AN ABSTRACT OF THE THESIS OF
Anna B. Marin for the degree of Master of Science in
Food Science and Technology presented on August 19, 1985.
Title:
An Analytical and Sensory Evaluation of the Aroma
Volatiles of Tuber gibbosum
Abstract approved:
(y
~ Leonard M. Libbey
The aroma of Tuber gibbosum, a native Oregon truffle, was
characterized using two distinct techniques.
Volatile aroma
constituents were identified by chemical analysis, and sensory
characteristics were determined by examining response of the
general populace to the truffle aroma.
Truffle samples from four maturity ranges, as determined
by microscopic examination of each specimen for spore maturity,
were compared for amounts and types of aroma volatiles present.
Aroma volatiles of frozen specimens of T. gibbosum were sampled
using a headspace concentration technique.
Volatile compounds
were analyzed by gas chromatography/mass spectrometry/data system
(GC/MS/DS) and the total volatile profile was found to contain up
to 20 compounds, depending on sample treatment and maturity.
Seven compounds were found to be the major aroma constituents and
were selected for further investigation.
These 7 compounds were
identified from their mass spectra and the identities
substantiated by determining their Kovats retention indices.
Amounts of volatile compounds present in each sample were
determined by capillary gas chromatographic (GO analysis using an
internal standard of n-tridecane for quantitative determinations.
The major aroma constituent for all the samples was oct-l-en-3-ol,
representing about 50% of the total aroma profile.
The other 7
major components were 8- and 6-carbon alchols, ketones and
aldehydes.
Quantitative data for all samples and components were
analyzed statistically.
A descriptive model for truffle maturity
was derived based on the amounts of 5 of the 7 compounds, and the
sum of the amounts of the 7 major constituents.
A comparison of volatiles from ascorbic acid treated and
untreated control truffle samples indicated an apparent reduction
in the amounts of several compounds.
The examination of public response to the aroma of T.
gibbosum was conducted as part of a display on truffles at the
Oregon Mycological Society Mushroom Show.
The aroma of this
truffle, and three other native Oregon truffles were rated for
desirability and preference.
Results indicate T. gibbosum was not
the favored sample, but was liked by about 2/3 of the population.
Comparison of responses by males and females showed no differences
for individuals liking the truffle aromas.
However, females rated
the truffle aromas of 3 of the 4 species as more unpleasant than
did males.
An Analytical and Sensory Evaluation of the Aroma Volatiles of
Tuber Gibbosum
by
Anna B. Marin
A Thesis
submitted to
Oregon State University
in partial fulfillment of
the requirements for the
degree of
Master of Science
Completed August, 19, 1985
Commencement June, 1986
APPROVED:
=
*■- w
u-' ~
"
jf
Professor of Food Science and Technology in charge of maj or
Chairman of Food Science and Technology Department
■y
—• ^—'
uate ^School
Dean af\the Graduate
Date thesis is presented
August 19, 1985
Typed by the author, Anna B. Marin
ACKNOWLEDGEMENTS
The author would like to express sincere appreciation to
the following individuals and organizations for contributing to the
completion of this thesis.
Special recognition is due to Dr. James
Trappe, U.S. Forest Service mycologist and minor department
committee member, who provided the inspiration and expertise to
initiate this research, and to his student, Wang Yun for his help
with spore counts.
Appreciation is also extended to my major
professor, Dr. Leonard Libbey, for his assistance in mass spectral
analysis, and for his patience in the completion of this thesis,
and to Dr. Ralph E. Berry of the Dept. of Entomology, and Brian
Arbogast of the Dept. of Agricultural Chemistry for use of the GC
and GC/MS/DS.
Special thanks to Frank Evans and his family, who
provided the use of their personal computer, "Ole Apple 738 plus,
plus, plus".
The assistance of A.B. Walters and Ethnobotanical
Research and Developement Enterprizes, Inc. (ERDE), and the members
of the North American Truffling Society (NATS) is gratefully
ackowledged for supplying truffles and training in truffling
expertise.
Recognition is also due to the Oregon Mycological
Society (OMS) who provided the site for the sensory testing, and
financial assistance in presenting a poster on that research at the
Institute of Food Technologists (IFT) National Meetings in Atlanta,
Ga., June 1985.
Finally, last but not least, sincere appreciation
is offered to Dr. Mina McDam'el, major department committee member,
mentor and friend,
who is responsible for inspiring the
introduction of the world to the sensory evaluation of truffles.
TABLE OF CONTENTS
I. Introduction
1
Truff1e ecology
Truffles as a source of nutrients
Truffles as a food source
Fungal odors
1
3
4
5
II. Characterization of the aroma volatiles of Tuber
gibbosum, a Native Oregon Truffle Species, Produced
During its Maturation
9
Abstract
Introduction
Materials and methods
. Truffle selection
Truffle sample preparation
Aroma volatiles concentration
Quantitative analysis of aroma volatiles
Qualitative analysis of aroma valatiles
Results and Discussion
Identification of the constituents from
T. gibbosum volatiles
Composition of T. gibbossum volatiles
A descriptive model of T. gibbosum
maturity based on changes in aroma
volatiles
A comparison of aroma volatiles from
ascorbic acid treated truffles with
untreated samples
Conclusions
References
10
11
12
12
13
13
15
16
18
18
18
21
24
25
26
III. A Comparative Sensory Analysis of the Public Response
to the Aroma Volatiles of Tuber gibbosum and Three
Other Native Oregon Truffles
Abstract ,
Introduction
Material s and Methods
Sample preparation
Sensory testi ng
Statistical analysis
Results and Discussion
Conclusions
References
IV. Bibliography
'.
28
29
30
32
32
32
33
34
40
41
42
LIST OF FIGURES
Figure
II.l.
Page
Change in the relative amount of n-hexanol
in the aroma volatiles of Tuber gibbosum
with increasing maturity
22
III.l. The relative frequency distribution of
hedom'c scores given for the aroma of four
truffle species by males and females
36
111.2. Mean hedom'c scores given for each of four
truffle aromas by male, female, and both
populations, for three ranges of hedom'c
scores
38
LIST OF TABLES
Table
II.l.
Page
Relative amounts of the compounds identified
as volatile aroma constituents from samples
of T. gibbosum at various stages of
maturity
19
III.l.
Percentage of population indicating one of four
truffles as favorite
35
111.2.
Aroma descriptors given for each of four
truffle species
39
AN ANALYTICAL AND SENSORY EVALUATION OF THE AROMA VOLATILES OF
TUBER GIBBOSUM
Chapter I.
INTRODUCTION
TRUFFLE ECOLOGY. Tuber gibbosum Harkness is a fungal species which
fruits abundantly in the Pacific Northwest conifer forests. Its
mycelium forms a mycorrhizae with a host tree, Douglas fir
(Pseudotsuga menziesii).
Mycorrhizae, which literally means
fungus-root, describes a symbiotic relationship which is formed
between the roots of higher plants and certain fungi.
mycelium colonizes the plant feeder roots,
The fungal
effectively extending
the host root system and assisting in the absorbtion of minerals,
water and nitrogen.
The tree, in turn, provides photosynthetic
products which are sources of energy for fungal growth and
reproduction.
The fruiting bodies of T. gibbosum are truffles, hypogeous
sporocarps which develop and mature underground.
Hypogeous fungi
are distinguished from epigeous, or above-ground fruiting fungi by
their spore dispersal mechanisims.
Epigeous fungi discharge their
spores into the air for wind dispersion, while hypogeous fungi
depend on animal mycophagy (fungal feeding) and spore dispersion
via the excreted fecal pellets.
These two distinct fungal types show evolutionary linkage
through the presence of genera that display both types of spore
dispersal mechanisms.
For example, the hypogeous fungus Geopora
cooperi is excavated and eaten by animals, but forcibly discharges
its spores when bitten open (Burdsall, 1965).
Mycologists agree
that this evolutionary link exists but until recently, disagreed
as to which fungal form was more advanced.
Trappe and Maser
(1977) pointed out that past arguments addressing this question
were based solely on morphological differences between the two
fungal forms.
They proposed convincing evidence that ecological
relationships, as well as morphology, are important in determining
the phylogenetic relationship.
