CORiFGSETEON BF REE EQUEGEES EXTRACT
8F BEEF éBEPOSE "EESSHE

flxmirs {:60 {its ammo o§ 95:. '53;
R’EECBEGRE‘? mm UREVERSEYY
Fred Hamid Pepper

E969

{has}:

  
   

its.
LIBRARY

Michigan S ta re
Universit 1-,!

This is to certify that the

thesis entitled

COMPOSITION OF THE AQUEOUS EXTRACT

OF BEEF ADIPOSE TISSUE

presented by

Fred Harold Pepper

has been accepted towards fulfillment
of the requirements for

Ph.D. degree in FOOd SCience

..§7,/¢.

Major professor

Date 1.2a 5 EIQ '6 /9/22

0-169

 

 

 

 

(‘3.-
\

I..-'

u‘d. 3";3 F
‘30

 

     
    
   
  
  
   
      
 
  
   

‘w Bsrev‘ z." r» '7 i

u

"was ingot-J1":-

i' V" .lt‘ir“: :nx‘n'... _ n ‘ ‘
t, .w,’ _
1 ., V W” ,1 W1 w _

.1 .‘
1, ' 5.5 a)! : L

me:~.-~;.. .
u‘n‘ugr': u.

.m 69%;: mm. . .vv ,- .
A
4 I

~

r ‘Sfih'ifll‘h: Uhi 1‘ ' ' .‘. V

St francs-:3 3:7.4 3“; 7 n ‘ m ’ ‘_‘
11B (laureates we 3‘9.in . x ».

.‘N m Shit"?! $9} ~31.” "Hg-firm: .

  

 
 

V 1‘” 4‘ ghl :‘i‘l frat on zzt'vt‘et'fiatt.tr.-.;r-~_w- '4 )~ ~.,‘- -

k
‘alL‘t
. .

   

p, 7 1-SiS 610‘. n01 g. "9 94396.}: 1113 .;-n; -- “1» ,whr

., 9‘!» the emttins ma 33631:;

a

 

 

 

ABSTRACT
COMPOSITION OF THE AQUEOUS EXTRACT 0F BEEF ADIPOSE TISSUE

by Fred Harold Pepper

Beef adipose tissue was separated into the water soluble, salt soluble
and insoluble (water-salt—ether) fractions, and the amount of hydrogen
sulfide evolved upon heating each of the fractions was determined. The
yield of hydrogen sulfide per 100 g of adipose tissue was 17.6, 1.7 and
5.5 pH for the water soluble, salt soluble and insoluble fractions, re-
spectively. Thus, the water soluble fraction contributed 71% of the
hydrogen sulfide, while the corresponding contributions from the salt
soluble and insoluble fractions amounted to approximately 7 and 22%.

The water soluble fraction of adipose tissue was subjected to four
treatments (air and nitrogen atmospheres and 3% sodium chloride under each
atmosphere). The sulfhydryl groups and other possible flavor precursors
were determined. The aqueous extract from each treatment was divided into
dialyzable and non-dialyzable fractions. The dialysates were concentrated
by pervaporation and dialysis against a 1 M sucrose solution. The sucrose—
protein concentrates were separated into 10 ml aliquots and stored at -23°C
until used. The diffusates from the treatments were concentrated by either
heating at reduced pressure or lyophilization.

The dialysates were subjected to column chromatographic analysis,
disc gel and starch gel electrophoresis and chemical analysis. Separation
employing gel filtration chromatography was not satisfactory. Starch gel
electrophoresis did not give adequate separation, however, satisfactory
separation of the proteins was accomplished utilizing disc gel electro-
phoresis. The dialySates were negative for glycoproteins, nucleoproteins

and lipoproteins, but positive for heme proteins. The sulfhydryl contents

 

 

 

Fred Harold Pepper

of the nitrogen, air, nitrogen with 3% sodium chloride and air with 3%
sodium chloride treatments were 9.05, 8.17, 16.00 and 15.79 uM per g of
protein. Thus, there was a slight atmospheric effect on the sulfhydryl
contents, since the air treatments for both the salt and salt-free treat-
ments were lower than those under nitrogen. The salt treatments under
either air or nitrogen contained almost twice the sulfhydryl content of
the salt-free treatments. The increase of sulfhydryls in the salt treat-

ment was also reflected in the disc gel electrophoretic patterns, which

revealed a greater number of sulfhydryl containing bands.

Amino acid analysis of the dialysates demonstrated the presence of
lysine, histidine, arginine, aspartic acid, threonine, serine, glutamic
acid, proline, glycine, alanine, cystine, valine, methionine, isoleucine,
leucine, tyrosine and phenylalanine. Glycine, alanine and glutamic acid
comprised approximately 25% of the amino acid residues. There was no
difference in the amino acid content of the proteins from the various
treatments.

The diffusates were satisfactorily separated on Dowex l anion exchange
resin (formate form), however, Bio-Gel P_2 resin gave inadequate separation.
The diffusates from the salt-free treatments under both nitrogen and air
atmospheres were similar to those of the corresponding salt treatments.
However, the diffusates from the salt treatments were more concentrated
because of different procedures of concentrating.

Chemical analyses of all diffusates demonstrated only traces of sulf-
hydryl groups. Diffusates from the salt and salt-free treatments contained
48.24 and 273.20 pH of free amino groups and 3.76 and 50.23 pH of aldoses
per g of lyophilized powder, respectively. All diffusates contained the

same amino acids as the dialysates, but glycine, alanine and glutamic acid

fl

Fred Harold Pepper

  
  
  
   
 
  
  
  
  
  
  
  
  
  
  

comprised over 50% of the amino acid residues. Thin layer chromatography
demonstrated that the diffusates contained creatine, creatine phosphate,
creatinine, uracil, cytosine, some unidentified phosphate containing com-
ponents and a number of fluorescent compounds. Purine bases, nucleosides,
nucleotides and lactic acid were absent.

- A quantitative method for the determination of sulfhydryl and disul-
fide groups in proteins based upon the ultimate formation of an azo dye
was investigated. After reduction of the disulfides, the resulting sulf-
hydryl groups were nitrosated with nitrous acid. The S-nitroso derivatives
were cleaved with mercuric chloride and the resulting nitrosyl chloride
was combined with sulfanilamide to form a diazonium salt. The diazonium

salt was coupled with N-l-naphthylethylenediamine to form an azo dye. The

dye absorbs maximally at 540 nm; and the thiol gives a 1 to 1 stoichio-
netric relationship with the dye. The only substances interfering with the
dye development are those that destroy either nitrous acid or the diazonium

salt.

r2

' k
..§
1 '. 3‘.“
e;-

 

  
   
   
   
  

GOWITION OF THE AQUEOUS BMW
in, ' : 0F BEEF ADIPOSE TISSUE

' 1 . By
Fred Harold Pepper

.A: , , , A THESIS

Submitted to ,2
Michigan State University g
in partial fulfillment of the requirements
for the degree of ‘_.‘
DOCTOR 0F PHILtBOPHY

Department of Food Science

1969 .

 

ACKNOWLEDGMENTS

The author wishes to express his appreciation to his major professor,

   
  
    
  
   
  
  

Dr. A. M. Pearson for his guidance throughout this research program and
for his assistance in the preparation of this manuscript.

Appreciation is extended to Dr. J. R. Brunner for his advice, to Dr.
B. S. Schweigert, Chairman of the Food Science Department, for the facili-
ties which were provided, and to Drs. R. H. Luecke and E. P. Reineke for
serving as members of the guidance committee.

Special thanks are extended to Dr. D. G. Farnum of the Department
,of Chemistry and to Dr. L. R. Dugan, Jr., for counsel on the synthesis of
organic compounds.

Lastly, the author is especially grateful to his wonderful wife,
Ann, for her understanding, encouragement and patience throughout his

graduate study and to his children, Krista and Diana, for their sacrifices.

.? }_; 11

 

__. “7—-

 

4?? .

TABLE OF CONTENTS

INTRODUCT I ON I I O O I O O O O O O O O O I O O I O C O O I O D O C 0

REVIEW OF LITERATURE . . . . . . . . . . . . . . . . . . . . . . . .
Definition of Flavor . . . . . . . . . . . . . . . . . . . . .
Flavor Components of Muscle . . . . . . . . . . . . . . . . . .

Water Soluble Components . . . . . . . . . . . . . . . . .

Meat extracts . . . . . . . . . . . . . . . . . . . .

Meat flavor precursors . . . . . . . . . . . . . . .

Meat flavor volatiles . . . . . . . . . . . . . . . .

Hater Insoluble Protein Fractions . . . . . . . . . . . .
Fatty Tissue Flavor Components . . . . . . . . . . . . . . . .
Flavor Components Due to Browning . . . . . . . . . . . . . . .
Flavor Components Involving Hydrogen Sulfide . . . . . . . . .
Organoleptic Evaluation of Meats and Model Systems . . . . . .

Investigation of Analytical Techniques in Sulfhydryl and
Disulfide Determinations . . . . . . . . . . . . . . . . . . .

General sulfhydryl analytical techniques . . . . . . . . .
General colorimetric, amperometric and spectrophotometric
technique 8 O O O O 0 O O I I O O O O I I O 0

Fluorescence . . . . . . . . . . . . . . . . . . . . . . .
lonicchromphores.oooooooouoooooooooo
Ninhydrin as a cysteine specific reagent. . . . .
Indirect suflhydryl analysis by spectrophotometricm eth hods

METHODS AND MATERIALS . . . . . . . . . . . . . . . . . . . . . . .
Adipose Tissue Sample Collection and Preparation . . . . . . .
Fractionation of Adipose Tissue Samples . . . . . . . . . . . .
Preparation of Muscle Sample . . . . . . . . . . . . . . . . .

Preparation of Dialyzable and Non-Dialyzable Components of
AdiposeTissue.......auo.....-......

Adipose Tissue Analysis . . . . . . . . . . . . . . . . . . . .

Page

Home

14
15
17
20

22

25
25
26
28

29

32
32
33

38

38

Page
Hydrogen Sulfide Determinations . . . . . . . . . . . . . . . . 41

Amino Acid Analysis of Dialyzable and Non-Dialyzable
Comonents I I I I I I I I I I I I I I I I I I I I I I I I 43

Detection of Various Non-Dialyzable Components by Disc Gel and
Starch Gel Electrophoresis . . . . . . . . . . . . . . . . 44

Staining the Disc Gel Electrophoretic Patterns of the Non—
Dialyzable Components . . . . . . . . . . . . . . . . . . 48

Column Chromatographic Separation of Non-Dialyzable and
Dialyzable Components . . . . . . . . . . . . . . . . . . 51

Thin Layer Chromatographic Analysis of Dialyzable Components . 52
Chemical Analysis of Dialyzable and Non-Dialyzable Components . 54
Development of a Method to Determine Sulfhydryl and Disulfide

GTO up 8 o o I O I o I o O O a o o I O I O 0 I O o o I o I I 57

RESUI‘TS AM) DISCIISSION I I I I I I I I I I I I I I I I I I I I I I I 66

~——- "—

Adipose Tissue Analysis . . . . . . . . . . . . . . . . . . . . 66
Hydrogen Sulfide Determinations o o o o I o o I o o o a o o I o 66

Analysis of Dialyzable and Non-Dialyzable Components of the
Water Soluble Fraction of Adipose Tissue . . . . . . . . . 7O

Amino ac id analysis I I I I I I I I I I I I I I I I I I I 71
Starch Gel Electrophoresis of Non-Dialyzable Components . . . . 72

Disc Gel Electrophoresis of Non-Dialyzable Components . . . . . 73

Column Chromatographic Separation of Non-Dialyzable and
Dialyzable COInponentS I I I I I I I I I I I I I I I I I I 79

Thin Layer Chromatographic Analysis of Dialyzable Components . 82
t Chemical Analysis of Dialyzable and Non—Dialyzable Components . 89
Development of a Method to Determine Sulfhydryl and Disulfide

} GI. oups I I I I I I I I I I I I I I I I I I I I I I I I I I 9 3
SUMY I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I 98

BIBLImRAHIY I I I I I I I I I I I I I I I I I I I I I I I I I I I I 100

iv

 

I
[7.2.

. W'F""" "" . ‘ -
.. _

LIST OF TABLES

       
   

Page

Hydrogen sulfide evolved upon cooking adipose tissue and its
mious fractions for 7 m. I I I I I I I I I I I I I I I I I 67

The amino acid composition of dialyzable and non-dialyzable
aqueous fractions of the various treatments . . . . . . . . 72

Rf values for standard and unknown compounds in diffusate II 85

Chemical analysis of dialyzable and non-dialyzable
components.........o...............89

 

avg-

WWI-

Figure

10

11

12

13.

14.

LIST OF FIGURES

Fractionation of beef adipose tissue - Study I . . . . . . .

Fractionation of adipose tissue for flavor precursor
analysis - Study II . . . . . . . . . . . . . . . . . . . .

Amount of hydrogen sulfide evolved with increased cooking

t me I I I I I I I I I I I I I I I I I I I I I I I I I I I I

Comparison of the water soluble components of beef muscle
andadiposetissue.....n...............

Comparison between protein and sulfhydryl distribution in
muscle and adipose tissue extracts . . . . . . . . . . . . .

Staining of various groups from adipose and muscle tissue
extr ac t8 I I I I I I I I I I I I I I I I I I I I I I I I I I

Gel patterns of the water soluble fraction of adipose tissue
employing 5% and 7.5% running gels . . . . . . . . . . . . .

Gel patterns of the water soluble fractions of adipose tissue
subjected to N2, 02, N28 and 023 treatments . . . . . . . .
Column chromatographic pattern of diffusate I components . .
Thin layer chromatographic patterns of diffusate II and
standard samples of creatine phosphate, creatine and

cre at inine I I I I I I I I I I I I I I I I I I I I I I I I I

Thin layer chromatogram of standard compounds and diffusate
II c omonents I I I I I I I I I I I I I I I I I I I I I I I

Thin layer chromatogram of standard compounds and diffusate
II components . . . . . . . . . . . . . . . . . . . . . . .

Thin layer chromatogram of diffusate II components . . . . .

Thin layer chromatogram of diffusate II components . . . . .

vi

Page

34

36

68

74

74

76

76

78

81

81

86

86

87

_-.-o~ .— ~—-‘. ~

INTRODUCTION

For years chemical identity of meat flavor has been an intriguing
problem. Since the primary criterion to consumer acceptance is meat
flavor, studies concerning its identity and origin are most important.
The literature contains many controversies pertaining to meat flavor
components and their origin.

Some researchers have indicated that whole muscle is responsible for
meat flavor (Barylko-Pikielna, 1957), while other workers have concluded
that the flavor components are located only in structural proteins
(Crocker, 1948). Generally, however, most research has indicated that
meat flavor components are located in the water soluble portion of muscle
(Kramlich and Pearson, 1958; Hornstein and Crowe, 1964).

While some investigators have suggested that fatty tissues play only
a minor role in meat flavor (Pippen £3 51., 1954; Hofstrand and Jacobson,
1960), other investigators have demonstrated that fatty tissues are re-
sponsible for differences in the flavor of meat from various species
(Hornstein and Crowe, 1960b; Iasserman and Talley, 1968).

Hydrogen sulfide has been shown to be a major contributor to the
characteristic meaty odor, since the aroma of cooked chicken persists
as long as hydrogen sulfide is evolved (Bouthilet, 1951a). Subsequent
studies on hydrogen sulfide have also implicated sulfur-containing comp
pounds, primarily in the meaty aroma of heated chicken (Minor gt 31.,
1965b). The water insoluble fraction of chicken muscle has been shown

to produce twice the quantity of hydrogen sulfide as the water soluble

    
   
    
  

j‘tion (necohi e_t g” 1964). Similarly, Ham and Hofmann (1965) have

whim that the myofibrillar fraction is responsible for practically all

I.

g .r ef the hydrogen sulfide evolved from beef muscle.
0

It

-. The present study was undertaken to determine the amount and source
. . i ' afichwdrogen sulfide in beef adipose tissues. In addition, other possible

‘5 "

flavor precursors in adipose tissues were identified and studied.

 

 

REVIEW OF LITERATURE
Definition of Flavor

Moncrieff (1967) stated that of all the qualities of flavor, taste
and odor are of the greatest importance. He indicated that texture (i.e.,
smoothness, roughness, particle size and solubility) is next in importance,
with less important qualities being the hotness of spices and the coolness
of menthol. Finally, he indicated that there are the metallic, alkaline
and meaty qualities. He also stated that in addition to the four true
tastes, i.e., sweet, sour, bitter and salty, there are metallic and alka-
line taste sensations.

In a theory that correlated different odor qualities to the physical
properties of the compound, Amoore (1967) categorized these qualities as
having a minimum of seven characteristic odors, namely, ethereal, camphor-

aceous, musky, floral, minty, pungent and putrid. He further concluded

-that molecular shape and/or charge determined the sensory response to

different odors.

Flavor Components of Muscle

Muscle can be divided into two principle components. One is water
soluble and the other is water insoluble. These components have been
studied to learn their role in the production of flavor precursors, and
also as to the origin of volatile compounds evolved upon heating. Each

will be studied herein.

 

‘r———__==—==q

Water Soluble Components

Meat extracts. Early investigators were concerned with the origin
and precursors of meat flavors. In 1956, Wood investigated beef extract
by exhaustively extracting it with methanol. The alcohol-soluble compounds
were identified as glutamic acid, d-alanine, B-alanine, glycine, proline,
methyl histidine, carnosine, anserine, creatine, creatinine, hypoxanthine,
inosine, choline, carnitine, betaine, urea and citrulline, as well as
other proteinaceous and inorganic materials.

In another study, Wood and Bender (1957) concentrated beef extract
to 45% of its initial volume at reduced pressure. Concentration was con-
tinued at atmospheric pressure by holding the solution at 75°C. The final
. concentrate was exhaustively dialyzed against distilled water and the
‘ diffusate was electrodialyzed. The compounds in the diffusate were iden-
tified by paper chromatography as: methyl histidine, alanine, serine,
methionine, leucine, isoleucine, histidine, taurine, carnosine, anserine,
creatine, creatinine, hypoxanthine, inosine, carnitine, choline, urea,
ammonia, citrulline, lactic acid, glycollic acid, succinic acid and B-
hydroxybutyric acid. The main differences between the extracts obtained
by two procedures were: 1) different amino acid profiles, and 2) the
presence of organic acids in the aqueous extract (Hood and Bender, 1957).

Bender gt 3;. (1958) filtered beef extract, concentrated the filtrate
in a rotary evaporator at 40°C and lyophilized the concentrated filtrate.
The freeze-dried powder was dialyzed against distilled water at room
temperature, and the diffusate was concentrated in a rotary evaporator

, at 40°C. All of the amino acids detected previously by Wood (1956) and

r" - —-—.——-

 

 

Wood and Bender (1957) were identified, plus aspartic acid, lysine,
phenylalanine, threonine, tyrosine and valine. Creatine and creatinine

were determined. It was also noted that creatine was found in much larger

quantities than creatinine. The extract was also shown to contain 2.1%
of reducing sugars.

Wood (1961) prepared a diffusate by dialyzing beef extract, but upon
lyophilization, the anhydrous powder rapidly absorbed moisture from the
air and turned brown. Subsequently, the beef extract was treated with
perchloric acid and filtered. The filtrate was neutralized, lyophilized
and solubilized with a small quantity of water. The slightly soluble
potassium perchlorate was removed by filtration, and the filtrate was
lyophilized and stored over phosphorous pentoxide at 0°C until analyzed.
Purine and pyrimidine derivatives were identified as adenosine mono-, di-
and triphosphate, adenine, hypoxanthine and inosine. Glucose-S-phosphate,
fructose-B-phosphate, ribose, glucose and fructose were identified as the
carbohydrates present along with ribose-S-phosphate, which was only
tentatively identified.

Bender and Ballance (1961) concentrated and heated commercial beef
extract. The volatiles evolved were gas chromatographically analyzed.
Many carbonyl compounds were identified, as well as a number of sulfur-
containing compounds, i.e., methyl mercaptan, ethyl mercaptan, methyl
sulfide and hydrogen sulfide. They concluded that raw meat has little
flavor and showed that the potential for flavor development resides in

the water soluble extract.

 

 

Meat flavor precursors. Bouthilet (1951b) declared that the "meaty"
flavor of chicken is found in the water extractable portions of the meat
fibers and not in the organic soluble fraction of the fatty tissues. Pippen
33 31. (1954) further demonstrated that chicken flavor precursors could be
extracted from raw chicken with cold water.

Batzer gt 2;. (1960) extracted fresh beef muscle with water, dialyzed
the extract and concentrated the diffusate. The concentrate was redialyzed
with dialysis tubing of a smaller pore size. This separated the original
diffusate into a high molecular weight fraction and a low molecular weight
fraction. They further separated the high molecular weight fraction into
two parts using a Sephadex column. One was composed of proteins, while
the second contained smaller polypeptides. The low molecular weight
fraction was divided into two portions by a Dowex ion exchange resin.

One fraction contained carbohydrates, while the second was composed of
amino acids. When they heated the polypeptide fraction with fat, a broiled
steak aroma was evolved. When the protein, polypeptide and carbohydrate
fractions were combined and heated in fat, a meaty aroma of the same inten-
sity as the original diffusate was evolved.

Batzer st 31. (1962) further analyzed the polypeptide fraction and
isolated a glycoprotein. 0n heating the glycoprotein with glucose and
inosinic acid, a meaty aroma resulted. Hydrolysis of the protein yielded
the following amino acids: o—alanine, B-alanine, isoleucine, leucine,
proline, serine, valine and two unknown ninhydrin-positive compounds. One

compound, comprising 80% of the unknown fraction gave a positive

 

- ‘-_.-

test for sulfur. Doty (1961) working in the same laboratory reported
that heating of the isolated polypeptide fraction from chicken and pork
loin resulted in almost the same odor as that from beef muscle.

