lipase upon the ripening of Blue cheese made from pasteurized homogenized milk. J. Dairy Sci. 33:611. Peters, I. I. and F. E. Nelson. 1961a. Manufacture of Blue cheese. Milk Prod. J. 33(1):10. Peters, I. I. and F. E. Nelson. 1961b. Quality of Blue cheese as influenced by methods of ripening. Milk Prod. J. 33(7):8. Peterson, M. H., Johnson, M. J., and W. V. Price. 1943. Determination of milk lipase. J. Dairy Sci. 33: 233. Pork§, J., Dolezalek, J. and Z. Pech. 1959. The influence of different strains of Penicillium roqueforti upon the ripening of cheese of the Roquefort type. XV. Int. Dairy Congr. 3:742. 160 Poznafiski, S., Jaworski, J., and T. D'Obyrn. 1967. Char- acterization of products of fat hydrolysis in the ripening of Roquefort type cheese in relation to NaCl content. Dairy Sci. Abstr. 33:1666. Pulay, G. 1959. Determination of the optimum drying temperature of mould powder, in particular Roque- fort mould. 15th Int. Dairy Congr. London. 3:736. Reineccius, G. A. and S. T. Coulter. 1969. Flavor reten- tion during drying. J. Dairy Sci. 33:1219. Roahen, D. C. and H. H. Sommer. 1940. Lipolytic activity in milk and cream. J. Dairy Sci. 33:831. Roberts, J. D. and S. Green. 1946. Absorption spectra of some 2,4 dinitrophenylhydrazones. J. Am. Chem. Soc. 33:214. Sadini, V. 1963. Fat of ewes' milk and of Italian ewes' milk cheese. Latte. 33:933. in Dairy Sci. Abstr. 33:1439. 1964. Salvadori, P., Bianchi, B. and V. Cavalli. 1964. Study on the action on amino acids of Penicillia used in making Blue veined cheese. Lait. 33:129. in Dairy Sci. Abstr. 33:3841. 1965. Salvadori, P. and B. B. Salvadori. 1967. Preliminary observations on the lipolysis of some long chain fatty acids by different strains of Penicillium roqueforti. Lait 33:605. in Dairy Sci. Abstr. 33:957. 1968. Sanders, G. P. 1943. A new method for dehydrating cheese. Saravacos, G. D. 1968. Retention of aroma compounds during dehydration of foods. In: Frontiers in Food Re- search. p. 63. Cornell University Proceedings. New York. St. Agr. Expt. Sta., Geneva, New York. Saravacos, G. D. and J. C. Moyer. 1968. Volatility of some aroma compounds during vacuum drying of fruit juices. Food Technol. 33:623. Sato, M., Honda, T., Yamada, Y., Tanaka, A. and T. Kawanami. 1966. A study of free fatty acids, volatile car- bonyl compounds and tyrosine in Blue cheese. In Proc. 17th Int. Dairy Congr. vol. D, International Milchwirtschfts kongress, Munchen, p. 539. 161 Schormuller, J. 1968. The chemistry and biochemistry of cheese ripening. Adv. Food Res. 33:231. Schultz, M. E. 1964. Suitability of different cheese varieties for freeze—drying. Preliminary communi- cation. Milchwissenschaft. 33:559. in Dairy Sci. Abstr. 33:429. 1966. Schwartz, D. P., Gould, I. A. and w. J. Harper. 1956. The milk lipase system. I. Effect of time, pH, and concentration on substrate activity. J. Dairy Sci. 39:1364. Schwartz, D. P., Haller, H. S. and M. Keeney. 1963. Direct quantitative isolation of mono-carbonyl compounds from fats and oils. Anal. Chem. 33:2191. Schwartz, D. P. and O. W. Parks. 1961. Preparation of car- bonyl free solvents. Anal. Chem. 33:1396. Schwartz, D. P. and O. W. Parks. 1963. Quantitative analy- sis of methyl ketones in Blue cheese fat. J. Dairy Sci. 33:989. Schwartz, D. P., Parks, 0. W. and E. N. Boyd. 1963. Methyl ketones in Roquefort cheese. J. Dairy Sci. 33:1422. Schwartz, D. P., Parks, 0. W. and M. Keeney. 1960. Chroma- tographic separation of 2,4-dinitropheny1hydrazine derivatives of aliphatic carbonyl compounds into classes on magnesia. Am. Chem. Soc. Abstr. 333:15B. Selim, S. 1977. Separation and quantitative determination of traces of carbonyl compounds as their 2,4 dini- trOphenyl hydrazones by high pressure liquid chro- matography. J. Chromatography 333:271. Sen, N. P. 1969. Analysis and significance of tyramine in foods. J. Food Sci. 33:22. Shchedushnov, E. 1961. Homogenization of cheese milk in the manufacture of Roquefort cheese. Mol. Prom. 33:17. in Dairy Sci. Abstr. 33;3401. 1961. Shchedushnov, E. 1962. The homogenization of cream in the manufacture of Roquefort-type cheese. Mol. Prom. 33:15. in Dairy Sci. Abstr. 33:2459. 1962. Shimp, J. L. and J. E. Kinsella. 