Epigeous fungi are generally
adapted to moist conditions; they take up water readily to produce
sporocarps with stems to lift large spore-bearing caps into the
air.
A large part of their total biomass is devoted to maximizing
air contact for random spore dispersal in the wind.
However, this
morphology makes them vulnerable to the weather; exposure to
freezing or warm dry weather causes their sudden death.
Consequently, mushrooms are short-lived; they emerge, discharge
their spores and senesce within days or weeks.
In contrast,
hypogeous fungi produce fruiting bodies without stems or caps,
devoting most of their biomass to spore-bearing tissue enclosed in
a peridium or skin.
They mature over a period of weeks or months
because, unlike epigeous fungi, they are protected from the
weather by moist soil or humus.
According to Trappe and Maser
(1977), these mechanisms of economizing energy and dealing with
environmental stress are indications of evolutionary advancement.
Specialization is also generally regarded as evidence of
evolution.
Sporocarps of hypogeous fungi are eaten by small
mammals that excrete the spores at sites likely to contain
mycorrhizal host roots (Trappe and Maser, 1977).
It has been
documented that fungal spores remain morphologically unchanged and
viable when the sporocarps are ingested and the spores are
excreted by small animals (Trappe and Maser, 1976).
Even though
epigeous fungi display elaborate mechanisms for spore discharge
into the air, their, dispersal is random and inefficient when
compared with the specific deposit of hypogeous fungal spores by
small mammals.
Hypogeous fungi had to co-evolve with animals
(Trappe, 1977) because they depend upon them for spore dispersal.
TRUFFLES AS A SOURCE OF NUTRIENTS.
Hypogeous fungi have a unique
ecological niche, being symbionts with host trees and also
depending on animals for spore dispersal.
Their chemical
composition reflects these two functional roles as providers of
nutrients for host trees and also for the animals that eat their
sporocarps.
Studies with radioactive tracers indicate that
virtually all macro- and some micronutrients are taken up from the
soil by mycorrhizal fungi and translocated to the host plants.
(Trappe and Fogel, 1977).
Stark (1977) has shown that fungal
tissues serve as sinks for elemental nutrients that otherwise
would be leached from the soil.
The truffle, in turn, utilizes
various carbohydrates from the host tree.
It is not known which
particular photosynthates are necessary to the fungi beyond
thiamin and simple carbohydrates (Hacskaylo, et al., 1973), but
generally the fungi will only fruit when growing with a host
plant.
The sporocarps have also been found to be a relatively
concentrated source of nutrients for saprophytic organisms and
animals (Trappe and Fogel, 1977).
TRUFFLES AS A FOOD SOURCE.
Consumption of fungi as a primary food
source for small mammals has been demonstrated (Fogel and Trappe,
1978).
Studies analyzing the digestive tract contents of several
genera of small mammals indicates the consumption of a higher
proportion of hypogeous to epigeous fungi (Maser, et al., 1978).
This may be due to the extended seasonal availability of hypogeous
fungi (Fogel, 1976).
However, this demonstrated preference also
prevails when epigeous fungi are abundant (Maser, et al., 1978).
Another explanation might be that truffles offer more nutrition
than mushrooms.
Proximate analysis of several different truffle
species (Andreotti and Casoli, 1968, Al-Delaimy, 1977, Delmas,
1978, Ahmed et al., 1981, Al-Shabibi, et al., 1982) indicated
that, when compared with analysis of several species of mushrooms
(Trione, 1978), truffles are significantly higher in protein,
carbohydrates, and fat, and lower in moisture content.
findings
These
may be an important consideration if truffles make up a
significant portion of the diet, as they do for small forest
mammals.
Also, certain species of desert truffles of the genus
Terfezia are considered food staples for some human communities in
North Africa (Gray, 1970).
Small mammals may also prefer truffles because they tend
to locate fungi by odor rather than sight;
a behavior that has
been demonstrated in the deer mouse, Peromyscus maniculatus
(Howard et al., 1968). Immature truffle specimens are odorless but
volatile aroma compounds are produced that increase in intensity
and complexity during spore maturation.
produce strong odors.
Epigeous fungi also
However, the unique ripe truffle aromas may
be the more enticing to mammals who seek them out.
These odors
are important for fungal spore dispersion since they signal the
truffle location to animals that feed on them.
FUNGAL ODORS.
The scientific community has shown interest in
studying fungal volatiles for some time.
In his presidential
address to the British Mycological Society, Hutchinson (1971)
reviewed the research to date on fungal volatile metabolites and
their functional significance.
Although his discussion included
information on all types of fungal volatiles and their ecological
impact, he made no mention of those volatiles produced by
hypogeous fungi.
An earlier review by Harper, et al., (1968)
dealt more specifically with fungal odors.
He summarized early
publications which attempted to classify fungal odors (Gilbert,
1932) and to use odors to classify fungal species (Heim, 1957).
In his classification scheme, Gilbert (1932) identified one fungal
aroma type as "musky".
He felt this aroma was exemplified by
several truffle species, but most prominantly in Tuber
melanosporum.
Heim (1957) agreed with Gilbert's classification of
the truffle aroma but noted that the aroma was complicated and
variable in character.
He described it as somewhat garlic-like
but sometimes more floral in character, like the essence of
lavender flowers that are rich in linalyl acetate.
Josserand
(1952) was more critical of Gilbert's classification of fungal
odors stressing that the same odors were likely to be perceived
differently by different people.
Josserand suggested
standardization with the use of a permanent set of odorous
reference chemicals but finally rejected the idea because he felt
fungal odors were too complex to classify or describe.
Benedict and Stuntz (1975) acknowledged the variability in
odor perception among individuals but still felt mushroom odor
characterization could be a.taxonomic aid.
They described 27
different odors in fresh mushrooms and also indicated the chemical
compounds involved, if known.
However, they did not reference
the isolation of particular compounds from the mushrooms but
rather suggested that the fungi had odors similar to known aroma
volatiles.
The elucidation of complex aroma mixtures has advanced
significantly since the introduction of sophisticated analytical
instrumentation, particularly gas chromatography/mass spectrometry
(GC/MS).
In recent years, investigators have used this technique
to identify the volatile compounds which characterize the aromas
of several fungal species, especially those of commercial
interest.
The aroma constituents of a steam distilled concentrate
of the common cultivated mushroom, Agaricus bisporus were reported
by Cronin and Ward (1971).
Picardi and Issenberg (1973) used
similar techniques to examine the changes in the composition of
the volatiles when this mushroom was heated.
Likewise, the
volatile aroma compounds from a popular European mushroom. Boletus
edulis (Thomas, 1973), and two favorite Japanese mushrooms,
shiitake, Lentinus edodes (Morita and Kobayashi, 1966), and
matsutake, Tricholoma matsutake (Yajima et al., 1981), have been
investigated.
Maga (1976) recognized the economic potential of
cultivating certain fungi as commercial sources of natural flavor
compounds.
He reviewed papers of several investigators who
examined the aroma volatiles of several non-commercial mushrooms
and other classes of fungi.
For example, Collins and Halim (1971)
examined the volatiles of the filamentous fungus, Leratocystis
variispora, and compared the aroma constituents of three species
of mushrooms from the genus Phellinus (1972).
Halim et al. (1975)
isolated and identified odorous volatiles produced by the mold,
Penicillium decumbens.
Unfortunately, few investigators have published results of
chemical analysis identifying the volatile aroma constituents from
hypogeous fungi.
The unique aroma volatile,
bis-(methylthio)methane was isolated from the Italian white
truffle. Tuber magnatum and found to be the character-impact
compound for that truffle aroma (Fiecchi, et al., 1967).
Recently, two odoriferous steroids were isolated and identified in
Jj, magnatum and also in fresh and canned specimens of T.
melanosporum, the famous French black truffle (Claus, et al.,
1981).
These steroids, 5o(-androst-16-en-3-one, and its
corresponding alcohol, are known to be porcine sex pheromones
(Melrose, et a!., 1971, Claus and Hoffmann, 1971) and have also
been detected as excretion products from humans (Brooksbank, 1962,
Ruokonen, 1973, Brooksbank et al., 1974), although their function
in humans is uncertain.