Macy £1 al. (1964a) compared the water soluble diffusates of beef,
lamb and pork. The diffusates of all three species contained alanine,
arginine, aspartic acid, cysteine, cystine, glutamic acid, glycine, his-
tidine, hydroxyproline, isoleucine, leucine, lysine, methionine, phenyla-
lanine, proline, serine, threonine, tyrosine, valine, anserine, carnosine,
glutamine, glycerophosphoethanolamine, l-methyl histidine, phosphoethanola—
mine, phosphoserine, taurine, urea, ribose, fructose, glucose, inosine,
creatinine, hypoxanthine and possibly maltose. The diffusate from the
three species differed in the fact that cysteic acid and ornithine were
only obtained from pork and lamb, while glutathione was found only in
lamb.

Wasserman and Gray (1965) fractionated the water soluble diffusate
of defatted beef muscle. The double dialysis method of Batzer gt 2;.
(1962) was utilized, but analysis showed that the dialyzate and diffusate
of the second dialysis contained the same components. Thus, separation
by this procedure was unsuccessful. The dialysate was separated into
three fractions employing a Sephadex column. Only one fraction possessed
a meaty aroma on heating. This fraction was lyophilized, but the dried
powder absorbed water and turned brown in a very short time. The dark
brown solution developed the aroma of commercial beef extract. The
diffusate was separated into three components by a cation exchange column.

Two components were eluted with water, while the third was released with

' £0).

ammonium hydroxide. After the components were lyophilized, the fraction
eluted with ammonium hydroxide was the only one that developed a meaty
aroma upon pyrolysis.

The ammonium hydroxide-eluted fraction was separated further by these
investigators (Wasserman and Gray, 1965) with an anion exchange resin.
This resulted in three additional components, which were eluted with water,
6 mM and 2 M HCl, respectively. The first two components gave off a meat-
like aroma, but combining these two fractions did not enhance the aroma.
The third fraction contained only hypoxanthine. The components eluted
with ammonium hydroxide contained both free amino acids and low-molecular-
weight water-soluble proteins. Wasserman and Gray (1965) readily deter-
mined the presence of free amino acids, but the water-soluble peptides
were not detected until acid hydrolysis revealed an increase of free amino
acids. Of all the amino acids, only tryptophan and serine showed decreases.
The authors expected a decrease in tryptophan, since it is destroyed by
acid hydrolysis. A decrease in serine was also expected, since glutamine,
asparagine and serine were eluted simultaneously, thus, masking serine.
They explained that acid hydrolysis converted glutamine and asparagine to
their respective amino acids, so that only serine itself was determined.

Seven amino acids were found in the diffusate by paper chromatography
(Wasserman and Gray, 1965). However, eighteen amino acids were found
using the amino acid analyzer. The latter discrepancy could not be explained.
The authors concluded that meaty aroma is primarily due to amino acids,
polypeptides and possibly hypoxanthine.

In a subsequent investigation, Zaika 22 El. (1968) fractionated the

low-molecular-weight water-soluble diffusate of beef muscle utilizing

“——

An

column chromatographic resins. Separation was achieved by gel filtrap

tion (Bio—Gel P-Z), adsorption (Amberlite XAD-Z) and ion exchange (DEAE-
Sephadex) chromatography. Many compounds were identified which evolved

a meaty aroma upon heating. At least a score of compounds were isolated
in the odoriferous fractions. Many of these were present only in trace
amounts. The authors did not elucidate the responsible compounds, but
they demonstrated that many of them could be removed without seriously
affecting the aroma. Such compounds were identified as: tyrosine, phenyl-
alanine, taurine, glutamic acid, inosinic acid, inosine, hypoxanthine,

and possibly creatine and creatinine.

Zaika gt gt. (l968) obtained seven different fractions by ion exchange
chromatography. Only three of them produced aromas, although none were
characteristic meaty aromas. Fraction 1 possessed an amine like odor;
fraction 3 had a potato-like aroma and was pungent, while fraction 4
produced a faint but sweet and acrid odor. When fraction 4 was combined
with either fractions 1 or 3, the aroma became more meaty. When all three
fractions were combined, the aroma was very similar to that of the original
diffusate.

Zaika gt gt. (1968) reported that fraction 4 contained tyrosine,
glucose, fructose, ribOSe, inosine and hypoxanthine, but stated that
tyrosine, inosine and hypoxanthine could be eliminated without seriously
affecting meaty aroma. In a previous investigation, Wasserman and Gray
(1965) had reported that no carbohydrates were found in the fraction re-
sponsible for meaty aroma. These results are completely contradictory,

since Zaika gt gt. (1968) had reported that fraction 4 is necessary for

10

meaty aroma, yet all of the compounds in this fraction were eliminated
as contributors to meaty aroma (Wasserman and Gray, 1965; Zaika gt gt.,
1968).

Koehler and Jacobson (1967) separated chicken into light and dark
meat fractions, slurried each with water and dialyzed the slurries. The
non-dialyzable portions were filtered. Unheated diffusates and filtrates
had similar "raw meat" and "salty' tastes, but heated diffusates of light
and dark muscles possessed a much stronger chicken flavor than their re-
spective filtrates.

The dialysates were lyophilized and separated into four fractions
by gel filtration. The second fraction from the light meat dialysate
possessed a characteristic chicken flavor, both before and after heating,
whereas, similar treatments of the corresponding dark meat fraction re-
sulted in a less pronounced chicken flavor. No chicken flavor was detected
in any of the remaining fractions.

Recombined unheated light meat fractions possessed a weak chicken
flavor, which was intensified upon heating. However, recombination of
the dark meat fractions produced "brothy", "meaty", "ricw' and "sulfury'
flavors, which were more fully developed with heat, but in neither case
were flavors produced that resembled chicken. These findings substan-
tiated the earlier results of Minor gt gt. (1965b; 1966), who demonstrated
that dark meat was responsible for "meaty' aroma, while light meat possessed
a characteristic "chickeny' aroma.

The second fraction of white meat contained sixteen amino acids,

ribose, fructose, glucose, lactic acid, inosine, IMP (inosine monophosphate),

9‘“

t
i
I
t
5

 

11

GM? (guanosine monophosphate), carbonyls, sulfhydryls and small polypep-
tides. The corresponding dark meat fraction contained all the components
of the light meat fraction except arginine, leucine and/or isoleucine,
threonine, tyrosine, valine and polypeptides. The pH of the light meat

fraction was 5.8 compared to 7.2 for the dark meat.

M_e§t flavor volatiles. Bouthilet (1950) steam distilled a chicken-

 

water slurry and concentrated the distillate. The concentrate had an
amine type odor until the pH was lowered by adding monobasic potassium
phosphate. The lowering of the pH eliminated the amine odor and produced
a strong distinct chicken aroma. The author then concluded that broth
concentrate could be used to determine the chemical properties of chicken
flavor.

Kramlich and Pearson (1958) observed that the water soluble components
from fresh beef muscle produced a meaty aroma upon heating. They were the
first to state that the flavor constituents of both cooked and raw meat
appeared to be in the water soluble fraction. The difference between the
aroma of the water soluble cooked fraction and the heated water extract
of raw meat was only one of intensity.

Hornstein gt gt. (1960a) water extracted beef muscle and filtered
the slurry. The filtrate was lyophilized and placed in an evacuated
system, which was heated to 100°C. The volatile components were collected
in a trap cooled with liquid nitrogen and identified as: acetone, formal-
dehyde, acetaldehyde, ammonia, hydrogen sulfide and possibly methyl amine.
These investigators observed that dried beef powder gave off an aroma
similar to roast beef upon heating, but upon boiling, a characteristic

boiled beef odor was evolved.

 

12

In a continuation of the previous study, Hornstein and Crowe (1960b)
fractionated the volatiles of beef and pork muscles. The volatiles
collected in a cold trap were separated into a volatile and non-volatile
fraction by allowing the cold trap to come to room temperature. The
volatile fraction was isolated by collection in an adjacent cold trap,
while the non-volatile fraction remained in the original trap. Hydrogen
sulfide, ammonia, carbon dioxide, acetone, formaldehyde and acetaldehyde
were found in the volatile fraction of water extracts from both species.
Upon analyzing the non-volatile fraction from both species, a maximum
absorbance at 290-2951nn was determined. Infra-red spectral analysis of
the non-volatile material showed beef and pork samples to be identical,
while tests for phosphorous and sulfur were negative. It was also noted
that the odor characteristics from the non-volatile fractions of both
species were the same. The authors stated that the ultra-violet absorbing
compound(s) from both samples may be responsible for the "meaty' aroma.

In elucidating their theory that the non-volatile fraction was re-
sponsible for "meaty' aroma (Hornstein and Crowe, 1960b), Hornstein gt gt.
(1963a) studied the water-soluble fraction of whale muscle. The whale
muscle extract was heated and fractionated as previously described (Horn-
stein and Crowe, 1960b). The non-volatile fraction had an amine type
odor, rather than the "meaty' aroma that was anticipated. However, upon
heating the fraction to 70°C to remove the trimethylamine, a similar
"meaty' aroma was evolved. Gas chromatographic profiles of beef, pork
and whale were the same, except for the trimethylamine peak found in the

whale volatiles.

 

l3

Hornstein and Crowe (1963b) subjected lamb to the same analysis as
beef and pork (Hornstein and Crowe, 1960b). In addition to analysis of
the collected volatile fractions, the original heated solution was also
investigated for any residual aromas, but no appreciable aroma was de-
tected. The collected volatile fraction from lamb had only a very weak
sulfide odor with hydrogen sulfide being barely detectable, whereas, beef
and pork had a high sulfide content. The non-volatile (room temperature)
fraction from lamb demonstrated absorption at 290-295 mm and had a "meaty"
aroma very similar to beef and pork.

Yueh and Strong (1960) slurried ground lean beef with water and
heated the slurry. The volatiles were collected and identified as hydrogen
sulfide, ammonia, acetaldehyde, acetone and diacetyl. In addition, formic,
acetic, propionic, butyric and isobutyric acids were tentatively identi-
fied along with methyl sulfide. No alcohols or esters could be detected,
while hydrogen sulfide, ammonia, diacetyl and acetone appeared to be the
major components responsible for cooked beef aroma.

Jacobson and Koehler (1963) water extracted and dialyzed cooked and
raw lamb. The same amino acids and sugars were found in both diffusates.
The volatile compounds identified in cooked lamb were carbonyls, ammonia
and hydrogen sulfide. Results led them to conclude that variations in
carbonyl compounds may be responsible for the aroma of lamb.

Kramlich and Pearson (1960) refluxed a beef-water slurry and swept
the volatiles into a cold trap with nitrogen gas. The volatiles were gas
chromatographed and identified as acetone, acetaldehyde, methyl mercaptan,

hydrogen sulfide and possibly methyl sulfide.

 

14

Macy gt gt. (1964b) quantitatively determined the components of beef,
lamb and pork diffusates both before and after heating. The lyophilized
diffusates were dissolved in water and divided into two aliquots. One
aliquot was analyzed for various components upon solubilization, while
the second was heated in a boiling water bath for 1 hr. . The heated
aliquot underwent extensive browning and evolved a "brothy' odor in all
cases. Quantitatively, the amino nitrogen compounds of all species seemed
to be quite similar. Taurine, anserine-carnosine and alanine were the
major constituents in both the heated and unheated samples. Other consti-
tuents degraded during heating were glutamic acid, glycine, lysine, serine,
cystine, methionine, leucine, isoleucine and methyl histidine. Phospho-
ethanolamine increased in beef, pork and lamb, whereas, histidine increased
only in beef and pork and methionine only in pork.

Macy gt gt. (1964b) also found that the carbohydrate content of all
three species was about the same. Glucose was present in the highest
quantity, followed by fructose, ribose and an unknown fraction. Ribose
appeared to be the most heat labile, while fructose showed the greatest

heat stability.
Water Insoluble Protein Fractions

Crocker (1948) stated that meat flavor resides in the meat fibers of
cooked meat and not in the expressible fluid. He also indicated that
flavor is not derived from the fat, but is probably obtained upon hydrolysis
of protein. However, Barylko-Pikielna (1957) indicated that the three

fractions of cooked beef, i.e., meat fibers, denatured water—soluble proteins

,
A
I
I
»

 

15

and extracted substances, all participate in the development of full
meaty flavor.

Greenwood gt gt. (1951) reported that the amino acid content of crude
protein was similar in different grades of beef and also in different cuts.
The amino acids in these cuts were stable to cooking. The cooked samples
contained the same percentage of amino acids as the raw samples, except
that isoleucine, histidine and serine all decreased by approximately 10%
in the cooked samples.

Hornstein (1967) water extracted beef muscle, centrifuged the slurry
and retrieved the protein residue. The residue was heated to 100°C under
vacuum and the volatiles were collected in a cold trap. Only traces of
ammonia could be detected and no meat-like aromas were discernible. The
author also stated that heating of the original aqueous extract resulted
in meaty aromas. This provided proof that regardless of their origin,
the flavor precursors are water soluble, while the insoluble substances

make no contribution to flavor.

Fatty Tissue Flavor Components

Barbella gt gt. (1936) found that lambs on a restricted ration yielded
less desirable flavor than similar lambs full fed on the same ration. The
amount of fat and protein was less for the lambs on the restricted diet,
so the authors concluded that the desirability of flavor varied directly
with fatness.

In further investigations, Howe and Barbella (1937) observed that a

characteristic flavor was produced by the fatty tissues. The flavor became

fl

 

16

more pronounced at higher temperatures, especially when browning occurred.
The authors postulated that the flavor characteristics reside chiefly in
the fatty tissues. In agreement, Jones (1952) reported that lean contri-
butes very little to meat flavor, while fat is mainly responsible for
flavor development. On the other hand, Pippen gt gt. (1954) stated that
the flavor precursors of chicken are water extractable and that fat only
contributes to the aroma of the broth.

Hornstein and Crowe (1960b) stated that there are no characteristic
flavor differences in the water-soluble fractions of beef and pork. Thus,
they suggested that the species characteristic flavor differences reside
in the fatty tissues. They melted and filtered fatty tissues from both
species. The filtrate was taken up in petroleum ether and passed over an
ion exchange resin to isolate the fatty acids. The fatty acids were then
measured quantitatively.

Hornstein and Crowe (1960b) also heated fatty tissues under reduced
pressure and in the presence of air. The fatty acids found upon heating
adipose tissue under vacuum were the same as those retained by the ion
exchange column. Upon heating in air, however, large concentrational
changes were noted in fatty acids due to the hydrolytic effect of water
on glycerides. They found that the fatty acid gas chromatographic profiles
varied both qualitatively and quantitatively between beef and pork fatty
tissues.

Similar results were obtained for carbonyl compounds by Hornstein
and Crowe (1960b). Heated adipose tissue showed a low carbonyl content.

However, the fatty tissues heated in air evolved a much greater quantity

‘W———

of carbonyl compounds due to oxidation. They concluded that similar meaty
flavors are obtained upon heating lean from both beef and pork, but that
flavor differences have their origin in the fatty portions. They further
concluded that fatty tissues may have different ratios of compounds and
different lipid soluble components which may contribute to flavor.
Hornstein gt gt. (1963a) analyzed the fatty acid fraction from fatty
tissues of the whale. The gas chromatographic profile was dissimilar to
the same fraction from beef and pork. Whale fat had a high percentage of
long chained polyunsaturated fatty acids. Since the water soluble fractions
of all three species were similar but the fatty fractions differed, the
authors concluded that the species characteristic flavor originated from
, differences in fat composition.
Hornstein and Crowe (1965b) cooked lamb adipose tissue in water at
. 100°C and noted a very strong mutton odor. They theorized that carbonyl
compounds in the fat were responsible for mutton odor, even though only
a small quantity of these compounds was found in comparison to beef and
pork. 0n the basis of several studies, Hornstein and Crowe (1964) con-
I cluded that species flavor differences are determined by components in

the fatty tissues.

Flavor Components Due to Browning

Howe and Barbella (1937) observed that the characteristic lamb flavor
is produced in the fatty tissues, but becomes more pronounced at higher
j temperatures, especially upon browning. Herz and Shallenberger (1960)

t reported that heating of glucose with free amino acids resulted in a

   

 

18

variety of aromas. They indicated that the aromas vary not only with
different amino acids, but also at different reaction temperatures. When
cystine and cysteine were heated with glucose, only sulfide aromas were
detected. Upon using methionine, it resulted in evolution of a potato-
1ike aroma. However, no meaty aroma was developed with any of the amino
acids. Pearson gt gt. (1962) spectrophotometrically measured the colored
components produced by browning of lean pork. They found that the color
developed was closely correlated to the level of reducing sugars in the
tissues. However, there was little correlation between color and free
amino nitrogen. In a subsequent paper, Pearson gt gt. (1966) concluded
that browning of pork muscle was primarily due to the Maillard reaction
with some of the brown color originating from caramelization.

Lobanov and Wolfson (1958) concluded that degradation products of
free amino acids and reducing sugars in meat products brought about the
formation of unsaturated melanoidins, which they stated are responsible
for the yellow tint of meat broths. Upon refluxing free amino acids and
reducing sugars, melanoidins were formed that smelled, tasted and had the
color of meat broths.

Wood (1961) reported that ribose-S-phosphate reacted more readily
with amino acids than other reducing sugars, producing the browning
reaction. Browning products of glucose and a number of amino acids
possessed a bitter flavor, but upon heating glucose with a combination
of components found in meat extracts, a meaty flavor was obtained. The
author then stated that heating developed the brown color and flavor,

which are characteristics of the Maillard reaction.

 

19

Macy gt gt. (1964b) stated that the interaction of carbohydrate and
amino nitrogen during heating is one of the most important aspects in the
production of desirable flavors and odors in meat. Hornstein and Crowe
(1964) separated a reducing sugar fraction and an amino acid fraction from
the water—soluble portion of beef muscle. They noted that no meaty aroma
was evident upon heating each individually. However, when both fractions
were recombined and heated, a meaty aroma occurred.

Wasserman and Gray (1965) separated the components of an aqueous
beef extract and obtained a specific fraction which was responsible for
the meaty aroma upon heating. The fraction was analyzed and contained
aspartic acid, threonine, serine, glutamic acid, proline, glycine, alanine,
valine, methionine, isoleucine, leucine, tyrosine, phenylalanine, lysine,
histidine, arginine, tryptophan and ammonia. However, glucose and ribose,
which are associated with the Maillard reaction, were not found.

Tonsbeek gt gt. (1968) reported the presence of 4-hydroxy-5-methyl-
3(2H)-furanone (I) and its 2,5-dimethyl homolog (II) in an ether extract
of beef broth. Compound I had a caramel type color, and component II had
the aroma of roast chicory root. These compounds could be synthesized by
combining reducing sugars and amines. Both compounds showed strong absorp-
tion at approximately 287 nm. The authors concluded that these odorous
dihydrofuranones contribute to the composite flavor of beef. Since
Hornstein and Crowe (1960b) isolated unidentified compound(s) from water
extracts of beef having similar absorption patterns (270-295 nm), it is

probable that they isolated the same or similar compounds.

 

 

Flavor Cogponents Involving Hydrogen Sulfide

Many papers have been published on the detection of hydrogen sulfide
in the volatile fraction of chicken (Sadikov gt gt., 1934; Bouthilet,
1949, 1950, 1951a, 1951b; Pippen and Eyring, 1957; Lineweaver and Pippen,
1961; Kazeniac, 1961; Mecchi gt gt., 1964; and Minor gt gt., 1965a, 1965b,
1965c, 1966), beef (Stahl, 1957; Yueh and Strong, 1960; Kramlich and
Pearson, 1960; Hornstein gt gt., 1960a; Hornstein and Crowe, 1960b; Hamm
and Hofmann, 1965), lamb (Hornstein and Crowe, 1963b; Jacobson and Koehler,
1963), and pork (Hornstein and Crowe, 1960a).

Sadikov gt gt. (1934) reported that hydrogen sulfide was produced
upon heating chicken muscle and appeared to be due to complete decomposi-
tion of glutathione, cystine and methionine. The authors postulated that
some of the hydrogen sulfide was reabsorbed by the protein and oxidized
to free sulfur. They also stated that glutathione was an important chicken
flavor precursor. Similarly, Bouthilet (1951b) reported that glutathione
was a possible hydrogen sulfide precursor in chicken volatiles. Mecchi
gt gt. (1964) observed that although glutathione released hydrogen sulfide
about 160 times as fast as chicken muscle, there was not enough gluta-
thione to play a significant role in hydrogen sulfide production. There-
fore, they stated that sulfur-containing proteins were the precursors of
hydrogen sulfide.

Minor gt gt. (1965a, 1965b, 1965c, 1966) refluxed chicken-water
slurries for various times under different atmospheres and compared the
volatile fractions of light and dark muscles. Hydrogen sulfide was de-

tected under all conditions at temperatures greater than 80°C. They

 

21

reported that upon heating older chickens and dark muscles more hydrogen
sulfide was evolved than from younger chickens or light muscles. Minor
a g. (1965b) also reported that breast muscle volatiles had a typical
"chickeny" aroma, whereas, leg muscle possessed a "chicken-like" aroma,
but with a definite "neaty" aroma reminiscent of roast beef. Upon passing
the volatiles through various solutions, they found that elimination of
sulfur components resulted in an almost complete loss of "meaty" odor,
while removal of the carbonyl compounds eliminated the "chickeny" odor,
but intensified the "neaty" aroma. They thus stated that sulfur com-
pounds were important to meaty odors.

Mecchi gt ‘a_.l. (1964) reported that during 2 hr of boiling, leg and
breast muscle produced hydrogen sulfide at about equal rates. Isolated
leg and breast muscle proteins produced 75 and 84 % as much hydrogen
sulfide as their respective intact muscles. Nonprotein muscle fractions
evolved no detectable amounts of hydrogen sulfide during the 2 hr cooking
period. They also stated that 9-12 ug of hydrogen sulfide
are evolved per min for every 100 g of chicken muscle. This is a
significant amount since hydrogen sulfide can be detected in concentrations
of as little as 1.1 ug per liter of air (Fieldner e_t fl. 1931).