1977. Lipids of Peni- cillium roqueforti. Influence of culture tempera- ture and age on unsaturated fatty acids. J. Agric. Food Chem. 33:793. 162 Shipe, W. F. 1951. A study of the relative specificity of lipases produced by Penicillium roqueforti and Agpergillus niger. Arch. Biochem. 30:165. Shriner, R. L., Fuson, R. C. and D. Y. Curtin. 1964. The systematic Identification of Organic Compounds. A Laboratory Manual. Fifth edition. John Wiley & Sons, Inc. New York. London. Sydney. p. 253. Simonart, P. and J. Mayaudon. 1956a. Chromatographic in- vestigation on cheese. II. Aliphatic acids. Netherlands Milk and Dairy J. 33:156. Simonart, P. and J. Mayaudon. 1956b. Chromatography in- vestigation on cheese. III. Aromatic acids. Netherland Milk and Dairy J. 33:261. Snyder, L. R. 1969. Rapid separations by liquid solid column chromatography. Qualitative and quantitative analysis of hydrogenated quinoline mixtures. J. Chromatog. Sci. 3:595. Snyder, L. R. and J. J. Kirkland. 1974. Introduction to Modern Liquid Chromatography. Wiley-Interscience Publication. John Wiley & Sons. New York, London, Sydney, Totanto. 534 pp. Stadhouders, J. 1956. The composition of Dutch butterfat. I. Seasonal variations in the unsaturated fatty acid composition of butterfat. Netherlands Milk Dairy J. 3:182. in Dairy Sci. Abstr. 33:383. Stadhouders, J. and H. Mulder. 1957. Fat hydrolysis and cheese flavor. I. The enzymes responsible for the hydrolysis of fat in cheese. Netherlands Milk and Dairy J. 33:164. Stadhouders, J. and H. Mulder. 1960. Fat hydrolysis in cheese from pasteurized milk. Netherlands Milk and Dairy J. 33:141. Starkle, M. 1924. Methyl ketones in oxidative decomposi- tion of some triglycerides (also fatty acids) by molds with regard particularly to the rancidity of cocoa fat. I. The significance of methyl ketones in the biochemistry of butter rancidity. (In Ger- many, English translation) Biochemische Zeit- schrift 333:371. 163 Stepaniak, L. and B. Habaj. 1974. Lipolytic activity of 3. roqueforti in strains as determined by manu- facturing procedures, in Proc. 19th Int. Dairy Congr., Vol. IE, International Dairy Congress, New Delhi, 496. Stoke, W. N. 1928. The rancidity of coconut oil produced by mold action. Biochemical J. 33:80. Svensen, A. and E. Ottestad. 1969. Qualitative and quan- titative analysis of flavor components in Normanna cheese. Meierposten. 33:50 and 77. in Dairy Science Abstr. 33:1506. 1969. Tarassuk, N. P. and E. N. Frankel. 1957. The specificity of milk lipase. IV. Partition of the lipase system in milk. J. Dairy Sci. 33:418. Thibodeau, R. and H. Macy. 1942. Growth and enzyme acti- vity of Penicillium roqueforti. Minnesota Agri. Expt. Sta. Tech. Bull. 152. 55 pp. Thomasow, J. Von. 1947. Secondary breakdown products of Blue mold cheese. Milchwissenschaft. 3:354. Titlar, E. S. 1960. A study of the isomerization of the 2,4-dinitrophenylhydrazones of acetophenone. M.S. Thesis. Pennsylvania State University. 42 pp. Van der Ven, B., Bergemann, P. H. and J. C. M. Schogt. 1963. Precursors of methyl ketones in butter. J. Lipid Res. 3:91. Vogel, A. I. 1962. A Textbook of Practical Organic Chemi- stry. Third Edition. John Wiley and Sons, Inc. New York, N.Y. p. 638. Voight, M., Eitenmiller, R., Koehler, P. and M. Hamdy. 1974. Tyramine, histamine, and tryptamine content of cheese. J. Milk Food Technol. 33:377. Walbaum, H. and O. Huthig. 1902. Ueber das ceylon zimmtol. J. Prakt Chem. 33:47. Watt, J. C. and W. 0. Nelson. 1963. Cheese flavoring pro- cess. U.S. Patent 3,072,488. Watts, F. 1886. On the essential oil of lime leaves (Citrus limetta). Preliminary notice. J. Chem. Soc. London Trans. 33:316. 164 Willart, S. and G. Sjostrém. 1959. The influence of sodium chloride upon the hydrolysis of milk fat in milk and cheese. XV. Int. Dairy Congr. 33:1480. Williams, G. G. 1858. XI. One the constitution of the essential oil of rue. Phil. Trans. Roy. Soc. London. 148:199. Wojtkiewicz, A. and G. Inikkoff. 1934. The acceleration of the ripening of cheese, 10th World's Dairy Congr. Milan. Section II:335-339. Wong, N. P., Patton, S. and D. A. Forss. 1958. Methyl ketones in evaporated milk. J. Dairy Sci. 33:1699. Yakhontov, P. 1956. Manufacture of powdered cheese. Mol. Prom. 33(7):16. in Dairy Sci. Abstr. 33:102c. 1957.