However, Kirk-Smith, et al. (1978) have
reported studies indicating that these steroids may demonstrate
pheromonal activity in humans.
The amounts of the steroids found
in truffles is above human perception thresholds (Griffiths and
Patterson, 1970).
However, the contribution of the odor of these
steroids to the overall truffle aroma has not been reported.
Their occurrence in truffles is significant, and offers an
explanation why pigs can detect these fungi up to 1 m underground
(Claus, et al., 1981).
The emphasis of this study has been to characterize the
volatile aroma compounds of the native Oregon truffle, T.
gibbosum, by two different methods:
evaluation.
chemical analysis and sensory
Volatile aroma constituents were chemically
identified and quantified for truffle samples at various stages of
maturity in order to characterize the changes that might occur
during maturation.
Sensory characteristics for T. gibbosum were
obtained by statistically analyzing public response for 1) degree
of liking or disliking the truffle aroma (hedonic response); and
2) truffle aroma preference when T. gibbosum was compared with
three other native truffles.
Chapter II.
Characterization of the Aroma Volatiies
of Tuber gibbosum, a Native Oregon Truffle Species,
Produced During Its Maturation
10
ABSTRACT
The aroma volatiles produced by Tuber gibbosum Harkness, a
North American truffle species, were chemically analyzed to
determine the constituents.
Quantitative examination of the
volatiles led to the development of a discriminant model for
maturity based on the amounts of 5 of the 7 volatile components
identified and the total volatiles present.
11
INTRODUCTION
Truffles, the fruiting bodies of hypogeous fungi, are known
to produce intense volatile aroma compounds at maturity.
It has
been hypothesized that the process of spore maturation is
correlated with intensified odor emissions in hypogeous fungi in
order to extend the chances for animal mycophagy and spore
dispersal (Maser, et. al, 1978).
The purpose of this study was to
examine the aroma volatiles of T. gibbosum at various stages of
spore maturity to see if a correlation does indeed exist between
odor production and spore maturation.
Several analytical techniques and experimental comparisons
were used to begin to describe the unique aroma volatiles of T.
gibbosum.
The identity of the compounds which compose the truffle
aroma volatiles was elucidated by gas chromatography/mass
spectrometry, GC/MS.
Quantitative analysis of the volatile
constituents by GC assisted in the developement of a descriptive
model of truffle maturity based on changes in the aroma volatiles.
Finally, samples of ascorbic acid treated truffles were examined
and compared to untreated samples to determine if this treatment
affected aroma volatile composition.
12
MATERIALS AND METHODS
TRUFFLE SELECTION.
Fresh and frozen untreated specimens of T.
gibbosum were supplied by members of the North American Truffling
Society (NATS).
Frozen ascorbic acid treated T. gibbosum
specimens were donated by A. B. Walters of Ethnobotanical Research
and Developement Enterprises, Inc. (ERDE).
The ascorbate
treatment involved dipping fresh truffle specimens in a 0.013 M
ascorbic acid solution for 3-5 min. prior to freezing in order
to help preserve color and flavor.
Ascorbic acid treated samples
were analyzed separately from untreated samples.
Prior to
chemical analysis for aroma volatiles, all specimens were
sectioned and slide-mounted in a 10% potassium hydroxide solution
for microscopic examination to determine approximate maturity.
Relative maturity was determined by taking a random count of at
least 25 spores from the slide mount of each specimen.
Immature
spores, which are hyaline, can be distinguished from mature spores
which are amber in color.
Four maturity ranges were established
depending on the percentage of mature spores in the sample:
1:0- 10%, class 2: 20 - 30%, class 3: 40 75%.
class
60%, and class 4: >
These four class ranges were selected because most of the
specimens examined fell into these maturity catagories.
Also,
maximizing differences between classes was desirable for
developing a staistical model of truffle maturity employing
discriminant analysis.
For the untreated samples, at least two
13
samples from each maturity range were analyzed for aroma
volatiles.
For the ascorbate treated samples, at least two
samples from the 40 - 60% maturity range were similarly analyzed.
Each truffle specimen was individually wrapped in aluminum foil,
labeled as to its treatment and maturity, and stored frozen at -10"
C in a sealed glass container until analyzed.
TRUFFLE SAMPLE PREPARATION.
When possible, a single specimen or
part of a large truffle was sampled.
However, if small, several
T. gibbosum specimens of the same maturity range were combined to
give a total sample weight of about 5 gms.
Samples were chopped
to fine pieces with a razor blade oh a glass surface and weighed
into a tared Pyrex screw-cap vial containing 4 gm. anhydrous
sodium sulfate.
A slurry was made of the sample by addition of 8
ml room temperature spring water.
A small Teflon-coated magnetic
stirrer was added to the vial before sealing it with a silicone
septum and cap.
The vial containing the truffle sample was then
ready for collection and concentration
of the aroma volatiles.
AROMA VOLATILES CONCENTRATION. Truffle aroma volatiles were
collected and concentrated by a modified entrainment system
similar to that originally used for dairy samples (Morgan and Day,
1965).
The system used for dairy products was a headspace
concentration technique where purified nitrogen carrier gas was
bubbled through a liquid sample, or a sample slurry in water and
14
sodium sulfate.
Aroma volatiles in the headspace were then
cold-trapped onto the front end of a gas chromatographic (GC)
column which was cooled by immersion in a Dewar of 2-methoxy
ethanol and dry ice.
The GC column was then reconnected to the
injection port, heated, carrier gas flow re-established, and GC
analysis conducted normally.
A somewhat similar method was used for the analysis of
canned corn volatiles (Boyko, et al., 1978), but aroma compounds
were concentrated in a stainless steel trap containing the porous
polymer, Porapak Q.
The advantage of using a porous polymer trap
is its preferred adsorption of organic compounds to
water, thus
providing an effective means of concentrating the sample aroma
volatiles without water.
The aroma volatiles were then
heat-desorbed from the Porapak Q trap and cold-trapped in a
stainless steel sample loop (Scanlan, et al. 1968) which was
cooled in dry ice.
The tightly-capped sample loop containing the
aroma volatiles was kept frozen until GC analysis.
It was then
connected to the inlet system of the GC, heated, and the sample
volatiles swept onto the GC column by the nitrogen carrier gas.
The volatiles concentration method finally adopted for the
truffle samples
was similar to that used for canned corn
volatiles (Boyko, et al., 1978) but with the following sampling
parameters:
1) the sample was immersed in a 45 - 550C water bath;
2) entrainment of volatiles was for 30 min. with a nitrogen flow
rate of 30 cc/min.;
and 3) the porous polymer trap was heated to
550C with a heat gun during sampling.
Modifications of the Boyko
15
method were the replacement of the stainless steel porous polymer
trap with a Pyrex glass trap and elution of the concentrated aroma
compounds with an organic solvent rather than heat (Jennings,
1978).
The Pyrex trap used was 10 cm X 4.0 mm I.D. and contained
a 5 cm segment of Porapak Q held in place with si lam* zed glass
wool plugs,- and the trap had a void volume of about 4 cm on one
end.
The trap was fitted with 1/4" I.D. brass Swage!ok fittings
with Teflon ferrules to form a gas-tight connection to the
entrainment assembly.
This trap was plumbed in line so that the
gas flow went through the packed end of the trap first.
After 30
min. of volatile entrainment, the trap was plumbed directly to the
nitrogen flow,
maintained at the same flow-rate, and heated at 55'
C for 25 min. to remove residual water from the column.
The trap
was then removed from the entrainment system, turned vertically so
that the void volume was at the top, and the volatiles eluted from
the column with 15 ml of spectro-grade dichloromethane (DCM, E.M.
Industries).
The sample in DCM was concentrated in a 125 ml
Kontes Kuderna-Dam'sh concentrator to about 2 ml.
A 10 ng spike
of n-tridecane (Sigma Chemical Co.) in spectro-grade trimethyl
pentane (TMP, Burdick and Jackson) was then added to each sample
as an internal standard.
Samples were concentrated further to a
100 ^il volume of TMP for GC analysis.
QUANTITATIVE ANALYSIS OF AROMA VOLATILES.
Truffle aroma volatiles
concentrated in TMP were analyzed by splitless-injection capillary
GC.