Hornstein a g. (1960a) determined that 0.1 mg of hydrogen sulfide
was evolved per gram of lyophilized aqueous beef muscle extract. They
also noted that dried beef powder evolved a roast beef aroma upon heating.
0n the other hand, boiling of the powder resulted in boiled beef aroma.

Ham and Hofmnn (1965) reported no loss in free sulfhydryl groups

upon heating beef myofibrills to 70°C, but upon attaining a temperature

A

 

 

22

of 80°C hydrogen sulfide was detected in the volatile fraction. When
myofibrils were heated to 120°C for 30 min , there was a decrease in
sulfhydryl groups. All of the decrease in sulfhydryl groups could be
accounted for by an increase in disulfide groups except for 7%.

They also found that heating the myofibrils under a nitrogen atmosphere

did not significantly decrease the free sulfhydryl groups. They reported
that the formation of hydrogen sulfide increased with temperature and
cooking time. Upon heating myofibrils at 120°C for 5 hr, the quantity

of both free sulfhydryls and disulfides declined. Whole beef muscle and
beef myofibrils contained essentially the same quantity of hydrogen sulfide
per unit myofibrillar protein, which suggests that all hydrogen sulfide
originates in the myofibrillar fraction. This led the authors to conclude
that hydrogen sulfide originates from the readily available free sulfhydryls
of the structural proteins and not from the water soluble fraction of beef

muscle.

Organoleptic Evaluations of Meats and Model Systems

In the investigation of chicken flavor, Minor gt gt. (1966) devised
three model systems, which were primarily composed of compounds found in
chicken broths and distillates. The first system was composed of sodium
sulfide, lactic acid, water and in some cases carbamyl phosphate. The
second system contained all the compounds present in the first system plus
glutathione and diacetyl. The third system consisted of methionine in
addition to all compounds in the second system. The systems were heated

to 180°F. They organoleptically evaluated the systems after heating.

A

 

 

 

23

The first system possessed an egg-like odor without carbamyl phosphate
and an odor similar to chicken volatiles upon addition of carbamyl phos-
phate.

The second system used by Minor gt gt. (1966) had a chicken odor and
an egg-like taste. The third system yielded a strong sulfide, oily
butter-like aroma with the connotation of breast meat broth odor. The
investigators reported that the addition of monosodium glutamate (MSG)
improved the taste of the first and third systems. An additional improve-
ment was noted when disodium inosinate (DSI) and disodium guanylate (DSG)
were added to the above systems in addition to MSG. However, the systems
with the added flavor enhancers did not produce a true chicken broth
flavor. The addition of taurine to the third system plus MSG, D61 and DSG
imparted a serum-like taste more similar to dark muscle broth. No change
was noted when creatine was added;

Klose gt gt. (1966) were somewhat skeptical of the conclusion drawn
by Minor gt gt. (1965b, c) that removal of the volatile carbonyls elimin-
ated the "chickeny aromw', because the acidic hydrazine solution would
also remove basic compounds. In subsequent studies, these authors con-
ducted odor evaluation tests on heated chicken volatiles by utilizing
solid absorbents, which would neither chemically modify nor concentrate
the effluent gases. They found that more hydrogen sulfide and ammoniacal
components were produced by dark meat than light meat. Results indicated
that chicken aroma was due to the blending of hydrogen sulfide with the
amoniacal compounds and another fraction with an unpleasant aroma was

not characterized. Upon recombination of various effluents, they could

24

not duplicate the original chicken aroma, although chicken aroma was
developed upon heating the solid absorbent and purging it with nitrogen
gas.

Wasserman and Talley (1968) conducted organoleptic identification
experiments on beef, lamb, pork and veal. When the panel was given samples
without standards for comparison, only 33% of the panelists iden-
tified all four meats correctly. Upon repeating the same experiment, but
with known samples present for comparison, 66% identified all four unknowns.

Upon conducting identification investigations on beef, Wasserman and
Talley (1968) reported that panelists identified roasted beef roasts,
with all visible covering fat removed, as beef 82.0% of the time.
lean ground and roasted beef (devoid of fat) was identified as beef by
42.5% of the panel, while naturally trimmed roasts were ground, roasted
and correctly identified 71.5% of the time, whereas, a mixture of lean
roast beef combined with 10% of beef adipose tissue was ground, roasted

and then identified correctly by 90.2% of the panel. These results led

 

the authors to investigate the role played by fatty tissues in species
characteristic flavors. They combined ground lean veal, a bland meat,
with 10% beef adipose tissue. The panel identified it as beef in 40.8%
of the cases. Beef adipose tissue was extracted according to Folch gt gt.
(1957) and the solvent-free fat extract was added to lean veal to a level
of 19% fat. The panel identified the mixture as beef 38.5% of the time.
On investigating pork and lamb flavor, Wasserman and Talley (1968)
found that compound(s) in the fatty tissues imparted a species specific
flavor. In pork, the fatty tissue flavoring factor(s) were water soluble,
L whereas, in lamb the factor(s) were due to component(s) of the fat or fat

soluble fractions.

 

r———

_—___-___. .-

 

25

Investigation of Analytical Techniques in Sulfhydryl and Insulfide Deter-
mination

Numerous methods have been developed for the determination of free
sulfhydryl and disulfide groups in proteins and other biochemical compounds.
The advantages and disadvantages of the available methods will be discussed

herein.

General sulfhydryl analytical techniques. In a review concerning the
estimation of thiol and disulfide groups, Leach (1966) discussed the ad-
vantages and disadvantages of several methods. He indicated that iodine,
ferricyanide, hydrogen peroxide and other oxidizing reagents should be
avoided since they lack sulfhydryl specificity. He further noted that
metal salts of copper, mercury, silver and zinc combine specifically with
thiols when used in very low concentrations. It was also noted that the
conCentration and pH of alkylating agents such as iodoacetate, iodoaceta—
mide, methyl bromide and N-ethylmaleimide (NEM) must be carefully controlled
to prevent reactions with -SCH3, -NH2 and -0H sites, however, these
alkylating agents are more -SH specific than either oxidizing agents or
metal salts. He finally stated that organic mercurials (RHgX) are the
most -SH specific compound available, but he also acknowledged that simple
alkylmercuric halides and alkylating agents reacted with labile disulfides.
Thus, these methods gave high sulfhydryl values due to slow hydrolysis of

the disulfide bonds.

I"
v
’n
1

 

 

26

General colorimetric, ggperometric and spectrophotometric techniques.

Leach (1966) designated three general techniques for determining
sulfhydryls: colorimetric, amperometric and spectrophotometric. He
classified the colorimetric methods as qualitative. Three common colori-
metric methods exemplified by Leach (1966) were sodium nitroprusside,
dithizone and thiofluorescein. Upon using sodium nitroprusside, he stated
that development of a pink-violet color indicated the presence of thiols.
Dithizone solutions changed from purple to yellow to confirm a thiol con-
tent, while thiofluorescein changed from colorless to blue in thiol-con-
taining solutions.

The second general technique for sulfhydryl determinations outlined
by Leach (1966) was amperometric titration. This method is particularly
useful for determining sulfhydryl groups in colored compounds and substances
that are partially or completely insoluble.

Leach (1966) stated that the disadvantages of titrating with silver
nitrate or mercuric chloride were their tendencies to complex with the
titrated mercaptides and the limited specificity of silver nitrate. He
listed two types of electrodes for use in amperometric titrations - rotating
platinum and dropping mercury electrodes. The advantages of the rotating
platinum electrode are its simple, inexpensive construction, its sensitivity

'5 M -SH) and its use with different metal ions (Cu, Ag, Hg). The dis-

(1o
advantages are its tendency to become "poisoned', micro-fractures at the
glass-platinum joint, low over-voltage for hydrogen discharge (thus limiting
the working plateau) and the inability to titrate disulfide groups with
mercuric chloride. The advantages of the dropping mercury electrode are

its trouble-free service and its high over-voltage. The main disadvantage

is its low sensitivity (10‘4 M .59).

 

27

The third general sulfhydryl technique cited by Leach (1966) was
spectrophotometric analysis. These methods, as with amperometric titra-
tions, are used for quantitative analysis. He reported that using p-
mercuribenzoate as the sulfhydryl reagent gives relatively sensitive

'4 -10'5 M -SH) and specific results. The disadvantages of this method

(10
were the common absorption of the reagent, products and protein, which

all absorb in the same ultra-violet region (250-255 nm). The molar ab-
sorptivity was affected by pH, urea and ethylenediamine tetraacetate.

Also, the wavelength was read on the slope of the absorption peak, thus,
small spectral shifts resulted in large absorbance errors. Finally, Leach
(1966) stated that other disadvantages were the limited reagent specificity
(steric hinderance) and its instability in solution.

In discussing the use of NEM to determine sulfhydryls, Leach (1966)
stated that the only advantage was that protein precipitation did not inter-
fere with the analysis. He pointed out several disadvantages. The product
had a low molar absorptivity, reactivity was low and the reagent was un—
stable in alkaline solutions. Absorbance was read at 300 nm. The low
sensitivity of the reagent and the necessary high protein content resulted

in high blank absorbance.

Fluorescence. Fluorescence, either development or quenching, for
determination of sulfhydryl or disulfide groups has the advantage of great
sensitivity. However, the methods suffer from limited application and

narrow experimental conditions.

28

McNeil and Beck (1968) developed a fluorometric determination of
glutathione employing o-phthalaldehyde (OPT). However, they reported the
method to be specific only for glutathione.

Karush gt gt. (1964) found that fluorescein mercuric acetate (FMA)
had a molar absorptivity of 7.8 x 104 at 497 nm in 1.0N NaOH. They also
stated that at a neutral pH sulfhydryls quench the fluorescence while
disulfides do not. However, at a higher pH, both groups quench the fluor-
escence. The pH of the solution, the concentration of the sulfur-containing
compounds and the FMA concentration must be carefully controlled, since a
stoichiometric relationship between quenching and disulfide concentration

exists only over very low disulfide concentrations.

Ionic chromophores. The use of chromophore producing disulfides is
advantageous because of the specificity in sulfhydryl determinations and
the existence of a molar absorptivity, thus, eliminating the construction
of a standard curve. Ellman (1959) employed DTNB [5,5'—dithiobis (2-
nitrobenzoic acid)] to determine free sulfhydryl groups. Grassetti and
Murray (1967) employed the same principle for thiol determinations, but
the compounds used were either 2,2'- or 4,4'-dithiodipyridine. The principle
disadvantage of these methods is the inability to determine disulfide groups.
However, Cavallini gt gt. (1966) used DTNB to determine disulfides in pro-
teins after reduction with sodium borohydride, but the method is quite
laborious.

Zahler and Cleland (1968) described dithioerythritol (DTE) as a sensi-
tive and specific reducing agent for disulfides. After disulfide reduction,

the sulfhydryl groups were determined with DTNB. When DTE was used to

29

reduce the disulfide bonds of bovine serum albumin, however, the results
were not satisfactory. DTE was useful as a disulfide reducing agent for

low molecular weight disulfides.

Ninhydrin as a cysteine specific reagent. Gaitonde (1967) determined
cysteine with an acid ninhydrin reagent. He reported that the chromophore
produced had a molar absorptivity of 2.8 x 104. Cystine was determined
after reduction with dithiothreitol, which did not interfere, since the
determination is specific for cysteine. The disadvantage of the method

is its inability to determine cysteine residues in proteins.

Indirect sulfhydryl analysis by spectrophotometric methods. Saville
(1958) developed a method for the indirect determination of thiols, and
Liddell and Saville (1959) slightly modified the method for the determina-
tion of cysteine and reduced glutathione. The method was based on the
determination of nitrous acid, which was quantitatively cleaved from the
S-nitroso derivative of the thiol. The thiol was reacted with an excess
of nitrous acid, which formed a stable S-nitroso derivative. The excess
nitrous acid was eliminated by treating the solution with ammonium sulfamate.
The S-nitroso compound was then combined with a solution containing mercuric
chloride and sulfanilamide. The mercuric chloride complex (HgC14'2) cleaved
the S-N bond to produce a mercuric mercaptide and nitrous acid. The ‘
sulfanilamide was added simultaneously with mercuric chloride because it
favorably competed with ammonium sulfamate for the liberated nitrous acid.
In an acid medium, the nitrous acid combined with sulfanilamide to form
p-sulfamoylbenzenediazonium chloride. N-l-naphthylethylenediamine‘ dihydro-

chloride was added to the diazonium salt solution and the electrophilic

3O

diazonium salt was rapidly coupled to the nucleophilic aromatic amine to
form.the intense reddish-violet azo dye p-E4-[(2-aminoethy1)-amino]azo;
benzenesulfonamide.

Liddell and Saville (1959) stated that the only critical points to be
observed were the initial amount of thiol to be determined and the final
dilution volume. They indicated that the timing, accuracy of other reagent
additions and concentration of acids are only of secondary importance. If
the concentration of nitrous acid was doubled, the calibration results never
deviated more than 2%. By allowing the excess nitrous acid to react with
the thiols up to 6 min, and the ammonium sulfamate to react up to 5 min,
little change was noted in the results. Upon increasing the acid content
tenfold in the sulfanilamide and naphthylamine solutions, the time for
color development was increased to 20 min, while the final results were
practically unchanged.

The possibility of interference in the thiol determination by various
substances was investigated by Liddell and Saville (1959). Secondary amines,
which form.NLnitroso compounds, were added to the reaction mixture. Up to
5 mM dimethylamine was added with no notable effect. A lOOO-fold excess of
glycine, alanine, tyrosine, arginine, valine, lysine, leucine, isoleucine,
histidine, threonine, serine and proline produced no interference. They
indicated that nitrosation of cysteine was completed before any deamination
of the excess amino acids occurred. Cystine, aldehydes, ketones and alcohols
did not interfere with the reaction. Chloride and bromide ions did not
interfere, but large amounts of mercuric, silver and cupric ions caused
serious interference. Zinc ions did not interfere, thus, cystine can be

determined as cysteine after reduction with zinc and HCl.

31

Leach (1966) stated that alkaline hydrolysis degraded both -SH and
-SS-groups and that acid hydrolysis destroyed -SH groups. Thus, the main
disadvantage of this method is its inability to determine cystine residues

in proteins without hydrolysis.

METHODS AND MATERIALS

Beef adipose tissue was analyzed and fractionated to determine the
amount of hydrogen sulfide evolved upon heating, to ascertain the principle
fraction responsible for hydrogen sulfide production, and to determine

hydrogen sulfide precursors and other flavor precursors in this fraction.
Adipose Tissue Sample Collection and Preparation

The trimmable subcutaneous and intermuscular fatty tissues from three
U.S. Choice grade steer carcasses were combined and ground twice at 4°C
through a 12.7 mm.plate, and twice through a 3.2 mm plate with thorough
mixing between each grinding. The samples were placed in Cry-o-vac bags,
evacuated, purged with nitrogen, reevacuated and stored at -29°C until used.
Samples were used to determine the amount of hydrogen sulfide evolved and
its primary source (Study I).

A second study (Study II) was designed to analyze for hydrogen sulfide
and other flavor precursors. The adipose tissue was collected from.three
different U.S. Choice grade steer carcasses and coarsely ground twice at
4°C with thorough.mixing. The ground tissue was mixed for 21 min in
an evacuated Hobart vertical cutter-mixer, and the mixer was brought to
atmospheric pressure with nitrogen.

The adipose tissue was separated into four portions and divided into
500 g aliquots. The first portion was placed in Cry-o-vac bags. Each bag
purged with nitrogen several times, evacuated and sealed (N2). The second

portion was similarly treated except that the aliquots were purged with air

32

33

(0 ). The third portion was prepared the same as portion one, except that

2
3% of solid sodium chloride was added (N28). Pertion four was prepared
the same as portion two, except that 3% of solid sodium chloride was added

(025). All samples were stored at -23°C for a two month period.

Fractionation of Adipose Tissue Samples

The fractionation procedures utilized in Study I are shown in Figure
1. All water used in this study was deionized and distilled. Aliquots of
adipose tissue (2 x 75 g) were mixed with cold (4°C) water in a 250 ml
stainless steel homogenizer flask and homogenized for l min with a Virtis
"45" homogenizer, while being cooled in an ice-water bath. The homogenate
was placed in centrifuge bottles from.which the necks and shoulders had been
removed, and was centrifuged in an International Refrigerated Centrifuge
(Model PR-Z) at 1600 x g at 0°C for 30 min. Centrifugation separated the
homogenate into three phases: (1) the bottom.phase - composed of a small
amount of mucous-like material, which was discarded; (2) the middle phase
or the aqueous layer; and (3) the top phase or the fatty fraction containing
the fat and remaining insoluble protein. The aqueous phase was decanted
and the bottom.phase was discarded. The top phase was extracted with water
and centrifuged as previously described until almost no color remained in
the aqueous layer. The aqueous extracts were combined, filtered through a
milk filter to remove all fatty particles, lyophilized and stored in a
dessicator at 4°C until used.

The top phase was extracted and centrifuged as previously described
using 0.45 M sodium chloride instead of water. The combined salt soluble

extracts were pervaporated and dialyzed against water at 4°C until no salt

34

Adipose Tissue Sample

Homogenized for l min with H O.
Centrifuged at 1600 x g for 20 min at 0°C.
1

 

I

Aqueous Fat phase Insoluble phase
phase (fat and insoluble (insoluble mucous-like
’ protein) material)(discarded)
Filtered and Extracted with 0.45 M NaCl.
lyophilized Centrifuged at 1600 x g for

20.min at 0°C.

A

salt soluble fat and insoluble
ph se protein

I.

Extracted with water at 0°C

 

 

 

Pervaporated ~until no saltjremained
and d'alyzed I '
ueous phase fat and insoluble
e47 , - discarded) protein
Diffusate Dialysate exhaustively extracted
(discarded) (lyophilized) with diethyl ether
Ether extract ResiLue
(discarded) (lyophilized)

Figure l. Fractionation of beef adipose tissue.- Study I

35

remained. The dialysis tubing had previously been boiled in water for 2
hr. The salt soluble fraction was then slurried with water, lyophilized
and stored in a dessicator at 4°C.

The top phase was extracted with water and centrifuged until all salt
was removed. The salt-free fatty residue was extracted with cold (4°C)
diethyl ether until no further fatty material was apparent. The excess
ether was removed under reduced pressure and the insoluble (water-salt-
ether extracted) fraction was slurried with water, lyophilized and stored
at 4°C in a dessicator.

Fractionation procedures for Study II are outlined in Figure 2. All
water used in this study was deionized, distilled and saturated with toluene
to prevent microbial growth. Aliquots of adipose tissue samples (100 g)
were successively homogenized with 125, 75 and 50 ml portions of borate
buffer (pH 7.8; u = 0.1) containing 2 mM EDTA (disodium ethylenediamine-
tetraacetate) for l.mdn under the conditions previously described.

Borate buffer was used because it does not interfere with nitrogen or
phosphorous determinations. An ionic strength of 0.1 and a pH of 7.8 was
selected since Ellman (1959) had previously suggested this concentration
and approximate pH for free sulfhydryl determinations. Free sulfhydryl
groups are stable at pH 7.8, and the low saltconcentration would have little
affect on the water soluble proteins, while the salt soluble fraction would
not be dissolved. A boric acid sodium borate buffer was prepared by calcu-
lating the necessary concentrations to give an ionic strength of 0.1 and a

pH of 8.0. Actually a pH of 7.8 resulted due to the low pH of the water.

36

Adipose Tissue Sample

Extracted with borate buffer
and centrifuged at 27,000 x g

 

Top layer Aqueous layer Bottom layer
(discarded) (discarded)

Filtered and
centrifuged at 27,000 x g

 

 

 

 

 

 

Top layer Aqueous layer
(discarded)
Filtered, pervaporated
and dialyzed
Diffusate Dialeate
(concentrated and
lyophilized)
Concentrated by dialysis
against sucrose solution
. 4 I?
‘Dlaly ate Dltrusate
(discarded)
Centrifuged 23,000 x g
for 15 min at 4°C
Superthant ReSldue
(frozen and stored) (discarded)

Figure 2. Fractionation of adipose tissue for flavor precursor analysis
- Stady II o

37

The extraction solvent contained 2 mM EDTA, since Scopes (1968) had pre-
viously utilized this concentration for extracting muscle. EDTA was used
since it complexes with and removes any divalent cations, which may act as
catalysts for oxidation of free sulfhydryl groups in the sarc0p1asmic pro-
' teins from muscle.

The aqueous solution was decanted after each homogenization. After
homogenizing, the solutions and fatty tissues were combined, placed in
stainless steel centrifuge cups and centrifuged at 0°C for 20.min at 27,000
x g in a Sorvall (Model RCZ-B) centrifuge equipped with a GSA rotor (5.25 cm
radius). The aqueous layers were collected, combined and filtered through
a milk filter to remove any large particles. The other layers were dis-
carded. The filtrate was then centrifuged. The fatty material was concen-
trated in the top layer and solidified in the cold (0°C) centrifuge. The
layer of fatty material was removed with a spatula, and the aqueous layer
was filtered through Ihatman No. 2 filter paper. The filtrate appeared
slightly turbid indicating the presence of suspended fatty particles.

Further filtration through various types of filter paper (Whatman Nos.
5, 40, 50) did not eliminate the turbidity. Ultracentrifugation at 95,000
x g (average) in an International Preparative Ultracentrifuge (Model B-60)
with an A-l70 type rotor also failed to remove the turbidity. However, the
turbidity was removed by suction filtration using a filter pad (1.5 cm.in
thickness) made by pulverizing Hhatman N0. 40 filter paper trimmings with
deionized distilled water in a Haring blender. A disc of Whatman No. 2
filter paper was placed in a filter funnel attached to a Hachner filter flask

and moistened with deionized distilled water. The paper-water slurry was

38

poured into the funnel and compressed into a pad by suction filtration.
The pad was covered with a disc of Whatman No. 2 filter paper and the
aqueous extract was filtered into a Bachner filter flask cooled in an ice-
water slurry. When filtration was completed, the pad was rinsed with 25
ml water. One drop of toluene was added to every 150 ml of filtrate to

prevent subsequent microbial growth.