The instrument used was a Hewlett Packard (HP) model 5710A
16
equipped with a flame ionization detector (FID) and an HP 18740B
capillary inlet system.
The GC column used was a 50 m X 0.25mm
I.D. flexible fused silica capillary column coated with AT-1000
liquid phase (Alltech Assoc), a polyethylene glycol ester liquid
phase very similar in retention to Carbowax 20M.
temperatures were:
Operating
injection port 2000C; detector 250oC; and oven
temperature: isothermal at 70oC for a 4 min. hold, then
temperature programmed, at 40C/min. to 160oC with an 8 min. hold.
Flow rates were: 0.6 cc/min. ultrapure helium through the column,
and with helium make-up gas to the detector at 30 cc/min.;
detector gases were prepurified hydrogen at 30 cc/min. and
breathing quality air at 240 cc/min.
Detector response was
integrated and quantified against the n-tridecane internal
standard peak by an HP 3390A recording integrator interfaced to
the GC.
QUALITATIVE ANALYSIS OF AROMA VOLATILES.
Two methods were used to
identify the volatile compounds isolated from the truffles:
calculation of the Kova'ts GC retention index (R.I.) for each
compound and qualitative analysis by gas chromatography/mass
spectrometry/data system (GC/MS/DS).
Kova'ts R.I. for each
compound was calculated by comparison of peak retention data with
retention times of n-hydrocarbons from an external standard
(Hewlett Packard, Inc.) analyzed using the same GC operating
conditions as for the samples.
After all sample aroma extracts
had been analyzed quantitatively and qualitatively by GC, all T.
17
gibbosum sample extracts for each maturity range and ascorbate
treatment were combined in order to have sufficient sample for
qualitative analysis by GC/MS/DS.
The same column used for
quantitative anaysis was installed in a Finnigan model 4023
quadrupole GC/MS/DS with a model 4500 ion source.
Chromatographic
flow rates and temperatures were the same as those used for
quantitative analysis with the HP 5710A GC.
lonization was by
electron impact (El) at 70 eV with a current flow of 300
microamps.
Scan time was 1 sec.
Aroma volatiles were identified
from their mass spectra which were matched to known compound
spectra via the Finnigan Incos computerized search routine
accessing the National Bureau of Standards/Mational Institutes of
Health (NBS/NIH) mass spectral library of approximately 25,000
compounds.
Spectra were also manually compared with the published
Environmental Protection Agency (EPA)/NIH mass spectral reference
data base (Heller and Milne, 1978).
18
RESULTS AND DISCUSSION
IDENTIFICATION OF CONSTITUENTS FROM T. GIBBOSUM VOLATILES.
Results from the qualitative and quantitative analysis of the
truffle aroma volatiles of the four maturity ranges of T. gibbosum
are given in Table II.l.
Seven compounds were identified by their
mass spectra and identities of six of the seven compounds were
further substantiated by their Kovats R.I.
Table II.l lists the
calculated R.I. for each compound (Kova'ts, 1958) on the AT-1000
liquid phase as compared with published R.I. on Carbowax 20M
(Buttery, et al., 1982, Seifert and King, 1982).
Calculated R.I.
values were in close agreement with the published values that were
available.
COMPOSITION OF T. GIBBOSUM VOLATILES.
identified
Amounts of each compound
were calculated in ng/gm of sample, or parts per
billion (ppb).
The amounts of the 7 compounds were then summed to
obtain an estimated value for total volatiles.
This total most
likely does not represent all the volatile compounds actually
present.
Rather, these are the components which have been
isolated by this particular analytical method.
Since all samples
were analyzed under the same conditions, these components
represent some portion of the total volatile profile, and
therefore, are useful estimators for describing differences
between the truffle maturity classes.
Comparison of these totals
indicated very immature truffles from class 1 had the lowest
concentration of volatiles, but samples from the next maturity,
TABLE II.l.
Relative amounts of compounds identified as volatile aroma constituents
from samples of T. gibbosum at various stages of maturity.
Compound
Maturity class
Average maturity
# of samples
Kova'ts R. I.
CBWX20M
AT-1000
1
2.8%
n = 5
2
24.0%
n = 6
3
46.0%
n = 3
3Ta
57.0%
n = 5
4
82.0%
n = 2
%
%
%
%
%
n-hexanal
1104
1108
b
11.4
14.6
12.8
0.0
13.6
octan-3-one
1261
1250
c
0.0
11.7
4.1
2.8
4.8
n-hexanol
1347
1355
c
0.0
6.8
8.0
0.0
14.0
octan-3-ol
1388
1390
b
3.0
8.8
7.8
6.4
9.9
octadienal
1417
11.4
2.3
4.0
0.4
4.3
oct-l-en-3-ol
1448
1450
C
71.3
49.8
60.2
86.3
51.6
oct-2-en-l-ol
1610
1610
C
2.9
6.0
3.1
4.1
1.8
d
a
3T designates truffle samples of maturity class 3 treated with ascorbic acid
b
reference:
Buttery et al., 1982
c
reference:
Seifert and King, 1982
d
octadienal tentatively identified as E-2,4-octadienal from its mass spectrum;
published Kova'ts retention data for this compound unavailable
lO
20
class 2, had the highest concentration.
From this data, the
relative amounts of each compound in the total volatile profile
were calculated and are presented in Table II.1
The major volatile component was found to be oct-l-en-3-ol
which has been identified as the typical "mushroomy" aroma
component in many fungal species including the common commerial
mushroom, Agaricus bisporus ( Cronin and Ward, 1971, Piccardi and
Issenberg, 1973).
Recently, Wurzenberger and Grosch (1984)
demonstrated the formation of oct-l-en-3-ol from the
10-hydroperoxide isomer of linoleic acid, the major fatty acid of
the mushroom Psalliota bispora.
The fatty acid composition of T.
gibbosum is not known, but Al-Shabibi (1982) reported Iraqi
truffles to be rich in linoleic acid.
It seems feasible, then,
that the oct-l-en-3-ol found in J. gibbosum, might be formed from
autoxidation of linoleic acid or enzymatic oxidation catalyzed by
a hydroperoxide lyase in a
mushrooms.
manner similar to that demonstrated in
Also, a number of the other aroma volatiles isolated
from J. gibbosum have been identified as products of lipid
oxidation (Forss, 1972, and Badings, 1970), a relatively common
source of flavor compounds in foods.
The concentration of oct-l-en-3-ol in the untreated
truffles ranged from 200 ppb in immature, class 1 truffles to 4000
ppb in samples of class 2 maturity.
However, even though this
range represents a 20-fold difference in concentration, the
relative amount of this compound in the total volatile profile was
between 50 - 70% for all maturity classes.
Even though other
compounds did not constitute as large a percentage of the total
21
volatiles as did oct-l-en-3-ol, they contributed more
significantly in describing differences in the volatile
composition during truffle maturation.
Figure II.1 illustrates
the change in the relative amount of n-hexanol, a minor component
of the total volatile profile, with increasing truffle maturity.
The linear regression coefficients were calculated by the method
of least squares, and the plotted line was fitted by eye.
A DESCRIPTIVE MODEL OF T. GIBBOSUM MATURITY BASED ON CHANGES IN
AROMA VOLATILES.
The relationship between production of volatiles
and truffle maturity was explored statistically with the data
obtained from qualitative and quantitative analysis of the
untreated T. gibbosum samples.
Amounts of all compounds in ppb
and total amounts for each sample in each maturity class, 1-4
were entered into an IBM personal computer.
Number Cruncher
Statistical System (NCSS) software (Hintz, 1984) was utilized for
discriminant analysis.
A stepwise variable addition process was
used until a satisfactory model was obtained which described a
significant relationship between maturity class (dependent
variable) and amounts of volatile compounds (independent
variables).
Variables found significant to the model describing
maturity were (in order of significance): n-hexanol, n-hexanal,
octan-3-one, octan-3-ol, the total and octadienal.
Computer
output for the model gave a matrix which included the discriminant
constants and coefficients for each variable in each class.
The
significance (ot.= 0.05) of each variable in the equations was
determined by an F-ratio calculation.
Variables which were found
FIGURE II.1.
Change in the relative amount of n-hexanol in the aroma volatiles
of T. gibbosum with increasing truffle maturity.
100
80
-
a:
60
40
-
*«
20
0
5
10
15
20
% n-HEXANOL
ro
23
to be insignificant to the model were oct-l-en-3-ol and
oct-2-en-l-ol.
The lack of significance of oct-l-en-3-ol to the
model is particularily interesting since it is the compound of
greatest relative abundance.
Relative amounts of this compound
indicate it is produced in early maturity and is therefore not a
useful indicator of the maturation process.
Appropriateness of the model was interpreted using the
Wilk's Lambda coefficient - a multivariate extension of R-Squared.
Wilk's Lambda values range from 0 to 1 where values near 1 imply
low predictability and values near 0 imply high predictability
(Hintz, 1984).
was 0.017.
The Wilk's Lambda value obtained using this model
Another calculated measure of model performance was
the apparent error rate (APER), which is the ratio of observations
that are misclassified to those that are correctly classified by
the model classification function (Johnson and Wichern, 1982).
The APER for this model was 3/14 or a 21% error rate in
classification.
The APER for this sampling however, is somewhat
inappropriate because the error rate is based on a very small
sample size, n = 17.
Samples which were misclassified were two
class 3 maturity samples which were predicted to be in class 1,
and one class 4 sample predicted to be in class 2.
Looking at the
amounts of individual compounds in these misclassified samples,
they do appear to have volatile profiles like less mature samples.
This model is by no means complete; however, it is a begining and
the additional classification
utility.
of more samples will enhance its
24
A COMPARISON OF AROMA VOLATILES FROM ASCORBIC ACID TREATED
TRUFFLES WITH UNTREATED SAMPLES.
Results of quantitative analysis
for ascorbate treated truffle samples were calculated as for
untreated samples, and the relative composition data is reported
in Table II.l.
Comparison of the results for treated and
untreated samples from the same maturity class, 3, indicate
several differences.
No trace of n-hexanal or n-hexanol was
detected in the treated samples and the amounts of octan-3-one and
octadienal were considerably reduced.
However, a greater amount
of oct-l-en-3-ol was found. .The amounts of each compound and the
amounts of the total volatiles for each sample were entered into
the discriminant analysis model, previously developed from the
data for untreated samples, to test the classification function.
The model misclassified all five samples into Class 1 or Class 2
maturities.
It is not known if the the ascorbic acid treatment
changed the components of the truffle aroma by inhibition of
oxidation, or if the treatment instead, caused changes in spore
coloration so that the samples were misclassified as to their
maturity.
It appears that this latter consideration is a
posibility.
Apparently, exposure of the truffle samples to higher
ascorbic acid concentrations, or to an exposure time longer than 5
min. will cause specimen shriveling and a yellow-brown
discoloration (A.B. Walters, personal communication).
It is not
known if this coloration occurs to any extent at the present level
of treatment.
Since mature T. gibbosum spores are amber in color
and immature spores colorless, a miscalculation of maturity might
occur if the asorbic acid treatment colored immature spores.
25
CONCLUSIONS
Results of the qualitative analysis of T. gibbosum
volatiles indicated the presence of 7 compounds:
oct-l-en-3-ol,
bct-2-en-l-ol, n-hexanal, n-hexanol, octan-3-one, octan-3-ol and
octadienal.
The latter 5 components, and also the sum of the
amounts of all components were found to be significant variables
to describe a model for truffle maturity.
Results of this
investigation indicate that truffle odors increase in intensity
and complexity with increasing spore maturation.
Comparison of
volatiles from ascorbic acid treated and untreated T. gibbosum
specimens shows an apparent change in the composition
treatment.
with
However, it is not known if this change is actually in
the composition of the volatiles, or is due to a change in the
spore coloration so that the maturity of the samples is
misclassified.
26
REFERENCES
Al-Shabibi, M.M.A., S.J. Toma and B.A. Haddad. 1982. Studies
on Iraqi Truffles. I. Proximate Analysis and Characterization of Lipids. Can. Inst. Food Sci. Technol. J. 15:200.
Badings, H.T. 1970. Cold-Storage Defects in Butter and Their
Relation to the Autoxidation of Unsaturated Fatty Acids.
H. Veenman & Zonen N.V., Wageningen, Nederlands. 118.
Boyko, A.L., M.E. Morgan, L.M. Libbey. 1978. Porous Polymer
Trapping for the GC/MS Analysis of Vegetable Flavors.
Analysis of Foods and Beverages Headspace Techniques. (G.
Charalambous, Ed.). Academic Press Ltd., New York. 57.
Buttery, R.G., J.A. Kamm, L.C. Ling. 1982. Volatile Components
of Alfalfa Flowers and Pods. J. Agric. Food Chem. 30:739.
Cronin, D.A., and M.K. Ward. 1971. The Characterisation of Some
Mushroom Volatiles. J. Sci. Fd. Agric. 22:477.
Forss, D.A. 1972. Odor and Flavor Compounds from Lipids.
Progress in the Chemistry of Fats and Other Lipids. (R.T.
Hoi man, Ed.). Pergamon Press. New York.13(4):287.
Heller, S.R. and G.W.A. Milne. 1978. EPA/NIH Mass Spectral Data
Base. Dept. of Commerce, U.S. Government, Washington, D.C.
Hintze, J.L. 1984. Number Cruncher Statistical System Version
4.1. Kaysville, Utah.
Jennings, W.G. 1978. Gas Chromatography with Glass Capillary
Columns. Academic Press. San Francisco. 117.
Johnson, R.A. and D.W. Wichern. 1982. Applied Multivariate
Statistical Analysis. Prentice Hall. Englewood Cliffs, NJ.
Kova'ts, E. von, W. Simon, and E. Heilbronner. 1958. Programmgesteverter Gas-Chromatograph zur Praparativen Trennung
von Gemischen organischer Verbindungen. He!. Chim. Acta
32:275.
Maser, C, J.M. Trappe, and R.A. Nussbaum. 1978. Fungal-Small
Mammal Interrelationships with Emphasis, on Oregon
Coniferous Forests. Ecology. 59(4):799.
27
Morgan, M.E. and E.A. Day. 1965. Simple On-Column Trapping
Procedure for Gas Chromatographi'c Analysis of Flavor
Volatiles. J. Dairy Sci. 48:1382.
Picardi, S.M. and P. Issenberg. 1973. Investigation of Some
Volatile Constituents of Mushrooms (Agaricus
bisporus): Changes which Occur during Heating. J. Agric.
Food Chem. 21(6):959.
Scanlan, R.A., R.G. Arnold and R.C. Lindsay. 1968. Collecting
and Transferring Packed-Column Gas Chromatographi'c
Fractions to Capillary Columns for Fast-Scan Mass Spectral
Analysis. J. Gas Chromatog. 6:372.
Seifert, R.M. and D.A. King, Jr. 1982. Identification of Some
Volatile Constituents of Asperqillus clavatus. J.
Agric. Food Chem. 30(4):787.
Wurzenberger, M. and W. Grosch. 1984. The Formation of
l-0cten-3-ol from the 10-Hydroperoxide Isomer of Linoleic
Acid by a Hydroperoxide Lyase in Mushrooms (Psalliota
bispora). Biochim. Biophys. Acta. 794:25.
28
Chapter III.
A Comparative Sensory Analysis of Public Response
to the Aromas of Tuber gibbosum
and Three Other Native Oregon Truffles
29
ABSTRACT
The aromas of three native Oregon truffle species,
generally only known to mycologists, were rated and compared with
Tuber gibbosum, a species recently recognized for its culinary
appeal.
Results show that the aroma of Gautieria monticola was
preferred to that of T. gibbosum, and therefore, shows consumer
appeal which may warrant further investigation.
For that segment
of the population that disliked the truffle aromas, females were
found to dislike them more intensely than males.
This response
should be investigated further considering that a similar response
has been reported for odoriferous steroids which have been
isolated from European truffle species.
30
INTRODUCTION
Truffles are an uncommon and frequently unknown food
commodity in this country, but use of truffles in Europe and Asia
has been recorded since the era of ancient Greek and Roman
civilizations (Wasson and Wasson, 1957).
Truffles are hypogeous
fungi, several species of which are both highly prized and priced
as a gourmet food because of their unique and enticing aromas.