Preparation of Muscle Sample

 

A sample of beef muscle was prepared in order to compare the water
soluble extracts from adipose tissue with those from.muscle. Beef muscle
(10.0 g) was trimmed of all fat and combined with a solution of 0.3 M sucrose
in the borate buffer (pH 7.8)(20 ml) previously described herein. The
.mixture was blended at 4°C for l min in a Waring blender. The slurry was
centrifuged in a Sorvall centrifuge (Type 88-1) at 4°C for 15 min at 19,400
x g. The extract was stored at 4°C and used within two days to develop

disc gel and starch gel electrophoretic patterns.
Preparation of Dialyzable and N0n-Dialyzable Compgnents of Adipose Tissue

The filtrate, which contained the water soluble components, was poured
into 2.1 cm (diameter) dialysis tubing. The dialysis tubing was prepared by
boiling in a 2.mM EDTA solution for 20 min, and thoroughly rinsing with water.
It was then boiled twice for 20 min with deionized distilled water, and
thoroughly rinsed. The filtrate was pervaporated and concentrated at 4°C
to a final volume of 50-75 ml. During pervaporation, the tubing was kept

wet with the internal solution in order to prevent drying and cracking.

39

Upon completion of the pervaporation, the dialysis tubing and its con-
tents was placed in a one liter Erlenmeyer flask and dialyzed against water.
The water was changed at 12 hr intervals, until approximately four liters of
diffusate had been collected. The non-dialyzable material was transferred
to 1.53 cm dialysis tubing and concentrated by dialysis for 24 hr at 4°C
against one liter of l M sucrose solution, which contained borate buffer
and EDTA as previously described herein. The sucrose solution had previously
been decolorized with activated charcoal.

Centrifugation of the concentrate at 23,000 x g for 15.min at 4°C in
a Sorvall centrifuge (Type SS-l) produced a supernatant solution and a small
amount of precipitate. The precipitate was discarded, and the supernatant
solution was placed in 6 oz Whirl-Pak bags, frozen and stored at -23°C.

The diffusates were concentrated by either lyophilization or heating
in a flasltevaporator at 40°C under reduced pressure. The flask evaporator
was not allowed to rotate, since rotation resulted in the formation of fine
protein strands. The precipitate resulting from concentration in the flash
evaporator was removed by filtration through Ihatman No. 2 filter paper.

The residue was washed with 10 ml of 4°C water, giving a final diffusate
volume of 90-95 ml. A 10 ml aliquot was lyophilized, and ultimately utilized
for detailed chemical analysis.

Concentration by lyophilization was accomplished by freezing a four
liter sample on a 45.5 cm.x 60.5 cm x 4.5 cm stainless steel tray. The
tray was covered with cheese cloth to prevent loss of the dried powder, and
placed in a RePP Sublimator (Model FFD 42"5). After concentration, a 150

ml portion of water was swirled in the tray to rinse it and to dissolve the

4O

powder. The tray was rinsed again with 50 ml water. Some powder adhered
to the cheese cloth and could not be recovered. The rinses were combined
and lyophilized using a laboratory constructed lyophilization apparatus.
The dried powder was slurried with 90 ml water, and the precipitate was
removed and washed as previously described. The final volume was approxi-

mately 95 m1.
Adipose Tissue Analysis

The moisture, fat and total nitrogen analysis were carried out the
same for all samples, except that the digestion procedure for nitrogen was
carried out at a lower temperature and for a longer time for the sucrose-
protein concentrates. The modification prevented the charred sample from
bubbling into the neck of the.micro-Kjeldahl flask.

The fat and moisture content was determined according to Benne 2£.El°
(1956). Aliquot samples (10 g) of beef adipose tissue were weighed in
disposable aluminum dishes and placed in a drying oven (approximately 100°C)
for a period of 24 hr. The dishes were removed from the oven and allowed
to cool to room temperature in a dessicator. The dried samples were weighed
and the percent moisture was calculated.

Fat content was determined by ether extracting the dried samples using
a Goldfisch fat extractor. After the samples had been extracted for 3.5 hr,
the ether was removed and the extractor flasks containing the ether soluble
material were placed in the oven to remove any residual moisture. The pre-
viously weighed flasks were reweighed after drying, and the fat content was

calculated.

41

Total nitrogen analyses were performed according to the.micro-Kje1dahl
method (American Instrument Company, 1961). The equipment included a rotary
digestion unit, 100 ml.micr0-Kjeldahl flasks and a steam distillation-
condensation apparatus. No attempt was made to differentiate between protein

and non-protein nitrogen.
Hydrogen Sulfide Determinations

All glassware used in this determination was rinsed successively in
chloroform, acetone, tap water and 6 N HCl. It was then washed with Alconox
detergent and rinsed several times with water. After cleansing in this
manner, no residual zinc sulfide could be detected.

Hydrogen sulfide was determined according to the method of Sands gt El°
(1949). A standard curve was prepared using sodium sulfide as the source
of the sulfide ion. One or two sodium.sulfide crystals were rinsed with
water to remove any residual sulfides, dried between acid washed filter
paper, weighed and made to volume with water. The sulfide concentration
was determined by adding an excess of standardized iodine solution to an
aliquot of the sodium sulfide solution, and back titrating with a standard-
ized sodium.thiosu1fate solution. The calibration curve was determined
from serial dilutions of the standardized sodium sulfide solution, but other-
wise employing the method of Sands 23 El. (1949).

The cooking apparatus was essentially the same as that described by
Mecchi Si.§l° (1964). The collection apparatus was equipped with the radial
manifold-trap system of Minor £3 21' (1965b). A series of traps and a Pyrex
brand flow meter, using the 1/4 mm orifice, were attached to the radial mani-
fold. The flow rate of nitrogen gas was.maintained to give a pressure

differential of 120-125 mm of ethylene glycol.

42

A slurry of 150 g adipose tissue and 300 ml water was gently refluxed
for a.period of 7 hr, starting from.the time of turning on the glass-col
heating mantle. The volatiles were swept through each trap with nitrogen
gas previously purified by scrubbing with an aqueous solution containing
2% KMNO4 and 3% HgClQ. Each trap contained 50 ml of 2% zinc acetate. The
volatiles were bubbled sequentially through each trap for a period of 1 hr.
The nitrogen gas flow rate and the heating rate were selected to give a
faint turbidity due to zinc sulfide precipitating in each trap after 1 hr
or heating.

After 7 hr of refluxing, each trap was emptied and rinsed with 2% zinc
acetate. The trap contents and its rinsings were combined and made to a
volume of one liter by adding the necessary amount of 2% zinc acetate solu-
tion. To quantitate the amount of zinc sulfide in each trap, 50 ml aliquots
of the collected zinc sulfide solutions were added to 125 ml Erlenmeyer
flasks and cooled to approximately 0°C in an ice-water bath. To this solution,
5.0 ml of a 0.1% solution of N,N-dimethyl-p-phenylenediamine sulfate in 24 N
sulfuric acid were added and gently stirred. Then 1.0 ml of ferric chloride
.6 H 0 dissolved in 50 ml concentrated H01 and

3 2
diluted to 100 ml) was added. The resulting solution was gently swirled

solution (2.7 g of FeCl

and allowed to stand for approximately 30 min until maximum color was devel-
oped. Absorbance was determined at 745 nm.

The residual zinc sulfide adhering to each trap was determined by adding
2% zinc acetate (50 ml) and developing the color in each trap. The analysis
was carried out as outlined above. Total hydrogen sulfide was obtained by
adding the hydrogen sulfide in the trap contents and rinsings to that adher-

ing to the trap. Hydrogen sulfide was determined for the water soluble,

43

salt soluble and insoluble lyophilized fractions in the same manner. The
slurry in all three cases was composed of 2 g lyophilized powder in 300 ml

water.
Amino Acid Analysis of Dialyzable and Non-Dialyzable Components

Amino acid analyses were conducted on the non-dialyzable and diffusate
fractions. Aliquots (10 ml) of the sucrose-protein concentrate were placed
in 2.1 cm.dialysis tubing and tied to allow a minimum of sample space.
Dialysis was performed against one liter of water for 12 hr, and the water
was changed once. Amino acid analysis was conducted according to the
method of Moore 23 El. (1958), except that the proteins were hydrolyzed
for 22 hr. Qualitative and quantitative amino acid determinations were
accomplished by using a Beckman (Model 120C) amino acid analyzer.

For determination of amino acids in the diffusate, an aliquot (5.0 ml)
was lyophilized. The dried powder was extracted three times with methyl
ethyl ketone containing 5% 6 N HCl (3 x 5 ml), and the extract was filtered
through a fine frited glass filter. The method described by Smith (1960)
was used except that no water was added to the extract before it was re-
moved at 4°C under a stream of nitrogen gas. Two ml of deionized distilled
water was added to the concentrate. The lyophilized diffusate (1.0-1.5 g)
obtained from the salt-containing adipose tissue was extracted twice with
methyl ethyl ketone containing 3% 6 N HCl (2 x 25 ml), and then once with
methyl ethyl ketone (25 ml). The extract was filtered through a fine
fritted glass filter and concentrated at room temperature in a rotary evapor-
ator until almoSt dry. Three ml of deionized distilled water was added and

the concentrate was refiltered.

44

Sufficient HCl was added to the concentrated diffusate to bring it to
6 N. The concentrates were hydrolyzed and amino acid determinations were

carried out as previously described herein.

Detection of Various Non-Dialyzable Components by Disc Gel and Starch Gel

Electrophoresis

 

The method and system of Davis (1964) was used with minor modifications.
Cyanogum.(E. C. Apparatus Co.) was substituted for acrylamide and N,N'-
methylenebisacrylamide. Riboflavin was utilized for photopolymerization of
the gels, but no sample gel was used. A.modification of the method of
Jolley 21 pl. (1967) was used to prepare the running gel solutions. A 7.5%
running gel solution was made by combining three solutions (A, B, C).
Solution A consisted of 6 ma 2 N HCl, 9.1 g Tris [Tris (hydroxymethyl) amino
methane], 0.06 ml TEMED (N, N, N,‘ N'-tetramethylethylenediamine) and deion-
ized distilled water to bring the final volume to 25 ml. Solution B was a
30% (i/V) aqueous solution of Cyanogum, which was filtered through Ihatman
No. 1 filter paper before using. Solution C was made by dissolving 1 ml
riboflavin in 25 ml deionized distilled water. The final running gel solu-
tion was composed of A, B, C and deionized distilled water in a ratio of
l.25:2.50:l.25:5.0, respectively. In making a 5% running gel solution,
solution B contained 20% Cyanogum. In making a 5-10% linear gradient gel,

a 5% gel solution was placed in one chamber and a 10% gel solution was
placed in the second chamber of a two chamber gradient device. The 10% gel
solution consisted of a 40% Cyanogum solution (Solution B). The running

gel was produced by degassing the gel solution, and filling the glass tubes

45

(6.5 cm.x 0.5 cm I.D.) to a depth of 5 cm. The gel solution meniscus was
leveled with deionized distilled water and photopolymerized with fluorescent
light on both sides for 15 min.

The spacer gel consisted of three solutions (I, II, III). Solution I
was prepared by mixing 4 ml 2 N H01, 1 g Tris and 0.06 ml TEMED, and was
diluted to 80 ml with deionized distilled water. Solution II was a 33.3%
aqueous solution of Cyanogum. Solution III was the same as Solution C,
which was described earlier herein. The three solutions were combined to
give ratios of 4.0:l.0:l.67 for solutions I, II and III, respectively.

After the running gel had been polymerized, the water was completely
removed. The spacer gel solution was degassed and immediately placed on
top of the running gel to a depth of 0.75 cm. Upon leveling the surface
with water; the gel solution was photopolymerized for 20 min as described
for the running gel.

After removing the leveling water and placing the tubes in the electro-
phoresis apparatus, 550 ml of tank buffer was poured into the lower reservoir
and 450 ml tank buffer with two drops 0.01% aqueous Bromphenol Blue was
added to the upper reservoir. A stock solution of tank buffer was made by
combining 6.0 g Tris, 28.8 g glycine and deionized distilled water to a
final volume of one liter. The stock solution was diluted tenfold for
ele ctrophoresis.

A 0.025 ml aliquot of sucrOSe-protein concentrate was carefully layered
on the surface of the spacer gel. A current of 2 ma was applied to each gel,
with the positive carbon electrode in the lower reservoir and the negative

carbon electrode in the upper reservoir. The power supply was a Heathkit

46

Model PS-4. When the blue dye front reached the bottom of each tube, the
gels were removed by rimming with a 26 gauge hypodermic needle while in-
jecting water. After the gels were removed, they were placed in their re-
spective staining solutions.

Attempts to separate the water soluble proteins in adipose tissue were
.made employing a system described by Rampton (1969). The spacer gel was

the same as described previously herein. The 7.5% running gel was prepared

‘1 .."‘ 1M5? —.'~

by combining solutions A, B and C (6.4:l.6:2.67). Solution A was made by
combining 4 m1 2 N HCl, 6.1 g Tris, 0.08 ml TEMED and diluting the solution
to 80 ml with distilled water. Solution B was composed of 50% (H/V) aqueous
Cyanogum. Solution C consisted of 1 mg riboflavin in 50 ml distilled water.

Urea gels were also used to separate the water soluble proteins in
adipose tissue (Hampton, 1969). A 5% spacer gel was produced by combining
solutions 1, 2 and 3 (1.6:0.4:0.67). Solution 1 was composed of 4 m1 2 N
HCl, 1 g Tris, 0.06 ml TEMED, 65 ml 10 M urea and distilled water to give a
total volume of 80 ml. Solution 2 contained 33.3 g Cyanogum, 25 ml 10 M
urea and was made to 100 ml with distilled water. Solution 3 consisted of
1 mg riboflavin, 35 ml 10 M urea and distilled water to bring the volume
to 50 m1.

A 6.5% running gel (Hampton, 1969) was.made by mixing solutions I, II
and III (6.4:1.6:2.67). Solution I contained 4 ml 2 N HCl, 6.1 g Tris, 0.08
ml TEMED, 65 ml 10 M urea and distilled water to a final volume of 80 ml.
Solution II consisted of 43.3 g Cyanogum, 25 ml 10 M urea and was made to
100 ml with distilled water. Solution III was identical to solution 3
utilized in preparation of the spacer gel. Polymerization, sample applica-

tion and electrophoresis were carried out as previously described herein.

47

Starch gel electrophoresis was carried out using an apparatus similar
to that described by Smithies (1959). The gel former was circumscribed with
removable Plexiglas blocks which defined the starch gel bed. Platinum
electrodes were used to conduct current from the power supply (Heathkit
Model PS-4).

Horizontal starch gel electrophoresis was carried out in a discontinuous
buffer system at 4°C. The starch gel was made according to a modified method
of Scopes (1964). The starch gel solution was produced by combining 30 g
hydrolyzed starch (Connaught Medical Research Laboratories, Toronto) with
55 ml buffer (0.012 M Tris -0.002 M diethylenetriaminepeItapacetic acid, pH
8.25 at 4°C), and then adding this mixture rapidly to 195 ml of heated
(95°C) buffer. The resulting viscous solution was shaken vigorously for 15
sec, degassed for 2 min and poured into the gel former. A slot former was
placed 11.5 cm from the end of the gel and the gel was allowed to cool at
least 90.min before use. After removing the slot former, samples were placed
in the slots and covered with Saran Wrap to exclude air bubbles in the sample.
The gel was connected to the buffer (0.06 M Tris -0.05 M Boric acid, pH 8.6
at 4°C) in the electrode chambers with filter paper wicks. A current of
10 mo at 400V was applied to the system. Electrophoresis was carried out
at 400V for 6 hr at 4°C.

The gel was stained for 5 min in a dye solution (250 ml distilled water,
250 ml methanol, 50 ml glacial acetic acid and 2.g Amido Black 10B) and
destained overnight in a solution containing 1.0 liter distilled water, 1.0
liter methanol, 200 ml glycerin and 200 ml glacial acetic acid. Destaining

was continued in 7% aqueous acetic acid.

48

Staining the Disc Gel Electrophoretic Patterns of the Non-Dialyzable

Comppnents

Disc gel electrophoretic patterns were stained for proteins, hemoproteins,
sulfhydryl groups, glycoproteins, nucleoproteins, ribonucleic acids, lipo-
proteins and esterase activity. Various dye producing reagents were studied
to determine their intensity and contribution to background staining.

Proteins were stained by immersing the gels in an Amido Black 10B
solution for 20 min. The staining solution was prepared by combining 250
ml deionized distilled water, 250 ml methanol, 50 ml glacial acetic acid
and 2 g Amido Black 10B. The gels were destained and stored in 7% aqueous
acetic acid.

The Coomassie Blue method of Crambach 23 El. (1967) was also used to
stain proteins. The gels were immersed in 12.5% TCA for 30 min, then in a
solution composed of one part 1% aqueous Coomassie Blue and 20 parts 12.5%
TCA for 30 min. The gels were destained in 10% TCA.

Hemoproteins were detected in the gels using a benzidine test. The
gels were immersed in a solution composed of benzidine (0.2 g), 30% hydrogen
peroxide (0.2 ml) and glacial acetic acid (0.5 ml) and.made to a final volume
of 100 ml with deionized distilled water (Smithies, 1959b). Staining re-
quired a total of 10 min. Destaining was carried out in distilled water.

Sulfhydryl groups were determined according to the DDD (2,2'-dihydroxy
-6,6'-dinaphthy1 disulfide) method of Zwaan (1965) as modified by Pitt-
Rivers and Schwartz (1967). The following solutions were prepared: A) 25
mg DDD were dissolved in 15 ml absolute ethanol at 50°C. Then 35 ml of 0.04

M sodium barbital-acetic acid buffer (pH 8.5) was added and the.mixture was

49

heated to 50°C; B) deionized distilled water was acidified with one drop
glacial acetic acid per liter (pH 4.0-4.5); C) 60% ethanol; D) 95% ethanol;
E) 50 mg of either Fast Black K, Fast Blue RR or Fast Violet B were dissolved
in 50 ml 0.04 M barbital-acetic acid buffer (pH 7.0). The gels were incu-'
bated in solution A for 4 hr at 50°C and washed two times for 15 min each
in solution B. Then the gels were washed and dehydrated three times in
solution C and once in solution D for 15 min periods. After the gels were
rehydrated in deionized distilled water, they were stained for 50 min in
solution E. The gels were destained and stored in deionized distilled water
and kept in the dark. Staining with Fast Black K resulted in black bands
with a very dark background. Fast Blue RR gave blue-purple bands with a
fairly dark background, however, Fast Violet B yielded red-violet bands with
practically no background color. Therefore, Fast Violet B was used in all
subsequent staining procedures.

Glycoproteins were stained by Alcian Blue (Gifford and Yuknis, 1965).
The gels were stained overnight in a 0.2% solution of Alcian Blue in 15%
acetic acid. Destaining was accomplished in 15% acetic acid.

The Acridine Orange method of Richards «31 2;. (1965) was used to detect
nucleoproteins. Acridine Orange staining solution was.made as follows:
2.5 g of lanthanum oxide (La203) was combined with 75 ml glacial acetic acid
and dissolved at 90°C. Deionized distilled water (75 m1 at 90°C) was added.
Clearing of the solution indicated the formation of the required lanthanum
acetate [La ((‘2'H302)3]. The solution was cooled and diluted to 490 ml with
deionized distilled water, and 10 g of Acridine Orange was added. The gels

were stained overnight and destained in 7% acetic acid.

50

Ribonucleic acid was determined by the Methylene Blue staining method
of Peacock and Dingman (1967). The staining solution was a 0.2% solution of
Methylene Blue in a 0.2 M sodium acetate-0.2 M acetic acid buffer. Gels
were immersed in 1 M acetic acid for 15-20 min, stained in the staining
solution for 2‘hr and destained in water.

Lipoproteins were stained by Sudan Black B according to Gifford and
Yuknis (1965). The gels were placed in 15% acetic acid for 30 min, stained
overnight in a solution of 40% ethanol saturated with Sudan Black B and
destained in 40% ethanol.

Lipoproteins were also stained by the Oil Red 0 method of Beaton 33 El.
(1961). The staining solution was made by saturating warm.methanol with Oil
Red 0, cooling and filtering. This solution was added to 20% TCA (1:1) and
thoroughly.mixed. The gels were stained overnight in the staining solution
and destained in a methanolsdistilled water-acetic acid (50:50:10) solution.

Esterase activity was detected in the gels according to the method of
Allen 2: El: (1965) with slight modifications. The substrate, coupling, de-
veloping and rehydrating solutions are described as follows: A (substrate
solution)-40 mg abnaphthyl butyrate (or acetate) in 100 ml 0.04 M Tris-HCl
buffer (pH 7.1); B (coupling solution)-70 mg Fast Blue RR in 100 ml 0.04 M
Tris-H01 buffer (pH 7.1); C (developing solution)-absolute ethanol: 10%
glacial acetic acid (3:2); D (rehydrating solution)-10% glacial acetic acid.
The gels were immersed in a.mixture of solutions A and B (1:1) for 1 hr at
37°C, placed in solution C overnight and rehydrated and stored in solution
D. The a-naphthyl butyrate substrate solution yielded darker, more intense
bands than o-naphthyl acetate. Therefore, aanaphthyl butyrate was used in

all determinations.

51

Column Chromatographic Sgparation of Non-Dialyzable and Dialyzable

Components

Attempts were made to separate the sucrose-protein concentrates into
various molecular weight fractions. All gel filtration resins were equili-
brated with borate buffer (pH 7.8, o = 0.1). Aliquots (5 ml) of the
sucrose-protein concentrates were applied to 2.5 x 37 cm columns packed with
either Bio-Gel P-30, Bio-Gel PL100, Bio-Gel A-l.5 m, Bio-Gel A-15.m, or
Bio-Gel A-50 nu Gel filtration eluants were monitored by an Isco column
monitor recorder at 254 nm during the collection of 10 ml fractions.

The dialyzable components were separated into various fractions by
column chromatography. A 5 ml aliquot was applied to the columns composed
of Bio-Gel PL2 (2.5 x 60 cm) and Dowex 1 anion-exchange resin (2.0 x 35 cm).