Recently, a native Oregon truffle species has won culinary appeal
and is being commercially exploited.
There are a number of other
native species which may have consumer appeal but remain virtually
unknown to the public.
For this reason, hedonic response to the
aroma of T. gibbosum was compared to the aromas of three other
native truffles.
Truffles produce aromas to attract animals since the
fungus is dependent on mycophagy for spore dispersal (Maser, et
al., 1978).
Only a few investigators have examined truffle odors
to elucidate the compounds responsible for these aromas.
Recently, Claus, et al (1981) isolated and identified two
odoriferous steroidal compounds from species of European truffles.
These compounds, 5o(.-androst-16-en-3-one (androstenone), and the
corresponding alcohol, are also known to be porcine sex pheremones
(Melrose, et al, 1971, Claus and Hoffmann, 1971).
Human response
to androstenone has been studied because this compound has been
identified as an unpleasant odor in pork (Griffith and Patterson,
1970).
At near threshold concentrations, androstenone was
described as "sweet, fruity and perfume-like", while at higher
31
concentrations, it was described as "sweaty, animal, urinous or
ammonia-like".
These investigators also found that women were
more sensitive to the odor, and that they found it significantly
more unpleasant than did men.
Recently, Koelega (1980) confirmed
this difference between female and male hedonic response and
sensitivity to androstenone.
The contribution of the steroids to the overall aroma of
the European truffles has not been reported, nor has the hedonic
response to these truffle aromas been investigated.
The truffle
species used for this study have not been chemically analyzed for
steroidal compounds.
However, it seemed worthwhile to
statistically analyze the results of this sensory study to
determine if there might be male/female differences in hedonic
scoring of the truffle aromas evaluated.
32
MATERIALS AND METHODS
TRUFFLE SAMPLE PREPARATION.
Four truffle species were selected
for this study because of availablity and distinct difference in
aroma between species.
Species selected were T. gibbosum, Picoa
carthusiana, Rhizopogon parksii and Gautieria monticola.
Several mature specimens of each species, raw, fresh or
frozen, were grated with a fine stainless steel grater and a 1
tablespoon (approximately 3 gm) subsample placed in a 7 oz. clear
stemmed wine glass.
One teaspoon (5 gms) of table salt and 1
tablespoon (5 ml) hot tap water (120 F) was added to each sample.
Samples were mixed with a spoon and then covered with a snug
aluminum foil cap.
Four glasses of each truffle slurry were
labeled with a number, 1-4 and corresponding specimen name, and
displayed on a table with instructions on the aroma evaluation
procedure.
SENSORY TESTING.
This truffle aroma evaluation was conducted as
part of a display on truffles presented by members of the North
American Truffling Society (NATS) at the Oregon Mycological
Society (OMS) Mushroom Show in Portland, Oregon.
Individuals
approaching the NATS truffle display area were asked if they would
like to evaluate truffle aroma.
Subjects were instructed on
sniffing procedure, how to fill out the ballot, and asked to
comment on the aromas if they wished.
Instructions for aroma
evaluation posted with the samples were: 1) swirl glass; 2) remove
cover; 3) sniff; 4) rate aroma desirability for each according to
33
this 9-point scale (Peryam and Pilgrim, 1957):
1
2
3
4
5
6
7
8
9
dislike extremely
dislike very much
dislike moderately
dislike slightly
neither like nor dislike
like slightly
like moderatly
like very much
like extremely
Information requested on the ballot was: subjects' sex; favorite
sample number; rating for the overall desirability of each sample,
based on the 9-point hedonic scale; and comments on the aromas.
STATISTICAL ANALYSIS.
After sensory testing, ballots were
collected, interpreted, and the data entered into an IBM personal
computer for data analysis utilizing Number Cruncher Statistical
System software (Hintze, 1984).
Significant differences between
male/female responses were determined using a 1-way analysis of
variance (ANOVA), and calculation of Fisher's Least Significant
Difference (LSD).
F-ratios from the ANOVA table were compared
with published values to confirm significance (Snedecor and
Cochran, 1980).
34
RESULTS AND DISCUSSION
Statistical analysis of the truffle aroma evaluation data
for favorite sample is given in Table III.l.
Male and female
populations tend to choose the same sample with similar frequency.
Analysis of the hedom'c scores given for these samples shows the
same trend with the highest scores being given to the aroma
preferred by the most indiviuals and the lowest scores to the
sample preferred by the least number of people.
Figure III.l. illustrates the distribution of hedom'c
scores given to each truffle aroma by male and female populations.
Relative frequency distributions for all four truffle species
appear to be bimodal:
aromas.
people either liking or disliking the
Land and Piggot (1980) discussed the statistical
treatment of data displaying diverse hedom'c responses, and
suggested a criterion for determining if data were um'modal or
bimodal.
They suggested condensing the data from the 9-point
hedom'c scale into three equal categories, pleasant, neutral and
unpleasant, and expressing the responses in each catagory as a
percent of the total.
Then, if the value obtained from the
product of [% pleasant responses)[% unpleasant responses) was
greater than 400, the distribution should be considered bimodal
and analyzed accordingly.
Their suggestion for treatment of this
type of data was to select individuals from the initial population
representative of the sub-population and ana-lyze the reponse data
separately.
Hedom'c response data from the truffle aroma
evaluation was categorized, and the bimodal criterion values
TABLE III.l.
Percentage of population indicating one of four truffles as
favorite.
Order of preference
(all populations)
Truffle
Population
male
n = 133
female
n = 160
total
n = 293
1.
Gautieria monticola
57.9 %
50.0 %
53.6 %
2.
Picoa carthusiana
20.3 %
25.6 %
23.2 %
3.
Tuber gibbosum
16.5 %
16.3 %
16.4 %
4.
Rhizopogon parksii
5.3 %
8.1 %
6.8 %
CO
tn
FIGURE III.l.
30 .
The relative frequency distribution of hedonic scores
given for the aroma of four truffle species by males
and females.
MAICC
FEMALES = 1
MALESs == "|
A. CAUTIERIA MONTICOLA
L
25
R
Z0
u
15
M
10
5
h
25
;:|:
!;;•
1
2
3
4
5
6
-
20
-
•llllllli, 1
B. PICOA CARTHUSIAHA
30
R
1
'■•l
■
15
-
10
-
a
HEDONIC SCORE
30
.
25
-
25
20
-
20
IS
-
IS
10
30 I
10
X;
5
12
IHI
3
4
5
6
HEDONIC SCORE
7
8
1i
9
Hi
123456789
HEDONIC SCORE
9
c. TUBER GIBBOSUH
1 =? »
iNik
5
i _.
7
;X
0. RII1Z0P0G0N PARKSU
1
^ 1
1
•'•••
11 i
-
5
ll 1 Li
12
IIULiiiil
3
4
5
6
HEDONIC SCORE
7
8
9
en
37
calculated for male, female and the total population for each
truffle.
In all cases,
the (% unpleasant)(% pleasant) calculated
value far exceeded 400, indicating hedonic frequency distributions
were indeed, bimodal.
Considering this diversity in hedonic
responses, comparisons of mean hedonic scores between males and
females were made for two subsets of the hedonic scale: scores 4:
5:
the dislike range and scores > 5:
the like range, as well as
for the entire scale.
Figure III.2. displays the mean hedonic score given to
each truffle aroma by male, female, and both populations for each
range of hedonic scores.
The number of individuals included
within each hedonic score range (expressed as percent of the
population) is included at the top of Figure 111.2. A and B.
In
the dislike range, significant differences were found between mean
hedonic scores given by males and females for three of the four
truffle species.
In the like range, no significant differences
were found between male and female responses.
For the entire
population, considering the full hedonic scale, significant
differences were found in mean hedonic score between three of the
four truffle species.
This indicates a real difference in degree
of liking between the truffle aromas.
Table 111.2. lists the most frequently given descriptors
for each of the truffle aromas.
Less than 25% of the participants
in the study described the aromas, but the resulting list
indicates some similarities and differences in the truffle aromas.
FIGURE III.2.