The Bio-Gel P-2 resin was equilibrated with deionized distilled water
and the fines were removed. The flow rate was approximately 40 ml per hr,
and 5 ml aliquots of the eluant were collected. The eluant was monitored
at 254 nm as described previously.

Dowex l anion-exchange resin was converted to the formate form by
washing the resin with l N NaOH and removing the excess alkali with deionized
distilled water. The resin was then washed with three bed volumes of 6 N
formic acid. The resin was poured into the column, allowed to settle and
then the column was washed with deionized distilled water until no ultra-
violet absorbing materials appeared in the effluent (Lento gt 21., 1964).
Two elution systems were used. The first system described by Lento gt El.

(1964) was a gradient system composed of water, formic acid (0.5 N) and

52

sodium formate (0.2 N). The second system was composed of water, 0.5 N
formic acid, 1 N formic acid, 3 N formic acid and 6 N formic acid. The pH
of the sample was adjusted to 8.0 with 5 N NaOH before application to the
column. The effluent was monitored at 254 nm and collected in 5 ml aliquots.
The column flow rate was approximately 40 ml per hr.

The fractions were collected, concentrated by lyophilization and stored
in a desiccator at 4°C until used. The lyophilized fraction was solubilized

in either 1 or 2 ml of deionized distilled water for thin layer chromato-

graphy-
Thin Layer Chromatographic Analysis of Dialyzable Components

The aqueous solutions obtained by column chromatography were applied to
thin layer chromatographic plates in 1 ul aliquots until sufficient material
had been applied to detect the various components.

Purine and pyrimidine derivatives were detected by several thin-layer
chromatographic methods (Randerath, 1966; Bolliger gt gt., 1965). Purine
and pyrimidine bases and nucleosides were chromatographed on thin-layer
plates coated with cellulose (Eastman Chromatogram Sheet 6064), Cellulose
Powder MN300 and MN300G (Laboratory prepared) and Silica Gel G (Eastman
Chromatograph Sheet 6061 and laboratory prepared) using water as the develop-
ing solvent for both the Silica Gel G and cellulose plates. A solvent
consisting of methanol, concentrated HCl and water (65:17:18) was also used
to develop the cellulose plates.

The Cellulose Powder (MN300 and MN300G) chromatograms were prepared
by combining 15 g of powder with 90 ml distilled water in a Haring blender
and blending for l min. The cellulose layers were approximately 0.35 mm

in thickness.

53

Silica Gel G thin-layer chromatograms (0.25 mm thick) were produced
by spreading a Silica Gel G and distilled water (l.0:2.0) slurry over the
glass plates. Methanol, concentrated HCl and distilled water (65:17:18)
and distilled water were used as the solvent systems to chromatographically
separate purine and pyrimidine bases and nucleosides on the cellulose chroma-
tograms. Distilled water was employed as the mobile phase to develop the
Silica Gel G chromatograms.

Nucleotides were detected by using MN300 and MN300G cellulose powder
MN300 ECTEOLA anion exchange chromatograms. MN300 cellulose powder chro-
matograms were developed with a solvent consisting of nebutanol, acetone,
acetic acid, 5% ammonium hydroxide and distilled water (4.5:l.5:l.0:l.0:2.0).
The solvent system used to develop the MN300G cellulose powder chromatograms
consisted of tert.amy1 alcohol, formic acid and water (3.0:2.0:l.0). MN300
cellulose ECTEOLA chromatograms (0.35.mm thick) were prepared by combining
8 g of exchange material, 2 g of unmodified cellulose powder MN300 and 90 ml
distilled water. The mixture was blended for l min in an electric mixer
and spread on the glass plates. The chromatograms were developed with a
0.15 M aqueous NaCl solution.

Purines, pyrimidines and their derivatives were detected with and with-
out spraying with 1.0 N HCl under ultrapviolet light (254 nm) and by spray-
ing with 2',7'-dichlorofluorescein prior to exposure with ultra-violet light.
Tests for carbonyl compounds were run by spraying the chromatogram.with 3.7%
2,4-dinitrophenylhydrazine in 2 N HCl. Phosphate derivatives were deter-
mined by spraying the chromatograms with either 4% ammonium.molybdate in 1

N HCl (l/V) or 1% aqueous ammonium.molybdate followed after drying completely

54

by 1% stannous chloride in 10% HCl (Randerath, 1966). Amino acids were
detected by spraying with either ninhydrin (Sigma Chemical Co.) or a cadmium-
ninhydrin reagent (Blackburn, 1966). The cadmiumpninhydrin reagent was
prepared by combining 1 g cadmium acetate, 100 ml water, 20 ml acetic acid
and 1 liter acetone. The cadmiumpninhydrin treated chromatograms were
allowed to develop at 4°C over sulfuric acid, whereas, the ninhydrin
treated chromatograms were developed from.3-5 min at 100°C.

Creatine, creatine phosphate and creatinine were detected using Silica
Gel chromatograms (Eastman Chromatogram.Sheet 6061) as the stationary phase,
and using isopropanol and ammonia (70:30 V/V) as the mobile phase. The
developed chromatograms were sprayed with 1% aqueous picric acid, dried for
1 hr at 110°C and then sprayed with 4% aqueous sodium hydroxide (Zaika gt_gl.,
1968).

Analyses for lactic acid were carried out by thin layer chromatography
on Silica Gel G and Cellulose MN300 plates. The Silica Gel G chromatogram
'was developed with diethyl ether and 90% formic acid (70:10, V/V), while the
Cellulose MN300 chromatogram was developed with formic acid, 2-butanone, tert.
butanol and water (l5:30:40:15, V/V). The chromatograms were sprayed with

0.3% bromphenol blue and 0.1% methyl red in 95% ethanol (Zaika g} 31. , 1968).
Chemical.Analysis of Dialyzable and NOn-Dialyzable CompOnents

Free sulfhydryl group determinations were conducted according to the
method of Ellman (1959) with modifications. DTNB [5,5'-dithiobis (2-nitro-
benzoic acid)] was reoxidized and crystallized, m.p. 239°C (237-238°C reported).

DTNB (39.6 mg) was.made to a volume of 10 ml with 90% ethanol.

55

Sulfhydryl determinations in study I were performed only on the water
soluble extract before and after heating. The lyophilized water soluble
fraction of adipose tissue (0.1 g) was diluted to 10 ml with phosphate
buffer (pH 8.0; u = 0.1). A 1 ml aliquot was placed in a 1 cm square Pyrex
cuvette. The aliquot was combined with 2.ml 8.5 M urea (recrystallized
from water and ethanol) and 0.02 ml DTNB solution. The cuvette was sealed
with Parafilm, and the contents were mixed by gently inverting the cuvette
several times until all schlieren lines were eliminated. The mixture was
then allowed to stand at room temperature for 45 min before absorbance was
measured. The sulfhydryl concentration was determined using a molar absorp-
tivity value of 13,600 at 412 nm. Absorbance was measured using a Beckman,
Model DU, spectrophotometer equipped with a Gilford (Model 220) absorbance
indicator. After heating, the protein-water slurry was lyophilized and the
free sulfhydryl groups were determined in the same manner. The salt-soluble
and insoluble fractions were not soluble in the buffer-urea mixture, so that
their sulfhydryl content could not be determined.

In study II, free sulfhydryl groups were determined as described above,
except that the non-dialyzable fraction (1 ml) was an aliquot of the sucrose—
protein concentrate, and the urea was crystallized from water and ethanol
and then passed over an Amberlite MB-3 mixed bed resin.

To determine the sulfhydryl content of the diffusate, isolated in study
II, 1 ml of urea was added to 2mm1 of sample. The determination was then
carried out as previously described herein.

The aldosaccharide content of the diffusate was determined according

"
to the method of Muller (1965). The reagent is composed of o-toluidine

56

(6%, V/V) and thiourea (0.15%, w/v) in glacial acetic acid. The sample
(0.5 ml) was.mixed with the reagent (3.5 ml), heated in a boiling water
bath for exactly 8.0 min and cooled in running tap water. Absorbance was
measured at 630 nm. The color developed was stable for 1 hr. The addi-
tion of trichloroacetic acid (TCA) and centrifugation were omitted from
the procedure, since the protein had already been removed by dialysis.

Free amino groups in the diffusate were determined by the method of
Florkin and Stolz (1963). Samples (0.1 ml) containing 0.2-1.0 mg protein
per 0.1 ml solution were combined with ninhydrin reagent (lml), covered
with aluminum foil caps and mixed. The mixture was heated for 20 min in a
boiling water bath, cooled and.mixed rapidly with 5 m1 of 50% aqueous isopro-
panol. Absorbance was measured at 570 nm within l5.min. Serial dilutions
of leucine were used to construct the standard curve. The ninhydrin reagent
was composed of citrate buffer and a ninhydrin Solution. The citrate buffer
(pH 5.0) was.made by dissolving 21 g citric acid in 200 ml 1 N NaOH and
diluting the mixture to 500 ml with deionized distilled water. The ninhydrin
solution was prepared by dissolving 20 g ninhydrin in 500 ml redistilled
methyl cellosolve. The ninhydrin reagent was produced by combining the
citrate buffer and ninhydrin solution (1:1 V/V).

Sulfate was detected in the diffusates by combining approximately 0.30
ml of the concentrated diffusates with approximately 10.m1 of 10% barium
chloride in l N HCl. The mixture was boiled for approximately 15 sec and
allowed to cool. The formation of a white precipitate indicates the presence

of the sulfate ion in the diffusate.

57

Development of a Method to Determine Sulfhydryl and Disulfide Groupg

After reviewing several sulfhydryl analytical techniques, a method
described by Saville (1958) and Liddell and Saville (1959) appeared to be
particularly adaptable for the determination of sulfhydryl and disulfide
groups. They determined low molecular weight thiols (cysteine and gluta-
thione) indirectly, employing the formation of an azo dye.

Several solutions were employed by Liddell and Saville (1959) for the
thiol determinations. The thiol concentrations ranged from 1 to 6 x 10-4 M.
Solution A was composed of a 0.01 M sodium nitrite solution (5.0 ml) in 1 N
H2S04 (35 ml) and water to a final volume of 100 ml. Solution B was 0.5%
aqueous ammonium sulfamate solution. Solution C was composed of a 4:1 (V/V)
mixture of 3.44% sulfanilamide in 0.4 N HCl and a 1.0% mercuric chloride
solution in 0.4 N H01, respectively. Solution D was produced by adding 1.0
g N-l-naphthylethylenediamine dihydrochloride to 1.0 liter of 0.4 N HCl.

Liddell and Saville (1959) allowed the thiol (1.0 ml) to react with
5.0 ml of solution A for 4-5.min. This solution was in turn allowed to
react with 1.0 ml of solution B for l-2.min. Solution C (10 ml) was rapidly
added to the reactants, and after a 2 min period the solution was diluted
to 25 ml with solution D. The colored solution was read at 540 nm after 2
.min.

The first modification of the Liddell and Saville (1959) method involved
reduction of the total volume of the various reagents and investigation of

the applicability of the method for disulfide analysis. Solutions A, B, C

and D were the same concentrations as described previously. The reducing

58

agent was 5% potassium cyanide in 1 N ammonium hydroxide. The source of
the thiol groups was DTNB. This disulfide was satisfactorily reduced by
the cyanide reducing agent to produce two thiols per disulfide as found by
determining the thiol concentration at 412 nm.

DTNB (0.5 ml) was combined with the reducing agent (0.1 ml) and allowed
to react for 45.min. Solution A (0.5 ml) was added to the reduced thiol
solution. After 2 min, 0.2 ml solution B was added to the mixture, and 5
.min was allowed for the destruction of excess nitrous acid. Solution C
(1 ml) was combined and mixed with the S-nitroso solution. Solution D (1 ml)
was added 2.min later, and 2 min was allowed for color development. The re-
sulting dye was read at 540 nm.

Catsimpoolas and Hood (1966) stated that both cyanide and sulfite re-
duction incompletely reduced disulfides in some proteins. Hamm and Hofmann
(1965) used 0.6 M sodium borohydride in 8.0 M urea to reduce protein disul-
fides. However, upon substituting the sodium borohydride reducing agent
for the cyanide solution in the present study, unsatisfactory results were
obtained. The borohydride solution generated hydrogen gas, which.made
accurate pipetting impossible. After decreasing the borohydride concentra—
tion to 0.1 M in 8 M urea and adding 0.5 ml of the reducing agent to the
disulfide solution, results were still unsatisfactory. Since urea destroys
nitrous acid, a 0.1 M aqueous borohydride solution was used. However, the
sodium borohydride was less stable in water than in urea, still giving un-
satisfactory results.

Several compounds were used to destroy the excess borohydride (ascorbic

acid, acetone, dihydroxyacetone, uric acid, H01). In order to destroy the

59

borohydride, either H01 had to be added to bring the pH to 2.0 or else high
concentrations of the other reagents were required. The concentrations were
determined by adding the reagents to the borohydride and then adding a solu-
tion of DTNB. The development of a yellow color indicated incomplete
borohydride destruction.

To study the effect of various reagent concentrations, three of the
four reagents (A, B, C and D) were held at 2 x 10'3 N except the reagent
under investigation. However, ammonium sulfamate was held at 4 x 10.3 N in
all cases except when its concentration was under investigation. This insured
complete destruction of excess nitrous acid, regardless of the concentrations
of nitrous acid or the other reagents. The investigation confirmed the
findings of Liddell and Saville (1959) that the only reagent detrimental to
the reaction was a large excess of the mercuric salt. However, much lower
concentrations of mercuric chloride than recommended were found just as
effective.

In order to simplify the method and thus insure greater repeatability,
the reagent concentrations were modified so that 0.5 ml of each reagent
could be added to the cuvette with 0.5 ml pipets (TD). This meant that 0.5
ml aliquots of the thiol, reducing agent and reagents A, B, C and D could
be added to the approximately 3.5 m1 cuvettes, leaving sufficient space for
adequate.mixing.

4

Kobayashi (1966) reported a molar absorptivity of 3.93 x 10 for the

azo dye, p-E4-[(2- aminoethyl) -amino] azo; benzenesulfonamide. Sawicki e_t
gt. (1963) found that the molar absorptivity of the azo dye was approximately

4.0 x 104. In the previous two reports, the method of determining the molar

60

absorptivity was not elucidated. Liddell and Saville (1959) did not report
the molar absorptivity of the dye, but they reported the absorbance of various
thiol concentrations and the light path length through the solution. These
values plus the volumes of the thiol and reagents, which gave the dilution
factor,.made determination of the molar absorptivity possible. The resulting
molar absorptivity was approximately 3.84 x 104. However, this is not the
true molar absorptivity of the dye, since this value is dependent upon the
yield of all the various reactions.

By using the modified Liddell and Saville (1959) method, the thiol con-
tent, absorbance, light path length and dilution factor were determined.
The calculated molar absorptivity was approximately 4.20 x 104. Therefore,
it was necessary to determine the actual molar absorptivity of the dye.

Approximately 1.0 g of the dye was synthesized by reacting sufficient
sodium nitrite in 1.0 N HCl with excess sulfanilamide in 0.4 N H01, which
gave the diazonium salt. The resulting diazonium salt solution was then
reacted with excess N-l-naphthylethylenediamine dihydrdchloride in 0.4 N H01
to give the azo dye. Upon neutralization of the reddish-violet dye solution
with sodium bicarbonate, the solution turned orange and the neutralized dye
formed a precipitate. The precipitated dye was quite soluble in ethyl acetate
but much less soluble in ethyl ether. It was decided that a much higher
recovery of the dye could be achieved by ether extraction. Therefore, the
aqueous solution containing the precipitated dye was subjected to liquid-
liquid extraction with ethyl ether over a four day period. Golden cyrstals
of the dye were isolated in the ether solution and were recovered upon

removal of the ether under reduced pressure in a Rinco rotating evaporator.

61

After storing the dye in an evacuated dessicator overnight, the molar absorp-
tivity was determined by solubilizing a known amount of dye to a final known
volume of 0.4 N H01. The volume of H01 added was adjusted to give an absor-
bance of approximately 0.300 at 540 nm. The 540 nm wavelength was determined
by noting the wavelength at which the highest absorabance of the dye (acid
form) occurred using a Spectronic 505 recording spectrophotometer. The molar
absorptivity of the dye was calculated to be approximately 4.93 x 104.

Theoretically, the coupling reaction of the diazonium salt and N-l-
naphthylethylenediamine could result in three products, since coupling
could take place in the ortho-, para- or both positions. Since the second-
ary free amino group on the coupling agent is in the a-position and is a
stronger para- than ortho-directing group, the majority of the coupling
reactions should occur in the para-position. However, all these theoretical
assumptions appear to be pre-empted by the fact that coupling occurs in an
acid medium. Since coupling occurs in an acid medium, equilibrium.must
exist between the free amine and amine hydrochloride.

The question as to why coupling occurs was resolved by titrating a
solution of N-l-naphthylethylenediamine (pH 1.15) with sodium hydroxide and
plotting the amount of sodium hydroxide used against the pH of the solution.
The pKa of the aromatic secondary amine group was found to be approximately
3.1. Thus, coupling could readily take place in the acidic medium since
approximately 50% of the free amine form is available for coupling at this
pH. Coupling could occur fairly rapidly since the reaction would be driven
to the right.

In order to determine if coupling had occurred in any position other

than the para postion, the neutralized dye was subjected to thin layer

62

chromatography employing a modification of the method of Ganshirt gt gt.
(1965). Alusil (Silica Gel G:A1umina G, 1:1) was the stationary phase.
Three mobile phases were used in the separation of the dye components. The
first mobile phase consisted of ethyl acetate, methanol and 5 N ammonium
hydroxide (60:30:10). The Second system consisted of ethyl acetate, tetra-
hydrofuran, methanol and 5 N ammonium hydroxide (60:20:10:10), while the
third phase was a combination of ethyl acetate, tetrahydrofuran and 5 N
ammonium hydroxide (60:30:10). All three solvent systems separated the dye
into three components, one major and two minor ones. All three components
had different Rf values in the three different solvent systems. The most
polar solvent system gave the highest Rf values for all three components.

After the chromatography was completed, the plates were dried at
approximately 55°C and the orange spots were sprayed with 0.4 N H01. All
three components wereconverted to reddish-violet azo dyes. The minor com-
ponents were not subjected to spectrophotometric analysis to determine their
molar absorptivities.

To prevent the formation of variously coupled dyes, a suitable coupling
agent was sought. Since none was available or reported in the literature,
the coupling agents had to be synthesized. er-acenaphthylethylenediamine
was chosen as a suitable coupling agent. Although the para position is
occupied, coupling should take place readily at the 4-position, i.e. ortho
to the amine group (Zollinger, 1961).

Cava gt gt. (1965) reported the synthesis of 5-aminoacenaphthene via
the reduction of 5-nitroacenaphthene. 5-nitroacenaphthene was formed by
adding 125 ml of 70% aqueous nitric acid over a 15 min period to a mixture

of 100 g acenaphthene in 800 ml of glacial acetic acid. After the addition

63

of nitric acid was completed, the.mixture was cooled to approximately 20°C
and added to 2 liter of distilled water. The precipitate was filtered,
employing a Bachner filter fitted with Ihatman No. 1 filter paper, and was
dried at 50°C. The yellow precipitate was cyrstallized from either methanol
or petroleum ether, m.p. 103°C. Cava gt gt. (1965) reported a m.p. of 104°C
for 5-nitroacenaphthene.

5-aminoacenaphthene was synthesized according to the method of Cava
21.2l' (1965) by combining 50 g of 5-nitroacenaphthene, 10% Pd-C (2.5 g),
60 ml of 85% hydrazine hydrate and 1.0 liter of 95% ethanol. The.mixture
was refluxed for 45.min and the hot solution was filtered through Celite.
Distilled water was added to the filtrate until the precipitation of 5-amino-
acenaphthene was complete. The precipitate was recovered by filtration and
crystallized from.petroleum ether. In the present study, a.m.p. of 105°C
was found for the white cyrstals of 5-aminoacenaphthene, which compares to
the m.p. of 104°C reported by Cava gt g. (1965).

Nel-acenaphthylethylenediamine was prepared by a modification of the
‘ methods of Newman (1891), Ing and Manske (1926) and Bratton and Marshall
(1939). 5-aminoacenaphthene was fused with a slight excess of B-bromoethyl-
phthalimide at approximately 100°C (heating to higher temperatures caused
oharring of the reactants). The liquid was washed with several volumes of
hot.(fi/90°C) distilled water. The B-(5-aminoacenaphthyl) ethylphthalimide
was then refluxed for 2 hr with 50 ml of 85% hydrazine hydrate in 200 ml of
95% ethanol. Excess H01 was added to the mixture and refluxing was continued
for an additional hour. Phthalylhydrazide was removed by filtration and

the filtrate was concentrated to 50% its original volume at reduced pressure

64

employing a Rinco rotating evaporator. The residual phthalylhydrazide was
removed by filtration and 100 m1 of distilled water was added to the filtrate.
The ethanol was removed at reduced pressure and the solution was neutralized
with aqueous potassium hydroxide. The free amine compound was extracted

with ethyl ether, dried over anhydrous sodium sulfate and, after removal of
the sodium sulfate, was precipitated from the ether solution by the addi-
tion of concentrated H01. The hydrochloride was dissolved in a minimum of
methanol and precipitated with ethyl ether. A white precipitate was formed
that rapidly changed to red.

Methanol, containing a trace of H01, was added to the ether solution
and the precipitate was redissolved upon removal of the ether. Activated
charcoal was added to the methanolic solution but failed to decolorize the
precipitate. The precipitate was added to an acidic solution of p-sulfamoyl-
benzenediazonium salt, and an intensely purple solution resulted. This
indicated that the precipitate possessed the properties of a coupling agent.
Further studies to determdne the usefulness of this compound as a coupling
agent will be carried out in the future.

Presently, the method has been utilized only for low molecular weight
disulfides. However, studies have shown that use of a 1% aqueous solution
of sodium lauryl sulfate as a protein denaturant in place of urea produced
no undesirable side reactions. But the Nel-naphthylethylenediamine solution
(D) must be composed of 0.1950 g of Mel-naphthylethylenediamine dihydro-
chloride in 10 ml of 0.4 N H01 and.made to a final volume of 25 ml with
acetone. The acetone not only destroys the foam resulting from the agita-
tion of the sodium lauryl sulfate solution but also introduces a bathochromic

effect, which changes the maximum absorption from 540 to 550 nm.

65

Since alkaline cyanide, sulfite and sodium.borohydride solutions are
not suitable reducing agents, aqueous sodium.trimethoxy borohydride and
tetramethylammonium borohydride are presently being investigated. Neither
reducing agent produces profuse hydrogen gas generation when combined with
water, but sodium trimethoxy borohydride appears to react very slowly. There-
fOre, tetramethylammonium borohydride appears to be the most desirable
reducing agent. However, tetramethylammonium borohydride should be almost
as alkaline as sodium borohydride, so it.may be necessary to decrease the
pH of the reducing solution to 9.0 to prevent any possibility of alkaline
scission of the disulfide bond.

Nitrous acid is used for two purposes in the method. First, nitrous
acid is used to destroy the excess redicing agent. Secondly, the residual
nitrous acid is employed in the formation of the S-nitroso derivatives of
the sulfhydryl groups. The amount of nitrous acid added must be regulated
by the amount of reducing agent to be destroyed, by the amount of thiols to
be nitrosated and by the amount of ammonium sulfamate added. Since the
ammonium sulfamate destroys the excess nitrous acid after nitrosation has
occurred, the nitrous acid content must be held at a lower concentration
than that of the ammonium sulfamate. It is not advisable to increase the
concentration of ammonium sulfamate, since the excess sulfamate would.more
favorably compete with the sulfanilamide for the cleaved nitrous acid.

Thus, the sensitivity of the determination would be decreased.

RESULTS AND DISCUSSION

Adipose Tissue Analysis

In study I, the moisture, fat and protein content of raw beef adipose
tissue averaged 15.36, 79.20 and 4.84%, respectively. The corresponding
values for the raw adipose tissue utilized in study II were 10.35, 86.07

and 3.27%.
Hydrogen Sulfide Determinations

The appearance of a white zinc sulfide precipitate in the collection
traps indicated that hydrogen sulfide was evolved upon heating the whole
adipose tissue, and by the water soluble, salt soluble and insoluble frac-
tions of adipose tissue. The precipitate was first noted when the heated
slurries reached a temperature of approximately 80°C. The temperature at
which hydrogen sulfide commenced to evolve is in agreement with that reported
by Ham and Hofmann (1965), who stated that hydrogen sulfide was initially
evolved from heated beef myofibrils at 80°C.

The amount of hydrogen sulfide evolved upon cooking the whole adipose
tissue and its various fractions for a 7 hr period is shown in Table l. The
hydrogen sulfide evolution rate upon cooking whole adipose tissue as compared
to the water soluble fraction is shown in Figure 3. The amount of hydrogen
sulfide obtained from the salt soluble and insoluble fractions could not be
plotted against time, because the total amounts obtained from each fraction

over a 1 hr period were too small to be accurately measured. Therefore, the

66

67

hydrogen sulfide generated from.a single fraction was collected in only one
trap over the 7 hr period in order to sufficiently increase the total amount
and allow accurate measurement.

Table 1. Hydrogen sulfide evolved upon cooking adipose tissue and its
various fractions for 7 hr.

 

 

 

Protein in ‘HzS/lOO g HzS/g

adipose tissue adipose tissue protein
Source gleO g pH pH
Adipose tissue (whole) 4.84 42.74 8.83
Water soluble fraction 0.43 17.63 41.01
Salt soluble fraction 1.18 1.71 1.45
Insoluble fraction1 3.23 5.46 1.69

 

lInsoluble in water, 0.45 M NaCl and ethyl ether.

The sulfhydryl content of the water soluble fraction before cooking
was 9.66 uM'(micro moles) of sulfhydryl per g of protein, but declined to
0.85 uM of sulfhydryl per g protein after cooking. The salt soluble and
insoluble fractions were not soluble in the buffer-urea mixture, so sulf-
hydryl analyses could not be made.

The rate at which hydrogen sulfide was evolved from the heated water
soluble fraction rapidly reached a maximum and then decreased. The amount
of hydrogen sulfide evolved was approximately 5.43 uM per g of protein
during the first hr of cooking, and was approximately 6.12 pH per g of
protein during the second hr. More hydrogen sulfide was collected during
the second hr of cooking than during the first. However, it should be borne

in mind that the first hr of cooking started upon connecting the heating

Figure 3.

68

 

M

8 - Whale Adipose hssue

i \_—Wam Soluble hauiun

 

 

l 2 3 I 5.8 7
InmelHust

 

Amount of hydrogen sulfide evolved with increased cooking time.
Values are for 150 g of whole beef adipose tissue and 0.64 g of
lyophilized water soluble fraction. The latter figure corres-
ponds to the amount of water soluble substance in 150 g of whole
adipose tissue.

69

mantle. The time required to heat the solution to 80°C was approximately
20.min. Therefore, the actual hydrogen sulfide evolution time during

the first hr of cooking was approximately 40 min. If the evolution time
(40 min) is extrapolated to 60 min, the hydrogen sulfide evolved would

be approximately 8.15 uM of hydrogen sulfide per g of protein. Since a
total of approximately 6.12 uM of hydrogen sulfide was collected during
the second hr of cooking, the rate of hydrogen sulfide evolution would

be greater during the first hr than during the second.

Figure 3 shows a decrease in hydrogen sulfide evolution after the
second hr. During the first 2 hr of collection, the amount of hydrogen
sulfide obtained was 11.55 uM. ‘Using the extrapolated value of 8.15 pH
of hydrogen sulfide during the first hr, it seems likely that hydrogen
sulfide began to decline during the second hr. The reason for assuming
that free sulfhydryls are responsible for the initial evolution of hydro-
gen sulfide is the decline in free sulfhydryls from 9.66 uM per g protein
before cooking to 0.85 pH per g protein after cooking.

Hydrogen sulfide was evolved at approximately the same rate during
the first hr of cooking from.hoth the heated whole adipose tissue and
an equivalent amount of water soluble protein (the quantity of water
soluble protein in the same amount of whole adipose tissue). After the
first 2 hr of heating, the amount of hydrogen sulfide from.the whole
adipose tissue continued to be evolved at almost a constant rate, whereas,
the hydrogen sulfide from the water soluble fraction steadily declined.
This is probably due to the limited exposure of the free sulfhydryl groups

entrapped in the fat and tissue matrices. The free sulfhydryl groups of

70

the freeze-dried, water soluble fraction were apparently much more
accessible, and therefore, more susceptible to degradation during heating.
Using chicken as substrate, Mecchi gt.gt. (1964) reported that the
water insoluble fraction was responsible for twice as much hydrogen sul-
fide as the water soluble fraction. Hamm and Hofmann (1965) found that
almost all of the hydrogen sulfide released upon cooking whole beef muscle
was derived from the structural proteins and not from.the water.solub1e
substances. With adipose tissue, the reverse was found to be true. 0f
the protein fractions isolated from adipose tissue, the water soluble
fraction evolved 71% of the hydrogen sulfide, while the salt soluble and
insoluble fractions yielded only 7 and 22%, respectively. Therefore, the
water soluble fraction contributed nearly 2.5 times as much hydrogen,

sulfide as the combined water insoluble fractions.

Analysis of Dialyzable and Non-Diatyzablg Components offithe Water Soluble

Fraction of Adipose Tissue

There.are four reasons for studying the water soluble fraction of
adipose tissue. First, the components responsible for meat flavor are
located in the water soluble portion of muscle (Kramlich and Pearson,-
1958; Hornstein and Crowe, 1964). Secondly, fatty tissues have been
reported to be responsible for differences in the flavor of meat from
various species (Hornstein and Crowe, 1960b; Uasserman and Talley, 1968).
Thirdly, it has been reported that hydrogen sulfide plays a major role
in the meaty aroma of chicken (Minor gt gt., 1965b). Fourth, the present
study indicates that the water soluble fraction of adipose tissue is pri-

marily responsible for hydrogen sulfide production upon heating.

71

The adipose tissue was stored under air and nitrogen to determine
if these atmospheres would affect the sulfhydryl content of the fatty
tissue. The adipose tissue was.mixed with solid sodium chloride to give
a 3% salt concentration and stored under air and nitrogen. A 3% salt
concentration was selected, since most sausages contain approximately

this level of salt.

.mmino acid analysis. The percent of the amino acids present in the

 

dialyzable and non-dialyzable water soluble fractions of the various

treatments (02 - air atmosphere; N - nitrogen atmosphere; 02S - air

2

atmosphere and 3% salt; N S - nitrogen atmosphere and 3% salt) are given

2
in Table 2. Since only trace amounts of free sulfhydryl groups were
determined in the diffusates of all treatments, the diffusates from
treatments N2 and 02 were combined, as well as those from N23 and 028
and designated as I and II, respectively.

There was no major difference in the amino acid composition of the
proteins isolated from the various treatments, except for the high glu-
tamic acid and low threonine and serine contents in the N28 treatment.
These values cannot be explained.

Alanine, glycine and glutamic acid composed over 50% of the amino
acids found in the diffusates, whereas, the same amino acids accounted
for only approximately 25% of the amino acid residues in the protein
fractions (Table 2). The half cystine content of the dialysates was

approximately three times that found in the diffusates. Under acid hydro-

lysis, some cystine sulfur appears in the humin of the hydrolyzate (Leach,

72

Table 2. The amino acid composition of dialyzable and non-dialyzable
aqueous fractions of the various treatments.

 

Percent amino acids

 

1 2 3 4 5 6

N2 02 028 N28 I II
Lysine 9.10 9.00 9.24 9.52 1.77 2.09
Histidine 2.80 2.67 2.84 3.41 3.22 3.96
Arginine 3.85 3.77 4.06 4.10 0.46 1.22
Aspartic acid 10.43 10.48 10.56 10.96 7.35 6.28
Threonine 5.99 5.92 6.11 1.80 1.17 3.33
Serine 6.24 5.75 5.86 0.43 1.30 4.49
Glutamic acid 12.54 12.79 12.72 16.55 21.80 13.46
Proline 5.12 5.07 5.11 6.27 6.83 4.93
Glycine 5.64 5.43 5.80 6.32 17.50 12.63
Alanine 8.44 8.06 8.04 8.80 28.21 27.78
Half cystine 2.90 3.23 2.80 2.28 0.69 0.72
Valine 6.83 6.93 6.88 7.83 3.28 4.99
Methionine 0.25 1.07 0.66 0.90 0.67 0.93
Isoleucine 3.27 3.29 3.34 3.36 1.37 2.49
Leucine 9.82 9.87 9.58 10.70 2.66 6.15
Tyrosine 2.55 2.53 2.42 2.21 0.48 1.75
Phenylalanine 4.24 4.12 3.99 4.54 1.23 2.80

 

1Adipose tissue stored under nitrogen.

Adipose tissue stored under air.

3Adipose tissue with 3% NaCl stored under air.

4Adipose tissue with 3% NaCl stored under nitrogen.

5Combined diffusates from N2 and 02 treatments.

6Combined diffusatesfrom.N25 and 028 treatments.

1966). Therefore, the percent of cystine found in the hydrolyzed solu-
tion is not an absolute but a relative percent, since all hydrolyzed

substrates were subjected to identical treatments.
Starch Gel Electrophoresis of Non-Dialyzable Components

The starch gel electrophoretic patterns obtained from.the water

soluble components of beef muscle and adipose tissue are shown in Figure

73

4. Patterns 1, 3, 4 and 6 were obtained from adipose tissue extracts
(02), while 2 and 5 are patterns of the beef muscle extract. The patterns
for the two substrates are not comparable, since the adipose tissue
extracts were not satisfactorily separated even though good separation
was achieved on the muscle extracts. These results were not surprising,
however, since proteins from.different substrates may not be completely
resolved by one electrophoretic system. Since starch gel electrophoresis
resulted in poor resolution of the water soluble components from adipose

tissue, the method was abandoned.
Disc Gel Electrophoresis of Non-Diatyzable Comppnents

Both disc gel electrophoretic methods (urea gels and aqueous gels)
described by Rampton (1969) for the separation of rabbit muscle contrac-
tile proteins did not adequately separate the water soluble proteins
of adipose tissue. However, a gel system.described by Jolley gt gt. (1967)
gave good protein resolution for both the muscle and adipose tissue ex-
tracts.

Figure 5 compares the proteins in both muscle and adipose tissue (02)
extracts, and also shows the sulfhydryl distribution for the two sub-
strates. The sulfhydryl distribution is more closely related to the
protein distribution in the muscle extract than in the adipose tissue
extract. Both the protein and sulfhydryl patterns, however, show the
dissimilarity between the two substrates. At 3.7 cm.a very broad protein
band (band A) was present in the muscle extract, while it was absent in

the adipose tissue substrate.

74

 

 

5 _—_ -_ ,1 .1 _
Figure 4. Comparison of the water soluble components of beef muscle and

Figure 5.

adipose tissue. (1-r, 1-6) 1, 3, 4 and 6 beef adipose tissue
extract; 2 and 5 beef muscle extract.

      

9
I:
b
2

c1“
“

 

 

7, III&$»«- fig .
CEEparlson between protein and sulfhydryl distribution in muscle
and adipose tissue extracts. (l-r. pattern l—adipose tissue
protein; 2-muscle protein; 3-muscle SH; 4-adipose tissue SH.)

75

When the gel from.the muscle extract was stained with Benzidine
reagent (Figure 6), band A (3.0 cm) stained very strongly. One other
band (band B) at 2.5 cm was also stained but less intensely. This stain
is specific for heme proteins. Scopes (1968) showed that rabbit myoglo-
bin.migrates faster than hemoglobin. Therefore:‘band A appears to be
myoglobin. The slower migrating smaller band B is probably hemoglobin.
When the adipose tissue gel was stained with benzidine reagent, a very
lightly stained band was observed, which had the same migration rate as
band B from the muscle extract. The staining and.migration characteris-
tics of this band indicate that it was probably hemoglobin. This is not
unreasonable, since at least a small quantity of residual blood was pro-
bably present in the adipose tissue.

Gels 1, 2 and 4 in Figure 6 show positive tests for ribonucleic
acids, nucleoproteins and glycoproteins, respectively. Gel 3 (Figure 6)
indicates that esterase activity is also present in the same general area
of the gel. To determine whether glycoproteins and nucleoproteins are
present in this region, the running gel was made of 5% Cyanogum. The
resulting gel patterns are shown in Figure 7. All proteins migrated at
a faster rate in the 5% running gel than in the 7.5% running gel. The
esterase migrated from 1 mm in the 7.5% gel to 4 mm in the 5% gel, however,
the substances that stained positive for glycoproteins and ribonucleic
acids did not migrate further with the 5% gel. Thus, these substances
appear to be ribonucleic acids. They stained extremely lightly with Amido
Black, so they contain practically no protein characteristics. However,
they stained positively with methylene blue, which is specific for ribo-

nucleic acids.

76

.Illll
;::

i
V from
1 » .
i

 

 

\u‘-

_ - ae-- *W—-

Figure 6. Staining of various groups from.adipose and muscle tissue ex-
tracts. [l-r. adipose tissue (02) extract patterns, 1-ribonucleic
acids; 2-nucleoproteins;'3-esterase; 4—glycoprotein; 6-protein;
5-heme proteins in muScle extract.)

 

 

 

:
u
i '-> a“
1

_ _~e_ _‘. "*h; .. .—

 

Figure 7. Gel patterns of the water soluble fraction of adipose tissue
employing 5 and 7.5% running gels. [l-r. Adipose tissue (02)
extract patterns, l-ribonucleic acids; 2-glycoproteins; 4-
esterase activity; 5-proteins; 3-esterase activity in a 7.5%
running ge1.]

77

Although the specific esterase was not identified, its activity
demonstrated that the extraction and concentration procedures used on
the tissue were relatively.mi1d. Therefore, the results are believed
to represent the water soluble proteins indigenous to adipose tissue,
and are probably not breakdown products.

Figure 8 illustrates the protein and sulfhydryl distribution in
all four treatments (N2, 02, N28 and 028) using 7.5% running gels.
Differences could not be ascertained between the protein and sulfhydryl
distribution in either the N28 and 028 treatments or between the N2 and

02 treatments. Therefore, neither nitrogen nor air appeared to have any
>affect on the proteins or sulfhydryl groups in the water soluble fractions
of adipose tissue. However, upon comparing the protein and sulfhydryl
patterns between the salt and salt-free treatments, definite differences
are apparent. The protein and sulfhydryl bands at 2.1 cm in Figure 8 are
much more intense in the salt treatments than in the salt-free treatments.
Conversely, the protein and sulfhydryl bands at 2.6 cm are more distinct
in the salt-free treatments than in the salt treatments (Figure 8). There
are also more distinct bands at 1.2 cm in the gels of the salt-free treat-
ment than in the salt treatments (Figure 8). The band at 5.8 cm appears
to be more intensely stained in the protein gels of the salt treatment
than in the corresponding gels of the salt-free treatment. However, when
the band at 5.8 om‘was stained for sulfhydryl groups, the results were
negative for all treatments (Figure 8).

Both.Amido Black 10B and Coomassie Blue were used to stain the gels

for proteins. The protein bands stained with Coomassie Blue became

Figure 8.

78

r
l

S

M?

!

NH

81

71

HHHWHHH

L
H

61

«vr'v:
ILIwHTH

 

 

 

Gel patterns of the water soluble fractions of adipose tissue
subjected to N2, 02, NZS and 02S treatments. (l-r. Adipose
tissue extract patterns, l-NZS protein; 2-N2S SH; 3-02S protein;
4-028 SH; 5-02 SH; 6-02 protein; 7-N2 SH; 8-N2 protein.)

 

79

discernible in a much shorter time than those stained with Amido Black, but
they were not as intensely stained. However, approximately the same de-
staining time was required to obtain distinct bands for both stains. Since
both stains gave identical protein patterns and Amido Black gave a more
intense color, the latter stain was used in all subsequent studies.

Lipoproteins appeared to be absent in the water soluble fraction of
adipose tissue in all treatments. No bands could be detected when the
gels were stained with either Sudan Black B or Oil Red 0.

The disc gel electrophoretic patterns from the water soluble protein
fraction of adipose tissue employing a 5-10% gel gradient were not as
satisfactory as those obtained from the 7.5% gels. Although many bands
were detected, the proteins failed to.migrate into the more concentrated
portions of the gel, whereas, the protein bands were dispersed throughout
the 7.5% gel. Therefore, the 7.5% gel gave more numerous and better

separated bands.

Column Chromatogpgphic Separation of NOn-Dialyzable and Dialyzable Components

 

Separation of the water soluble proteins of adipose tissue could not
be accomplished employing any of the gel filtration resins. The proteins
were eluted in a single peak using columns packed with Bio-Gel P-30, Bio-
Gel PL100, Bio-Gel A-l.5 m, Bio-Gel A-15 m and Bio-Gel A-50 m.

Several peaks were observed when Bio-Gel P-Z resin was used to separate
the diffusate of the water soluble components from adipose tissue. However,
the peaks were not separated sufficiently to give adequate resolution of

the components.

80

The formate form of Dowex l anion exchange resin yielded satisfactory
separation of the diffusate. Diffusate I (N; plus 02) was separated into
five peaks when the gradient system described by Lento gt gt. (1964) was
used. However, a discontinuous gradient system composed of water, 0.5 N
formic acid, 1 N formic acid, 3 N formic acid and 6 N formic acid divided
diffusate I into eight peaks (Figure 9). Therefore, this elution system
was employed to separate diffusate II (N28 plus 028). Diffusate II was
also separated into eight peaks, which were eluted at approximately the
same elution time using the same system. The principle difference was the
increased absorbance of the peaks, which indicated that diffusate II was
more concentrated. This was expected because of the different method of
concentration.

Diffusate I was concentrated in a flash evaporator at 40°C. However,
fine white strands of material appeared in the solution, so rotation was
stopped and concentration was completed under reduced pressure at 40°C.
When a volume of approximately 100 ml was reached, appreciable amounts of
precipitate were present, which was removed by filtration. It is believed
that agitation and heat may have exerted adverse affects on the diffusate
components. Since a RePP Sublimator became available for remaining studies,
diffusate II was concentrated by lyophilization, which is a much.milder
method of concentration. Upon solubilizing diffusate II in a minimum
amount of water, a very light yellow solution was obtained. Whereas, the

concentrate from diffusate I was a much darker yellow.

LO“

Time (Hm)

0.0. (254 mu.)

 

 

 

 

 

 

 

 

 

 

 

-

Figure 9. Column chromatographic pattern of diffusate I components.
[Elution systems, water (oez hr), 0.5 N formic acid (2-5 hr),.
1 N formic acid (5-12.5 hr), 3 N formic acid (12.5-17 hr),
6 N formic acid (17-22 hr)].

 

 

L—n—h—M—fl

Figure 10. Thin layer chromatographic patterns of diffusate II and
standard samples of creatine phosphate, creatine and creatinine.
(l- r, 1-4) l-creatine, 2- diffusate II, 3-creatine and creati-
nine, 4-creatine phosphate, creatine and creatinine.

 

82

Thin Layer Chromatogpgphic Analysis of Dialyzable Components

Preliminary investigations showed that peaks 1-3 contained creatine
and creatine derivatives in both diffusates I and II. Peaks 1-6 of both
diffusates contained ninhydrin positive components. Peak 2 gave the most
intense ninhydrin reaction, while peak 6 gave the least intense. The
amino acid and sulfhydryl contents of both diffusates were also similar.
Since both diffusates demonstrated these similarities, detailed analyses
were conducted only on diffusate II because of its greater concentration.