Mean hedonic scores given for each of four truffle aromas
by male, female, and both populations for three ranges of
hedonic scores. MALES = z
FEMALES =|
B0TH =|
TRUFFLE SAMPLE » 1=G. monticola; 2=P. carthusiana; 3=T. gibbosum; 4=R. parksii
S of
Hale I9.0X
total Feoale 19.7
n • 273 Both 19.4
37.21
45.4
41.8
39.71
49.3
45.1
~r
81.01
82.2
81.7
A. HEDONIC SCORES <5:
DISLIKE*
Hale 86.8»
Female 82.9
Both 84.6
66.U
58.9
61.5
-r
24.81
23.1
25.6
B. HEDONIC SCORES 2 5:
~
E
1
1
:
:?
^ %
5
::y
l•1
^ :••?
_ •:•?
^ ■:?
^ ??
:?:
^
C. FULL SCALE OF HEDONIC
SCORES
LIKE*
—
§
~
- i*
!;$
~ :$
*
^
—
1
^
:X;
_ •■§
^ ij-li
■$•
1 E
1 =rl
:3
=
- :|
~ ■'?
■■S*
— :•?
—
—
~
z
6
r
5 1
p z
"
1 z
1 31
m ii
2
^
2
3
TRUFFII SAMPII f
70.21
58.7
63.7
Z
:::"
—
"~
::•:
Z.
1 r
1 =
3
TRUFFLE SAMPLE #
2
3
TRUFFLE SAMPLE #
* Hedonic score of S: neither like nor dislike, scale Midpoint, Is Included Kith both like and dislike scores.
** Indicates Bean hedonic scores for males and females found to be significantly different (p <0.05).
a.b.c
Indicates nean hedonic scores between truffle species to be significantly different (p<0.05) for both nale
and female population (n ■ 273).
CO
00
TABLE 111.2.
Aroma descriptors given for each of four truffle species.
Truffle:
G. monticola
P. carthusiana
T. gibbosum
R. parksii
sweet
apple
cheese
earthy
mushrooms
cheese
nuts
musty
cake
nuts
fungus
spoiled wine
cheese
coconut
potatoes
stale nuts
buttery
spoiled prunes
bread
saw dust
coconut
wine
fruity
varnish
marmelade
oysters
wet dog
bleach
CO
10
40
CONCLUSIONS
Results of the truffle aroma study indicate 6. monticola
to be the preferred sample;
it was given a mean hedonic score of
6.8, indicating it was liked slightly to moderately.
Considering
the large percentage of individuals who rated this truffle aroma
very favorably, 84.6% of the total population giving a mean score
of 7.5, this truffle warrants further investigation and
utilization studies.
Relative frequency distributions for the hedonic scores
indicated bimodal distributions for all responses:
either liked the truffle aromas or disliked them.
individuals
For individuals
liking the truffle aromas, there was no significant difference in
mean hedonic scores between males and females.
However, for those
disliking the truffle aromas, females rated three of the four
truffles significantly lower than did males.
Considering this
difference in female/male degree of dislike for these aromas, it
may be worthwhile to analyze these species for the presence of
steroidal compounds, since similar hedonic responses have been
reported for androstenone, an odoriferous steroid recently
identified in European truffles.
41
REFERENCES
Claus, R. and B. Hoffmann. 1971. Determination of A.16-Steroids
in Boars. Acta Endroc. Suppl. 155:70.
Claus, R., H.O. Hoppen and H. Karg. 1981. The Secret of
Truffles: A Steroidal Pheremone? Experientia. 37:1178.
Griffths, N.M. and R.L.S. Patterson. 1970. Human Olfactory
Response to 5oL-androst-16-en-3-one - Principal Component
of Boar Taint. J. Sci. Fd. Agric. 21:4.
Hintze, J.L. 1984. Number Cruncher Statistical System Version
4.1. Kaysville, Utah.
Koelega, H.S. 1980. Preference for and Sensitivity to the Odours
of Androstenone and Musk. 01 faction and Taste VII. (H. van
der Starre, Ed.). IRL Press Ltd. Washington, D.C. 436.
Land, D.G. and Piggot, J.R. 1980. A Criterion for Identifying
Stimuli with Diverse Hedonic Responses. 01 faction and
Taste VII. (H. van der Starre, Ed.). IRL Press Ltd.
Washington, D.C. 389.
Maser, C, J.M. Trappe, and R.A. Nussbaum. 1978. Fungal-Small
Mammal Interrelationships with Emphasis on Oregon
Coniferous Forests. Ecology. 59(4):799.
Melrose, D.R., H.C.B. Reed, and R.L.S. Patterson. 1971.
Androgen Steroids Associated with Boar Odour as an Aid to
the Detection of Oestrus in Pig Artificial Insemination.
Br. Vet. J. 127:497.
Peryam, D.R. and F.J. Pilgrim. 1957. Hedonic Scale Method for
Measuring Food Preferences. Food Technol. 11:9.
Snedecor, G.W. and W.G. Cochran. 1980. Statistical Methods. The
Iowa State University Press. Ames, Iowa. 507.
Wasson, V.P. and R.G. Wasson. 1957. Mushrooms, Russia and
History. Pantheon Books. New York. 166.
42
BIBLIOGRAPHY
Ahmed, A.A., M.A. Mohamed, and M.A. Hami. 1981. Libyan Truffles
"Terfezia bouderi Chatin:" Chemical Composition and
Toxicity. J. Fd. Sci. 46:927
Al-Delaimy, K.S. 1977. Protein and Amino Acid Composition of
Truffle. Can. Inst. Food Sci. Techno!. J. 10(3):221.
Al-Shabibi, M.M.A., S.J. Toma and B.A. Haddad. 1982. Studies
on Iraqi Truffles. I. Proximate Analysis and Characterization of Lipids. Can. Inst. Food Sci. Techno!. J. 15:200.
Andreotti, R. and Casoli, U. 1968. Composition Chimique et
Technologie de Conservation de !a Truffe. Congr. Int.
Truffe - Spolete.
Badings, H.T. 1970. Cold-Storage Defects in Butter and Their
Relation to the Autoxidation of Unsaturated Fatty Acids.
H. Veenman & Zonen N.V., Wageningen, Nederlands. 118.
Benedict, R.G. and D.E. Stuntz. 1975. Mushroom Odors. Pacific
Search. 9:42.
Boyko, A.L., M.E. Morgan, and L.M. Libbey. 1978. Porous Polymer
Trapping for GC/MS Analysis of Vegetable Flavors.
Analysis of Foods and Beverages Headspace Techniques. (G.
Charalambous, Ed.). Academic Press Ltd., New York. 57.
Brooksbank, B.W.L. 1962. Urinary Excretion of Androst-16-en3-ol Levels in Normal Subjects, and Effects of Treatment
with Trophic Hormones. J. Endocrin. 24:435.
Brooksbank, B.W.L., R. Brown and J.A. Gustavsson. 1974. The
Detection of 5 o(-androst-16-en-3o(-o! in Male Axillary
Sweat. J. Endocr. 30:864.
Burdsall, H.H. Jr. 1965. Operculate Asci and Puffing of Ascospores in Geopora (Tuberales). Mycologia. 57:485.
Buttery, R.G., J.A. Kamm, L.C. Ling. 1982. Volatile Components
of Alfalfa Flowers and Pods. J. Agric. Food Chem. 30:739.
Claus, R. and B. Hoffmann. 1971. Determination of A.16-Steroids
in Boars. Acta. Endocr. Suppl. 155:70.
Claus, R., H.O. Hoppen and H. Karg. 1981. The Secret of
Truffles: A Steroidal Pheromone? Experientia. 37:1178.
43
Collins, R.P., and A.F. Halim. 1971. Production of Monoterpenes
by the Filamentus Fungus Leratocystus variispora.
Llyodia 33:481
1972. An Analysis of the Odorous
Constituents produced by Various Species of Phellinus.
Can. J. Microbiol. 18:65.
Cronin, D.A., and M.K. Ward. 1971. The Characterisation of Some
Mushroom Volatiles. J. Sci. Fd. Agric. 22:477.
Del mas, J. 1978. Nutritive and Gastronomic Value of the
Truffle. Biology and Cultivation of Edible Mushrooms. (S.
T. Chang and W.A. Hayes, Eds.). Academic Press. New York.
677.
Fiecchi, A., M. Galli Kienle and A. Scala. 1967. Bis-methylthio
-methane, an Odorous Substance from White Truffle,
Tuber Magnatum pico. Tetrahedron Letters 18:1682
Fogel, R.D. 1976. Ecological Studies of Hypogeous Fungi. II.
Sporocarp Phenology in a Western Oregon Douglas Fir Stand.
Can. J. of Bot. 54:1152.
Fogel, R. and J.M. Trappe. 1978. Fungus Consumption (Mycophagy)
by Small Animals. Northwest Sci. 52:1.
Forss, D.A. 1972. Odor and Flavor Compounds from Lipids.
Progress in the Chemistry of Fats and Other Lipids. (R.T.
Holman, Ed.). Pergamon Press. New York. 13(4):287.
Gilbert, M.E.J. (1932) Osmologie Mycologique. Extrail du
Bulletin de la Societe Mycologique de France. 48(3):241.
Gray, W.D. 1970. The Use of Fungi as Food and in Food
Processing. Buttersworth and Co. Publishing Ltd. London.
Griffths, N.M. and R.L.S. Patterson. 1970. Human Olfactory
Response to 5<*-androst-16-en-3-one - Principal Component
of Boar Taint. J. Sci. Fd. Agric. 21:4.
Halim, A.F., J.A. Narciso, and R.P. Collins. 1975. Odorous
Constituents of PeniciIlium decumbens. Mycologia
67:1158.
Harper, R., E.C. Bate-Smith and D.G. Land. 1968. The Odours of
Fungi. Odour Description and Odour Classification.
American Elsevier Pub. Co., Inc. New York. 50.
44
Hacskaylo, E., J. Palmer, and J. Vozzo. 1965. Effect of
Temperature on Growth and Respiration of Ectomycorrhizae
- Their Ecology and Physiology. Mycologia. 57:748.
Heller, S.R. and G.W.A. Milne. 1978. EPA/NIH Mass Spectral Data
Base. Dept. of Commerce, U.S. Government, Washington, D.C.
Heim, R. 1957. L'Odeur. Son Importance Taxinomique.
Champignons. N. Boubee et Cie. Paris. 141.
Hintze, J.L. 1984. Number Cruncher Statistical System Version
4.1. Kaysville, Utah.
Howard, W.E., R.E. Marsh and R.E. Cole. 1968. Food Detection by
Deer Mice Using Olfactory Rather than Visual Cues. Animal
Behavior. 16:13.
Hutchinson, S.A. 1971. Biological Activity of Volatile Fungal
Metabolites. Tr. Br. Mycol. Soc. 57(2)185.
Jennings, W.G. 1978. Gas Chromatography with Glass Capillary
Columns. Academic Press. San Francisco. 117.
Josserand, M. 1952. Les Proprietes Organoleptiques.
Encyclopaedia Mycologique. (P. Chevalier, Ed.). Paris.
Johnson, R.A. and D.W. Wichern. 1982. Applied Multivariate
Statistical Analysis. Prentice Hall. Englewood Cliffs, NJ.
Kirk-Smith, M. and D.A. Booth. 1978. Human Social Attitudes
Affected by Androstenol. Res. Commun. Psycho!. Psychiat.
Behav. 3:379.
Koelega, H.S. 1980. Prefernce for and Sensitivity to the Odours
of Androstenone and Musk. 01 faction and Taste VII. (H. van
der Starre, Ed.). IRL Press Ltd. Washington, D.C. 436.
Kovats, E. von, W. Simon, and E. Heilbronner. 1958. Programmgesteverter Gas-Chromatograph zur Praparativen Trennung
von Gemischen organischer Verbindungen. He!. Chim. Acta
32:275.
Land, D.G. and Piggot, J.R. 1980. A Criterion for Identifying
Stimuli with Diverse Hedom'c Responses. 01 faction and
Taste VII. (H. van der Starre, Ed.). IRL Press Ltd.
Washington, D.C. 389.
Maga, J.A. 1976. The Potential of Certain Fungi as Sources for
45
Natural Flavor Compounds. Chemical Senses and Flavor.
2:255.
Maser, C, J.M. Trappe, and R.A. Nussbaum. 1978. Fungal-Small
Mammal Interrelationships with Emphasis on Oregon
Coniferous Forests. Ecology. 59(4):799.
Melrose, D.R., H.C.B. Reed, and R.L.S. Patterson. 1971.
Androgen Steroids Associated with Boar Odour as an Aid to
the Detection of Oestrus in Pig Artificial Insemination.
Br. Vet. J. 127:497.
Morgan, M.E. and E.A. Day. 1965. Simple On-Column Trapping
Procedure for Gas Chromatographic Analysis of Flavor
Volatiles. J. Dairy Sci. 48:1382.
Morita, K. and S. Kobayashi. 1966. Isolation and Synthesis of
Lenthionine, an Odorous Substance of Shiitake, an
Edible Mushroom. Tetrahedron Lett. 6:573.
Peryam, D.R. and F.J. Pilgrim. 1957. Hedonic Scale Method for
Measuring Food Preferences. Food Technol. 11:9.
Picardi, S.M. and P. Issenberg. 1973. Investigation of Some
Volatile Constituents of Mushrooms (Agaricus
bisporus): Changes which Occur during Heating. J. Agric.
Food Chem. 21(6):959.
Ruokonen, A. 1973. Free and Sulphate-Conjugated 16-Unsaturated
C19 Steroids in Human Testis Tissue. Biochim. Biophys.
Acta. 316:251.
Scanlan, R.A., R.G. Arnold and R.C. Lindsay. 1968. Collecting
and Transferring Packed-Column Gas Chromatographic
Fractions to Capillary Columns for Fast-Scan Mass Spectral
Analysis. J. Gas Chromatog. 6:372.
Seifert, R.M. and D.A. King, Jr. 1982. Identification of Some
Volatile Constituents of Aspergillus clavatus. J.
Agric. Food Chem. 30(4):787^
Snedecor, G.W. and W.G. Cochran. 1980. Statistical Methods. The
Iowa State University Press. Ames, Iowa. 507.
Stark, N. 1972. Nutrient Cycling Pathways and Litter Fungi.
BioSci. 22:355.
Thomas, A.F. 1973. An Analysis of the Flavor of the Dried
Mushroom, Boletus edulis. J. Agr. Food Chem. 21(6):955.
46
Trappe, J.M. 1977. Biogeography of Hypogeous Fungi: Trees,
Mammals, and Continental Drift. 2nd Internet. Mycol. Cong.
Abstr. 675.
Trappe, J.M., and R.D. Fogel. 1977. Ecosystem Functions of
• Mycorrhizae. The Belowground Symposium: a Synthesis of
Plant-associated Processes. (J.K. Marshall, Ed.). Colo.
State Univ. Range Sci. Dep. Sci. Ser. 26:205.
Trappe, J.M. and C. Maser. 1976. Germination of the Spores of
Glomus macrocarpus (Endogenaceae) after Passage
through a Rodent Digestive Tract. Mycologia. 68:433.
1977. Ectomycorrhizal Fungi:
Interactions of Mushrooms and Truffles with Beasts and
Trees. Mushrooms and man, an Interdisciplinary approach
to Mycology. (A.B. Walters, Ed.). Forest Service, U.S.D.A.
165.
Trione, E.J. 1977. The Nutritional Value of Mushrooms.
Mushrooms and Man, an Interdisciplinary Approach to
Mycology. (A.B. Walters, Ed.) Forest Service, U.S.D.A.
183.
Wasson, V.P. and R.G. Wasson. 1957. Mushrooms, Russia and
History. Pantheon Books. New York. 166.
Wurzenberger, M. and W. Grosch. 1984. The Formation of
l-0cten-3-ol from the 10-Hydroperoxide Isomer of Linoleic
Acid by a Hydroperoxide Lyase in Mushrooms (Psalliota
bispora). Biochim. Biophys. Acta. 794:25.
Yajima, I., T. Yanai, M. Nakamura, H. Sakakibara and K.
Hayashi. 1981. Volatile Flavor Compounds of Matsutake Tricholoma matsutake (Ito et Imai) Sing.-. Agric. Biol.
Chem. 45(2):377.