Creatine phosphate, creatine and creatinine were identified by come
paring the Rf values of the unknown to standard compounds, and by the orange
color developed on spraying both the unknowns and standards with picric acid
and sodium hydroxide solutions. The Rf values for standard samples of
(creatine phosphate, creatine and creatinine were 0.08, 0.23 and 0.56,
respectively (Figure 10). The Rf values for the unknown orange spots were
0.09, 0.24 and 0.56, thus, establishing the presence of creatine phosphate,
creatine and creatinine, respectively, in the adipose tissue diffusate.
Although peaks 1-3 contained creatine and its derivatives, the heaviest
concentration of these components was found in peak 2.

Lactic acid was not detected by employing either the Silica Gel G or
the Cellulose MN 300 plates with their respective solvent systems. Carbonyl
compounds could not be detected when the chromatograms were sprayed with
2,4—dinitrophenylhydrazine solution. When aliquots of all the peaks were
spotted on Silica Gel plates and sprayed for phosphate ,. all peaks except 4

and 5 gave positive reactions.

83

Peaks 1 and 2 streaked extensively when cellulose plates were used.
Peak 3 possessed one u.v. absorbing spot which was also present in peak 4.
Peak 3 contained no components that fluoresced at 254 nm, while fluorescing
substances were just barely visible at 366 nm. Peaks 4-8 contained come
ponents that fluoresced at both 254 and 366 nm, however, peaks 7 and 8
demonstrated the most intense fluorescence. Upon developing the cellulose
plates with water, peaks 4-8 possessed yellow fluorescent material in the
region of the origin, whereas, the yellow fluorescent material of these
same peaks appeared at the solvent front when the cellulose plates were
developed with the methanol-HCl-water solvent system. .All fluorescing
components, except those that fluoresced yellow, demonstrated light blue
fluorescence at both wavelengths. Apparently, most of these fluorescent
compounds possessed rather strong electronegative group(s), since they were
only eluted with relatively concentrated acid solutions. On the other hand,
nucleotides, which possess a phosphate group, are eluted with 0.5 N formic
acid and 0.2 N sodium formate (Lento gt gt., 1964).

It also appears likely that all fractions collected after the elution
of peak 3 contained these fluorescent substances. Compounds absorbing u.v.
light are responsible for the peaks. If these fluorescent compounds are
continually eluted, there would be no noticeable baseline fluctuations.
These compounds could be non-enzymatic browning products, since Spark (1969)
reported that browning products fluoresce under u.v. light. Also, chemical
analyses revealed the presence of both aldoses and free amino groups, which
are necessary for non-enzymatic browning.

The browning reaction probably accounts for the dark yellow color of

the concentratefrom diffusate I. To concentrate this diffusate, the

84

solution was held at an elevated temperature (40°C) for a prolonged period.
These conditions would greatly enhance the browning reaction. However, the
concentration of diffusate II required no heating, except that required for
sublimation, thus, the browning reaction would probably be largely inhibited.
The lack of browning probably accounted for the light yellow appearance of
diffusate II. However, it is not possible to competely prevent browning
fromtoccurring when the solution contains aldoses and free amino acids.

Figure 11 shows the thin layer chromatogram.of various standard come
pounds and peaks 4, 5, 6, 7 and 8 upon development with water. Figure 12
is a chromatogram, which is identical to the chromatogram in Figure 11 ex-
cept that it was developed with a methanol-HCl-water solvent system. Table
3 gives the Rf values for the standard and unknown compounds in the above
chromatograms.

Since peaks 6, 7 and 8 contained components that were phosphate positive,
they were next subjected to thin layer chromatographic systems designed for
nucleotide separation. Cellulose MN300 plates (laboratory prepared) were
spotted with peak 6, 7 and 8 and developed with a tert.amyl alcohol-formic
acid-water solvent system.(Figure 13). Peak 6 was separated into three
components. Spot 1 (Rf = 0.33) fluoresced but was phosphate negative.
Spots 2 and 3 (Rf = 0.39 and 0.54, respectively) did not fluoresce but were
phosphate positive. Peaks 7 and 8 resulted in identical patterns with four
components being detected in each of the two peaks. Spots l, 2 and 3 (Rf =
0.35, 0.44 and 0.49, respectively) fluoresced but contained no phosphate.
Spot 4 (Rf = 0.57) appeared to absorb u.v. light very faintly and gave a

fairly intense positive phosphate reaction.

85

 

 

Table 3. Rf values for standard and unknown compounds in diffusate II.
Cellulose MN300 plates (Laboratory prepared).

Compound 12,-~1 sz Peak R,~1
Adenine 0.41 0.48 4 0.80, 0.75a, 0.64°, 0.51a, 0.03b
Adenosine 0.63 0.57 5 0.64a, 0.03b
Cytidine 0.77 0.65 6 0.60, 0.05b
Cytosine 0.70 0.62 7 0.95, 0.90“, 0.82a, 0.69a, 0.38a, 0.13b
Guanidine 0.00 streak 8 0.93, 0.31, 0.08b
Guanosine 0.64 0.47
Hypoxanthine 0.66 0.45 Peak sz
Inosine 0.84 0.59 4 0.99b, 0.81a, 0.73, 0.65, 0.62a
Thymidine 0.87 0.88 6 0.99”, 0.73, 0.65, 0.62a
Thymine 0.81 0.81 6 0.99b, 0.82, 0.36
Uracil 0.80 0.74 7 0.99b, 0.83, 0.68a, 0.61a, 0.53a, 0.43a' 0.35a
Uridine 0.88 0.76 8 0.99b, 0.84, 0.67a, 0.54a, 0.44?, 0.34a

 

 

ISolvent-water.

ZSolvent-0H30H, H01, 320 (65:17:18).
aIndicates light blue fluorescence (254 and 366 nm).
bIndicates yellow fluorescence (254 and 366 nm).

Figure 11.

Figure 12.

86

 

 

 

k ?I \;’.’.._l N . I
7* r-
. A...‘ M;
_ A
yo. ~\
53 l) ; f
‘g 1.1.. .
,’. 17 ‘ 7
—-~-— 1, .‘ . *., . . . \. Y . . an; 3?. . .4. - .
49442524162272.118u1

 

Thin layer chromatograms of standard compounds and diffusate

II components. Cellulose MN300 plate (laboratory prepared);
Solvent-water. t-inosine, 2- hypoxanthine, 3-cytosine, g-
cytidine, 5-adenine, 6-uridlne, 7-uracil, 8- thymine, 9-thymidine,
lO-adenine,ll-adenoslne,12-guan03ine,lS-guanidine, 4-peak 4,
5-peak 5, 6-peak 6, 7-peak 7, 8-peak 8.

's" 1';

1 .. ,
L" .3,‘
r | -'
t. _3, . ’_
- . t 1 ‘ . . O.
' .‘uu \‘t'irr'. .
1 ' , I
‘ ,
. O .- t
. ).|‘ r:- c ) .
‘ 0' : .§ ‘
3 1

I

 

 

Solvent- 0H30H, H01, H20 (65. 17: 18).

4942125987635713118101

Thln layer chromatograms o__ffiuflh~“

I standard compounds and dift‘usate
II components. Cellulose MN300 plate (laboratory prepared);

(See Figure 11 for sample
identification.)

 

1) 3 “3‘93 i o
(g x) Q

7

8

 

 

 

 

 

 

‘Figure 13. Thin layer chromatogram.of diffusate II components. Cellulose
MN300 plate (laboratory prepared); solvent-tert. amyl alcohol,
formic acid, water (3:2:1)(l-r. Diffusate II, 8-peak 8, 7-
peak 7, 6-peak 6.)

(X)

G
‘ /.,~:7 ,1 t
“ c " y I '
\ ' I. t. . x . \I I N

M

‘ _' M )
\ ' ‘\ éW’J
N..." ‘ M

 

 

 

Figure 14. Thin layer chromatogram.of diffusate II components. ECTEOLA
plate (laboratory prepared); solvent-0.15 M NaCl (l-r.
Diffusate II, 6-peak 6, 7-peak 7 and 8-peak 8.)

88

To determine whether spot 4 (Figure 13) was a nucleotide, peaks 7 and
8 were subjected to anion exchange thin layer chromatography. ECTEOLA
plates were prepared according to Randerath (1966) using 0.15 M sodium
chloride as the developing solvent. After the solvent was allowed to
travel 10 cm, both peaks 7 and 8 were separated into three components.

Peaks 7 and 8 both contained fractions that streaked from the origin to a
height of 1.5 and 1.0 cm, respectively. The Rf values for spots 2 and 3 of
peak 7 were 0.47 and 0.83, respectively, while peak 8 revealed two spots
with Rf values of 0.44 and 0.81 (Figure 14).

Peak 6 was also applied to ECTEOLA plates to confirm the absence of
nucleotides. Peak 6 was separated into two components. The first component
streaked from the origin to a height of 2.0 cm, while the second component
had a Rf value of 0.48. All spots were phosphate negative in peaks 6, 7 and
8. Peaks 7 and 8 were similar, since both revealed blue streaks from above
their second components (Rf = 0.47 and 0.44, respectively) to the solvent
front. Peak 6 demonstrated blue streaks from component 1 (0-2.0 cm streak)
to the solvent front. However, none of the components in peaks 6, 7 and 8
contained any phosphate positive.materia1. Therefore, nucleotides were
absent in the diffusate. The components of peaks 6, 7 and 8 that streaked
from the origin to 2.0, 1.5 and 1.0 cm, respectively, gave strong white
fluorescence under u.v. light. However, after spraying the chromatograms
with 2,7-dichlorofluorescein, the components that streaked at the origin
fluoresced giving a brilliant purple color under u.v. light.

All thin layer chromatographic systems except those designed for lactic

acid, creatine and its derivatives and nucleotides, indicated the presence

89

of uracil and cytosine in the diffusate. Therefore, thin layer chromato-
graphic analysis indicated the presence of uracil, cytosine, creatine,
creatinine, creatine phosphate and other phosphate-containing compounds,

as well as several fluorescent compounds. However, purine bases, nucleosides,

nucleotides and lactic acid were not detected in the diffusate.

Chemical Analysis of Dialyzable and Non-Dialyzable Components

The protein and sulfhydryl content of the water soluble protein frac-
tion of adipose is given in Table 4. Free amino group, aldose and sulfhydryl
analyses of diffusates I and II.are also listed in Table 4.

Table 4. Chemical analyses of water soluble dialyzable and non-dialyzable
adipose tissue components.

 

 

IfiProtein Sulfhydryl 4_:Free amino groups Aldoses
Treatment 13341111 (Tm/grow in umlg powder,7 wig powder7
N21 9.81 9.05
022 9.30 8.17
N283 12.16 16.00
0284 11.61 15.79
15 Trace 273.20 40.23
116 Trace 48.24 3.76

 

lAdipose tiesue stored under nitrogen
ZAdipose tissue stored under air

3Adipose tissue with 3% NaCl stored under nitrogen
4Adipose tissue with 3% NaCl stored under air
5Combined diffusates from.N2 and 02 treatments
6Combined diffusatesfrom.N28 and 02S treatments

7Lyophilized extract

90

On studying the precursors of hydrogen sulfide in the water soluble
fraction of adipose tissue, the effect of different atmospheres (nitrogen
and air) on the protein sulfhydryl groups was determined. Since sausage
products contain high fat levels and approximately 3% sodium chloride, the
influence of air or nitrogen atmospheres on adipose tissue is of special
interest. It seemed probable that air would reduce the sulfhydryl content
due to oxidation. As sodium chloride has been shown to accelerate autoxi-
dation of adipose tissue (Ellis gt gt., 1968; Gaddis, 1952; Chang and Watts,
1950) and because autoxidation continues during freezer storage (Ellis gt
gt., 1968; Gaddis, 1952), it was anticipated that the sulfhydryl content
of the salt treated samples would decrease to a greater extent than that
of the salt-free samples.

Results suggest that there was a slight atmospheric effect on the
sulfhydryl content in both the salt and the salt-free treatments. In both
treatments, the sulfhydryl content was slightly lower in air than in the
nitrogen atmosphere. However, the.most significant and unexpected finding
was the difference in sulfhydryl content between the salt and salt-free
treatments. Under either nitrogen or air, the sulfhydryl content of the
salt treatments was almost twice that of the salt-free treatments. The
greater sulfhydryl content in the salt treatments was also verified by
disc gel electrophoresis (Figure 8) as shown by a larger number of sulfhy-
dryl bands. .

Ellis 5 g. (1968) reported that the fat immediately adjacent to the
lean streaks in bacon had higher peroxide and monocarbonyl values. They

found an increase in the lean to fat ratio greatly enhanced autoxidation

91

during freezer storage in lean and fat mixtures containing 4% sodium
chloride. It was also noted that the C6 alkanal was the primary oxidation
product, while there appeared to be a selective oxidation of linoleate.
They suggested that the lean in the sodium chloride-cured bacon exerted

a special effect on tissue oxidation, and that some component(s) in the
lean was/were activated by sodium chloride, which changed the oxidation
characteristics of pork adipose tissue. They finally concluded that a
characteristic sodium chloride-accelerated oxidation occurred during
freezer storage of pork. They stated that the meat pigments were relatively
inactive in the salt-free samples during freezer storage of pork, although
they acknowledged the non-specific autoxidation effect of hematin. They
then speculated that heme pigments.may specifically decompose linoleate
hydroperoxide to form monocarbonyls. They further stated that nothing

is known about the separate catalytic effects of heme pigments.

Lea (1937) reported the presence of a lipoxidase in pork. However,
Tappel (1952) was unable to demonstrate the presence of lipoxidase in
animal tissues. Ellis gt gt. (1968) acknowledged the absence of lipoxi-
dase, but stated that knowledge of the independent oxidative influence of
sodium chloride and lipoxidase would serve to clarify the mechanisms in-
volved.

In the present study, disc gel electrophoresis separated the water
soluble extract of adipose tissue into a number of bands. The sulfhydryl
content of these bands was greater in the salt treatments (Figure 8). It
is possible that sodium chloride could bring about an increase in the
number of sulfhydryl bands in two ways. First, salt could stimulate re-

duction of disulfides, or secondly, it could stabilize the sulfhydryls

92

present by preventing their oxidation. The stabilizing action of salt
seems to be more reasonable than the reduction of disulfides, yet the
mechanism for sulfhydryl stabilization is not understood.

The sulfhydryl content of both diffusates I and II could not be
determined. There was a very slight increase in the yellow color of the
samples, but the amount of chromophore developed was too small to be
accurately.measured.

Both aldoses and free amino groups were present in the diffusate,
therefore, a system conducive to development of browning existed. On
heating adipose tissue to 100°C, browning occurred. Thus, the carbonyl
compounds and free amines from adipose tissue may, in part, be responsible
for browning of meat. Tonsbeek gt a. (1969) ether extracted the aqueous
fraction of a heated beef-water slurry, which possessed a typical meaty
odor, and detected two furanone derivatives. They stated that the deri-
vatives resulted from.the browning reaction and could contribute to the
aroma of beef broth, thus, indicating that browning products.may be in-
volved in the production of meat aroma. Therefore, the aldoses and amine
compounds of the diffusate may participate in production of meat aroma.
Since amino acids have been found in the diffusates, they could be re-
sponsible for some of the free amino groups.

Peaks 7 and 8 of diffusate II were very tightly held by the anion
exchange resin. Thus, it is possible that they may contain sulfate groups.
White gt gt. (1964) have previously reported that sulfate groups are pre-
sent in many biological systems, such as blood, urine, lipids, mucopoly-

saccharides and others. Therefore, the diffusate fractions were subjected

93

to the barium chloride test for sulfate. However, the amount of precipi-

tate was so minute that the test was inconclusive.
Development of a Method to Determine Sulfhydpyl and Disulfide Groupg

The analysis of sulfhydryl and disulfide groups was based on the
fermation of a S-nitroso derivative with ultimate formation of an azo dye.
The first step involved disulfide reduction with tetramethylammonium

borohydride (Equation 1).

125312 fill—e 2 RSH (1)

The excess reducing agent was destroyed with an excess of nitrous

acid (Equation 2).

NH BO + 4 N + 7 H 0

3 (CH3)4 NBH4 + 8 HONO -—-> 3 (0H3)4 2 3 2 2 (2)

The excess nitrous acid reacted with the thiol to form.a S-nitroso

derivative (Equation 3).

>RSNO+H0 (3)

RSH + HONO 2

 

The remaining nitrous acid was destroyed with ammonium sulfamate

(Equation 4).

 

HONO + NH4S03NH2 > NH4HS04 + N2 + H20 (4)

The S-nitroso derivative was cleaved with mercuric chloride and

nitrosyl chloride was generated (Equation 5).

RSNO + HgCl ——> (RS)2 Hg + 2 N001 (5)

2

94

The nitrosyl chloride combined with sulfanilamide to form a diazonium

salt (Equation 6).

NH2 N 5 N,C1
0 + N001 -—>© + H20 (6)
502th 50an2

The diazonium salt was coupled with N-l-naphthylethylenediamine to
form.an azo dye (Equation 7). The dye gives.maximum absorbance at 540 nm
and exhibits a l to l stoichiometric relationship with the thiols, i.e.,

one mole of dye is formed per mole of thiol.

 

+ -
N ; N,01 HN—CHZCHZNHZ HN—CHZCHZNHZ
+ 5 7
0 H
60 NH N
2 2 N

‘2...

2 2

Because urea destroys nitrous acid, it was not suitable as a protein
denaturating agent. In the present study, sodium lauryl sulfate (1%) in
0.1 M phosphate buffer (pH 8.0) was successfully used in studying the re-
duction of DTNB. Since sodium lauryl sulfate (0.9%) has been used by
Damjanovich and Kleppe (1967) as a denaturating agent for exposing the
sulfhydryl groups in proteins, it would appear to be applicable to the
present method. In the present study, a period of 5 min was allowed for
the destruction of nitrous acid by ammonium sulfamate. The concentration
of mercuric chloride was kept to a.minimum to prevent formation of a mercuric

chloride-diazonium slat complex, which gives a precipitate.

95

Since Kobayashi (1966) demonstrated that the diazonium salt decomposed
with increasing temperatures, it is suggested that the method be modified
by reducing the temperature of the reaction mixture to 5°C after destruction
of nitrous acid with ammonium sulfamate. The mercuric chloride-sulfanila-
mide solution should also be cooled to 5°C to insure that diazonium salt
decomposition is held at a minimum. In modifying the method, it is also
believed that cooling of the N-l-naphthylethylenediamine solution to 5°C
would increase the stability of the diazonium salt prior to coupling,
even though Kobayashi (1966) reported that coupling did not occur below
14°C. Since temperatures higher than necessary are to be avoided, it is
believed that coupling would occur at the minimum critical temperature by
allowing the reaction mixture to warm.up gradually at room.temperature.
Sodium lauryl sulfate caused foaming on agitation of the reaction.mixture,
so acetone was added to the Nel-naphthylethylenediamine solution as an
antifoaming agent. The acetone, however, changed the maximum absorbance
from 540 to 550 nm.

When using DTNB as the potential source of sulfhydryl groups, 5%
potassium cyanide in 1.0 N ammonium hydroxide was used for'reducing the
disulfides. The method was very sensitive and reproducible. However,

DTNB possesses no u-amino group, so formation of an iminothiazolidine

ring was not possible as is the case for cystine. Since potassium cyanide
could not be used as a reducing agent for cystine residues in proteins, a
suitable reducing agent was sought. The reducing agent must be stable in
aqueous solutions, quantitatively reduce disulfides in an alkaline solution

at a pH < 9.0, be completely destroyed by nitrous acid and produce no

96

decomposition products detrimental to the reaction. Tetramethylammonium
borohydride appears to be a suitable reducing agent and is presently under
investigation to determine whether it is compatible with protein solutions.

In order to allow addition of 0.5 ml aliquots of each reagent to 0.5
ml of the sample, the following concentrations of reagents are necessary:

Thiol - Dissolve 59.5 mg DTNB in 100 ml 95% ethanol. Dilute

50 to l with ethanol.

Reducing agent - Dissolve 22.5 mg tetramethylammonium borohydride

in 10 ml water.

Solution A - Dissolve 0.5000 g sodium nitrite in 100 ml 1.0 N H01.

Solution B - Dissolve 0.6000 g ammonium sulfamate in 100 ml water.

Solution 0 - Make a solution containing 0.0490 g mercuric chloride

in 100 ml 0.4 N H01. Add 0.2580 g sulfanilamide to 50 ml
of the solution.

Solution D - Dissolve 0.1950 g Nel-naphthylethylenediamine dehydro-

chloride in 10 ml 0.4 N H01 and make to a final volume of
25 ml with acetone.

To 0.5 ml of disulfide, add 0.5 m1 reducing agent (tetramethylammonium
borohydride) and allow the necessary time for complete disulfide reduction.
To the reduced solution, add 0.5 ml solution A (nitrous acid) and allow
1 to 2 min for the destruction of the reducing agent and the formation of
the S-nitroso derivative. Add 0.5 ml of solution B (ammonium sulfamate)
and allow 5 min for complete destruction of excess nitrous acid. 0001 the
mixture to 5°C in an ice bath and add 0.5 ml solution 0 (sulfanilamide-

mercuric chloride) at 5°C, allow 1 to 2 min for the formation of the

mercuric mercaptide and the diazonium salt. Add 0.5 ml of solution D
(N-l-naphthylethylenediamine), at 5°C, to the cool reaction.mixture and
allow time for maximum color development. The dye has been reported to
be stable for at least 3 hrs by Kobayashi (1966).

Elemental analysis of the synthesized N-l-acenaphthylethylenediamine
has not been carried out to date. It has, however, undergone coupling in
an acid medium with a diazonium salt, which indicates that it is an aromatic
amine. The dye formed is similar in color to the azo dye of N-l-naphthyl-
ethylenediamine.

The method is rapid for both sulfhydryl and disulfide groups in low
molecular weight disulfides. After protein denaturation and exposure of
the disulfide groups, however, the method should also be rapid and effi-
cient. The time required for protein denaturation is dependent on the
nature of the protein or proteins being studied. The actual time to com-
plete color development from thiol formation is approximately 20 min for
triplicate samples. The method is specific for thiols and none of the
reagents absorb in the region of the dye. However, the method cannot be
used if the material under investigation contains substances that will
destroy nitrous acid or diazonium salts. Fortunately, substances of this
nature are not normally present in proteins, so the method should be use-

ful for disulfide and sulfhydryl analyses of proteins.

SUMMARY

Beef adipose tissue was separated into the water soluble, salt soluble
and insoluble (water-salt-ether) fractions, and the amount of hydrogen
sulfide evolved upon heating each of the fractions was determined. The
yield of hydrogen sulfide per 100 g of adipose tissue was 17.6, 1.7 and
5.5 pH for the water soluble, salt soluble and involuble fractions, re-
spectively. Thus, the water soluble fraction contributed 71% of the
hydrogen sulfide, while the corresponding contributions from.the salt
soluble and insoluble fractions amounted to approximately 7 and 22%.

The water soluble fraction of adipose tissue was subjected to four
treatments (air and nitrogen atmospheres and 3% sodium chloride under each
atmosphere). The sulfhydryl groups and other possible flavor precursors
were determined. The aqueous extract from.each treatment was divided into
dialyzable and non-dialyzable fractions. The dialysates were concentrated
by pervaporation and dialysis against 1 M sucrose. The diffusates were
concentrated by either flash evaporation or lyophilization.

The dialysates were successfully separated by disc gel electrophoresis,
and demonstrated the presence of heme proteins and the absence of glyco-
proteins, nucleoproteins and lipoproteins. The sulfhydryl contents of the
nitrogen, air, nitrogen with 3% salt and air with 3% salt treatments were
' 9.05, 8.17, 16.00 and 15.79 uM per g of protein. Thus, there was a slight
atmospheric effect on the sulfhydryl contents, since the air treatments

for both salt and salt-free treatments were lower than those under nitrogen.

98

99

The salt treatments under either air or nitrogen contained almost twice
the sulfhydryl content of the salt-free treatments. The increase in
sulfhydryls in the salt treatments were also reflected in the disc gel
electrophoretic patterns, which contained a greater number of sulfhydryl
containing bands.

Amlno acid analysis of both dialysates and diffusates demonstrated
the presence of lysine, histidine, arginine, aspartic acid, threonine,
serine, glutamic acid, proline, glycine, alanine, cystine, valine, methio-
nine, isoleucine, leucine, tyrosine and phenylalanine. Glycine, alanine
and glutamic acid comprised approximately 25 and 50% of the amino acid
residues in the dialysates and diffusates, respectively.

The diffusates were separated into a number of fractions employing
a Dowex l anion exchange resin (formate form). These fractions were sub-
jected to thin layer chromatography and contained creatine, creatine
phosphate, creatinine, uracil, cytosine, some unidentified phosphate
containing compounds and a number of fluorescent compounds. Purine bases,
nucleosides, nucleotides and lactic acid were absent.

A quantitative method for determination of sulfhydryl and disulfide
groups in proteins based upon formation of an azo dye was investigated.
After reduction of the disulfides, the sulfhydryl groups were nitrosated.
The S-nitroso derivatives were cleaved with mercuric chloride. The re-
sulting nitrosyl chloride was combined with sulfanilamide to form a diazon-
ium salt. The diazonium salt was coupled with Nel-naphthylethylenediamine
to give an azo dye. The dye absorbs maximally at 540 nm. The thiol:dye

exists in a 1:1 stoichiometric relationship.

BIBLIOGRAPHY

Allen, R. 0., Popp, R. A. and Moore, D. J. 1965. Separation and relative
quantitation of mouse plasma esterases with disc electrophoresis. J.
Histochem. Cytochem. 13:249.

American Instrument Co. 1961. The determination of nitrogen by the
Kjeldahl procedure including digestion, distillation and titration.
American Instrument Co. Reprint No. 104.

Amoore, J. E. 1967. Stereochemical theory of olfaction. In Chemistr
and Physiology of Flavors. Ed. H. W. Schultz, E. A. Day and L. M.
Libbey. p. 122. The AVI Publ. 00., Inc., Westport, Conn.

Barbella, N. G., Hankins, O. G. and Alexander, L. M. 1936. The influence
of retarded growth in lambs on flavor and other characteristics of
the meat. Proc. Am. Soc. An. Prod. 29:289.

Barylko-Pikielna, N. 1957. The components of meat flavor. Przemysl.
Spozywczy. 11:26 (CA 54,18821F).

Batzer, O. F., Santoro, A. T., Landmann, W. A. and Schweigert, B. S. 1960.
Precursors of beef flavor. J. Ag. Food Chem. 8:498.

Batzer, O. F., Santoro, A. T., and Landmann, W. A. 1962. Identification
of some beef flavor precursors. J. Ag. Food Chem. 10:94.

Beaton, G. H., Selby, A. E. and Wright, A. M. 1961. Starch gel electro-
phoresis of rat serum proteins. I. Procedure and designation of
components. J. Biol. Chem. 236:2001.

Bender, A. E., Wood, T. and Palgrave, J. A. 1958. Analysis of tissue
constituents extract of fresh ox muscle. J. Sci. Food Agr. 9:812.

Bender, A. E., and Ballance, P. E. 1961. A preliminary examination of
the flavor of meat extract. J. Sci. Food Agr. 12:683.

Benne, E. J., Van Hall, N. H. and Pearson, A. M. 1956. Analysis of fresh
meat. J. Assoc. Offic. Agr. Chemists 39:931.

Blackburn, S. 1966. Micro techniques for amino acid analysis and peptide
separation based on high-voltage electrophoresis. In Anal ical Methods
of Protein Chemistpy. Eds. P. Alexander and H. P. Lundgren. Vol. 4.
p. 89. Pergamon Press, New York.

 

100

101

Bouthilet, R. J. 1949. A note on the nature of a flavor constituent from
poultry meat. Food Technol. 3:118.

Bouthilet, R. J. 1950. Chicken flavor: Separation and concentration of
its volatile components from broth. Food Res. 15:322.

Bouthilet, R. J. 1951a. Chicken flavor: The fractionation of the vola-
tile constituents. Food Res. 16:137.

Bouthilet, R. J. 1951b. Chicken flavor: The source of the meat flavor
component. Food Res. 16:201.

Bratton, A. C. and Marshall, Jr., E. K. 1939. A new coupling component
for sulfanilamide determination. J. Biol. Chem. 128:537.

Cava, M. P., Merkel, K. E. and Schlessinger, R. H. 1965. Pleiadene Systems
- II. On the mechanism of acepleiadylene formation -- a vinylogous
elimination in the acenaphthene series. Tetrahedron 21:3059.

Cavallini, D., Graziani, M. T. and Dupre, s. 1966. Determination of
disulfide groups in proteins. Nature 212:294.

Chang, I. and Watts, B. M. 1950. Some effects of salt and moisture on
rancidity in fats. Food Res. 15:313.

Crambach, A., Reisfeld, R. A., Wyckoff, M. and Zaccari, J. 1967. A pro-
cedure for rapid and sensitive staining of protein fractionated by
polyacrylamide gel electrophoresis. Anal. Biochem. 20:150.

Crocker, E. C. 1948. Flavor of meat. Food Res. 13:179.

Damjanovich, S. and Kleppe, K. 1967. The number of SH groups in rabbit
muscle phosphorylase. Biochem. Biophys. Res. Comm. 26:65.

Davis, B. J. 1964. Disc electrophoresis. 11. Method and application to
human serum.proteins. Ann. N. Y. Acad. Sci. 121:404.

Doty, D. M., Batzer, 0. F., Landmann, W. A. and Santoro, A. T. 1961.
Meat flavor. Proceedings Flavor Chemistpy Symposium. Campbell Soup
Company, Camden, New Jersey, p. 7.

 

Ellis, R., Currie, G. T., Thornton, F. E., Bollinger, N. 0., and Gaddis,
A. M. 1968. Carbonyls in oxidizing fat 11. The effect of the pro-
oxidant activity of sodium chloride on pork tissue. J. Food Sci.
33:555..

Ellman, G. L. 1959. Tissue sulfhydryl groups. Arch. Biochem. Biophys-
82:70.

102

Fieldner, A. 0., Sayers, R. R., Yant, W. P., Katz, S. H., Shohan, J. B.
and Leitch, R. D. 1931. Warning agents for fuel gases. U.S.Dept.
Commerce, Bureau of Mines, Mono. 4.

Florkin, H. and Stolz, E. H. 1963. Proteins. In Qgpprehensive Biochemis-
tgy. Ed. H. Fraenkel-Conrat. Vol. 7. p. 73. Elsevier Publ. 00.,
New York.

Folch, J., Lees, M. and Stanley, G. H. S. 1957. A simple method for the
isolation and purification of total lipides from.animal tissues. J.
Biol. Chem. 226:497.

Gaddis, A. M. 1952. Effect of pure salt on the oxidation of bacon in
freezer storage. Food Technol. 6:294.

Gaitonde, M. K. 1967. A spectrophotometric method for the direct deter-
mination of cysteine in the presence of other naturally occurring
amino acids. Biochem. J. 104:627.

H

Ganshirt, H., Waldi, D. and Stahl, E. 1965. Synthetic organic materials.
In Thin Lgyer Chromatography. Ed. E. Stahl. p 344. Academic Press
Inc., New York.

Gifford, G. T. and Yukins, L. 1965. Protein patterns in human parotid
saliva. J. Chromat. 20:150.

Grassetti, D. R. and Murray, Jr., J. R. 1967. Determination of sulfhydryl
groups with 2,2'- or 4,4'-dithiodipyridine. Arch. Biochem. Biophys.
119:41.

Greenwood, D. A., KraybiIL H. R. and Schweigert, B. S. 1951. The amino
acid composition of fresh and cooked beef cuts. J. Biol. Chem. 193:23.

Hamm, R. and Hofmann, K. 1965. Changes in the sulfhydryl and disulfide
groups in beef muscle proteins during heating. Nature 207:1269.

Herz, W. J. and Shallenberger, R. S. 1960. Aromas produced by simple
amino acid, sugar reactions. Food Res. 25:491.

Hofstrand, J. and Jacobson, M. 1960. The role of fat in the flavor of
lamb and mutton as tested with broths and with depot fats. Food Res.
25:706.

Hornstein, I., Crowe, P. F. and Sulzbacher, W. L. 1960a. Constituents
of meat flavor: Beef. J. Ag. Food Chem. 8:65.

Hornstein, I. and Crowe, P. F. 1960b. Meat flavor chemistry: Studies
on beef and pork. J. Ag. Food Chem. 8:494.

Hornstein, I., Crowe, P. F. and Sulzbacher, W. L. 1963a. Flavor of beef
and whale meat. Nature 199:1252.

103

Hornstein, I. and Crowe, P. F. 1963b. Meat flavor: Lamb. J. Ag. Food
Chem. 11:147.

Hornstein, I. and Crowe, P. F. 1964. Meat flavor -- A review. J. Gas
Chromatog. 2:128.

Hornstein, I. 1967. Flavor of red meats. In Chemistry and Physiology
of Flavors. Ed. H. W. Schultz, E. A. Day and L. M. Libbey. p 228.
The AVI Publ. 00., Inc., Westport, Conn.

Howe, P. E. and Barbella, N. G. 1937. The flavor of meat and meat products.
Food Res. 2:197.

Ing, H. R. and Manske, R. H. F. 1926. A modification of the Gabriel
synthesis of amines. J. Chem. Soc. 2348.

Jacobson, M. and Koehler, H. H. 1963. Meat flavor. Components of the
flavor of lamb. J. Ag. Food Chem. 11:336.

Jolley, W. B., Allen, H. W. and Griffith, O. M. 1967. Ultracentrifugation
using acrylamide gel. Anal. Biochem. 21:454.

Jones, 0. 1952. The flavoring of meat and meat products. I. Perfume and
Essential Oil Rec. 43:336.

Karush, F., Klinman, N. R. and Marks, R. 1964. An assay method for disul-
fide groups by fluorescence quenching. Anal. Biochem. 9:100.

Kazeniac, S. J. 1961. Chicken flavor. Proceedings Flavor Chemistry
Symposium. Campbell Soup 00., Camden, New Jersey. p. 37.

Klose, A. A., Palmer, H. H., Lineweaver, H. and Campbell, A. A. 1966.
Direct olfactory demonstration of chicken aroma. J. Food Sci. 31:638.

Kobayashi, Y. 1966. Spectrophotometric determination of cyclohexanone
oxime in sulfuric acid solution of epsilon-caprolactam. Anal. Chem.
38:917.

Koehler, H. H. and Jacobson, M. 1967. Characteristics of chicken flavor
-containing fraction extracted from raw muscle. J. Ag; Food Chem.
15:707.

Kramlich, W. E. and Pearson, A. M. 1958. Some preliminary studies on
meat flavor. Food Res. 23:567.

Kramlich, W. E. and Pearson, A. M. 1960. Separation and identification
of cooked beef flavor components. Food Res. 25:712.

Lea, C. H. 1937. The influence of tissue oxidases on rancidity. Oxidation
of the fat of bacon. J. Soc. Chem. Ind. 56:376T.

104

Leach, S. J. 1966. The estimation of thiol and disulfide groups. In
Analytical Methods of Protgin Chemistry. Eds. P..Alexander and H. P.
Lundgren. p 3. Pergamon Press, New York.

Lento, H. G., Ford, J. A. and Denton, A. E. 1964. A method for deter-
.mining 5'-nucleotides. J. Food Sci. 29:435.

Lineweaver, H. and Pippen, E. L. 1961. Chicken flavor. Proc. Flavor
Chemistrijymposium, p. 21, Campbell Soup 00., Camden, New Jersey.

Lobanov, D. I. and Wolfson, G. 1958. The formation of melanoidins during
preparation of meat and its effect on the gastric secretion of dogs.
Nahrung. 2:660. (CA 53,12513i).

Macy, R. L., Naumann, H. D., and Bailey, M. E. 1964a. ‘Water-soluble
flavor and odor precursors of meat. 1. Qualitative study of certain
amino acids, carbohydrates, non-amino acid nitrogen compounds, and
phosphoric acid esters of beef, pork, and lamb. J. Food Sci. 29:136.

Macy, R. L., Naumann, H. D. and Bailey, M. E. 1964b. Water-soluble flavor
and odor precursors of meat. II. Effects of heating on amino nitrogen
constituents and carbohydrates in lyophilized diffusates from aqueous
extracts of beef, pork and lamb. J. Food Sci. 29:142.

MCNeil, T. L. and Beck, L. V. 1968. Fluorometric estimation of GSH-OPT.
Anal. Biochem. 22:431.

Mecchi, E. P., Pippen, E. L. and Lineweaver, H. 1964. Origin of hydrogen
sulfide in heated chicken muscle. J. Food Sci. 29:393.

Minor, L. J., Pearson, A. M., Dawson, L. E. and Schweigert, B. S. 1965a.
Gas chromatographic analysis of volatile constituents from cooked
carcasses of old and young chickens. Poultry Sci. 44:535.

Minor, L. J., Pearson, A. M., Dawson, L. E. and Schweigert, B. S. 1965b.
Chicken flavor: the identification of some chemical components and
the importance of sulfur compounds in the cooked volatile fraction.
J. Food Sci. 30:686.

Minor, L. J., Pearson, A. M., Dawson, L. E. and Schweigert, B. S. 1965c.
Separation and identification of carbonyl and sulfur compounds in
the volatile fraction of cooked chicken. J. Ag. Food Chem. 13:298.

Minor, L. J., Pearson, A. M. and Stine, C. M. 1966. Chicken flavor studies.
J. Ag.. Food Chem. 14:416.

Moncrieff, R. W. 1967. Introduction to the symposium. In Chemistry and

Phyaiglggy 9f thvorg. Ed. H. W. Schultz, E. A. Day and L. M. Libbey.
p. 5. The AVI Publ. 00., Inc., Westport, Conn.

105

Moore, S., Spackman, D. H. and Stein, W. H. 1958. Chromatography of amino
acids on sulfonated polystyrene resins. Anal. Chem. 30:1185.

"
Muller, G. 1965. Blood-sugar determination with o-toluidine. Dt. Z.
Verdau.-u. Stoffwechs Krankh. 25:77. (Anal. Abst. 14, 1559, 1967).

Newman, H. E. 1891. Ueber einige Derivate des Aethylenphenyldiamins und
Homologe desselben. Ber. chem. Ges. 21:2198.

Peacock, A. C. and Dingman, C. W. 1967. Resolution of multiple ribonucleic
acid species by polyacrylamide gel electrophoresis. Biochemistry
6:1818.

Pearson, A. M., Harrington, G., West, R. G. and Spooner, M. E. 1962. The
browning produced by heating fresh pork. 1. The relation of browning
intensity to chemical constituents and pH. J. Food Sci. 27:177.

Pearson, A. M., Tarladgis, B. G., Spooner, M. E. and Quinn, J. R. 1966.
The browning produced on heating fresh pork. II. The nature of the
reaction. J. Food Sci. 31:184.

Pippen, E. L., Campbell, A. A. and Streeter, J. V. 1954. Flavor Studies.
Origin of chicken flavor. J. Ag. Food Chem. 2:364.

Pippen, E. L. and Eyring, E. J. 1957. Characterization of volatile nitro-
gen and volatile sulfur fractions of cooked chicken and their relation
to flavor. Food Technol. 11:53.

Pippen, E. L., Nonaka, M., Jones, F. T. and Stitt, F. 1958. Volatile
carbonyl compounds of cooked chicken. I. Compounds obtained by air
entrainment. Food Res. 23:103.

Pitt-Rivers, R. and Schwartz, H. L. 1967. The thiol groups of thyroglobin.
Biochem. J. 105:28c.

Rampton, J. H. 1969. Separation, identification and characterization of
some myofibrillar proteins. Unpublished Ph.D. Thesis. Mich. State
Univ., East Lansing, Mich.

Randerath, K. 1966. Thin-Layer Chromatogtaphy. Academic Press, New York.

Richards, E. G., 0011, J. A. and Gratzer, W. B. 1965. Disc electrophoresis
of ribonucleic acid in polyacrylamide gels. Anal. Biochem. 12:452.

Sadikov, V. S., Shoshin, A. F., Starukhina, K. M. and Livshitz, M. E.
1934. Origin of hydrogen sulfide produced when chicken flesh is
cooked. Compt. Rend. Acad. Sci. U.S.S.R. 3:39. (CA 28,68664).

Sands, A. E., Grafius, M. A., Wainwright, H. W. and Wilson, M. W. 1949.
The determination of low concentrations of hydrogen sulfide in gas
by the methylene blue method. U.S. Bureau of Mines. Report of In-
vestigation 4547.

106

Sawicki, E., Stanley, T. H., Pfaff, J. and D'Amico, A. 1963. Comparison
of fifty-two spectrophotometric methods for the determination of
nitrite. Talanta 10:641.

Scopes, R. K. 1964. The influence of post-mortem conditions on the solu-
bilities of muscle proteins. Biochem. J. 91:201.

Scopes, R. K. 1968. Methods for starch-gel electrophoresis of sarcoplasmic
proteins. An investigation of the relative mobilities of the glycolytic
enzymes from the muscles of a variety of species. Biochem. J. 107:139.

Smith, I. 1960. Solvent extraction techniques. In Chromatographic and
Electrophoretic Techniqugg. Ed. I. Smith. Vol. 1. p. 60. Inter-
science Publ., Inc., New York.

 

Smithies, O. 1959a. An improved procedure for starch-gel electrophoresis.
Further variations in the serum proteins of normal individuals. Biochem.
J. 71:585.

Smithies, O. 1959b. Zone electrophoresis in starch gels and its application
to studies of serum proteins. In Advances in Protein Chemistty.
Eds. M. L. Anson, K. Bailey and J. T. Edsall. Vol. 14. p. 76.
Academic Press, New York.

Spark, A. A. 1969. Role of amino acids in non-enzymatic browning. J.
Sci. Fd. Agr. 20:308.

Stahl. W. H. 1957. Gas chromatography and mass spectrometry. In Chemistry_
of Natural Food Flavors; a Symposium. p. 58. U.S. Quartermaster Food
and Container Institute for the Armed Forces, Chicago.

Tappel, A. L. 1952. Linoleate oxidation catalyzed by hog muscle and
adipose tissue extracts. Food Res. 17:550.

Tonsbeek, C. H. T., Plancken, A. J. and v. d. Weerdhof, T. 1968. Come
ponents contributing to beef flavor. J. Ag. Food Chem. 16:1016.

Wasserman, A. E. and Gray, N. 1965. Meat flavor l. Fractionation of
water-soluble flavor precursors of beef. J. Food Sci. 30:801.

Wasserman, A. E. and Talley, F. 1968. Organoleptic identification of
roasted beef, veal, lamb and pork as affected by fat. J. Food Sci.
33:219.

White, A., Handler, P. and Smith, E. L. 1964. Principles of Biochemistpy.
MoGraw-Hill Book Co., New York.

Wood, T. 1956. Some applications of paper chromatography to the examin-
ation of meat extract. J. Sci. Food Agr. 7:196.

107

‘Wood, T. and Bender, A. E. 1957. Analysis of tissue constituents.
Commercial ox-muscle extract. Biochem. J. 67:366.

Wood, T. 1961. Browning of oxemuscle extracts. J. Sci. Fd. Agr. 12:61.

Yueh, M. H., and Strong, F. M. 1960. Some volatile constituents of cooked
beef. J. Age. Food Chem. 8:491.

Zahler, W. L. and Cleland, W. W. 1968. A specific and sensitive assay
for disulfides. J. Biol. Chem. 243:716.

Zaika, L. L., Wasserman, A. E., Monk, Jr., 0. A. and Salay, J. 1968.
Meat flavor. 2. Procedures for the separation of water-soluble beef
aroma precursors. J. Food Sci. 33:53.

Zollinger, H. 1961. Azo and Diazo Chemistpy. Interscience Publ., Inc.,
New York 0

Zwaan, J. 1966. Detection of protein-bound sulfhydryl groups after agar
gel electrophoresis. Anal. Biochem. 15:369.

H
u
u
H
II.
u
"I
III"
In.'
E N

03142 8745