_1
IO“

  

(9—)

 

\IWIWIHHNIlMWlHHHlWIWNW(\INIHHHI

    

C0010?

 

LIERARY
Michigan State
University

 

 

 

This is to certify that the

dissertation entitled

A Relationship between Carbohydrates and
Freezing Survival in Winter Barley

presented by

David Palmer Livingston III

has been accepted towards fulfillment
of the requirements for

PhD degree in Crop and Soil Science

 

gzéiw/OWJZQ ,.

Major professor

Date 25 April, 1988

 

.MS U is an Affirmative Action/Equal Opportunity Institution 0-12771

 

 

MSU
LIBRARIES
-_

 

RETURNING MATERIALS:
Place in book drop to
remove this checkout from
your record. FINES will
be charged if book is
returned after the date
stamped below.

 

 

 

 

 

 

A RELATIONSHIP BETWEEN CARBOHYDRATES
AND FREEZING SURVIVAL
IN WINTER BARLEY

By

David Palmer Livingston III

A DISSERTATION

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY

Department of Crop and Soil Science

1988

ABSTRACT
A RELATIONSHIP BETWEEN CARBOHYDRATES
AND FREEZING SURVIVAL
IN WINTER BARLEY
By

David Palmer Livingston III

Past studies have shown that sugar distribution within
cells immediately before the onset of stress is critical to
a realistic understanding of cryoprotection related to
carbohydrates. Because spatial distribution of
carbohydrates in cells are difficult to determine,
compositional distributions were calculated from total
carbohydrate amounts and related to freezing stress. Three
barley (Hordeum vulgare) cultivars, differing in
carbohydrate composition and in response to a controlled
freezing test, were hardened under six light levels. At
each level plants were freeze-tested and total crown sugars
measured by water/ethanol extraction and High Pressure
Liquid Chromatography. The log of a sugar to fructan ratio
was linearly related to total carbohydrates. Experimental
variability was eliminated by calculating simple sugar and
fructan amounts at varying carbohydrate levels from the
equations for the lines. The relationship between kill
temperature and total carbohydrate was different for the
three cultivars, suggesting a difference in sugar
availability for cryoprotection. At approximately 100 mg

carbohydrate/g dry wt. all three cultivars were killed at

approximately -10‘C, the temperature where adhesive stress
was reported to begin. When hardened under similar low
light conditions, however. the less hardy cultivar Durani
accumulated about one half as much carbohydrate as the other
two cultivars. F1, F2 families and backcrosses (BC) of the
cultivars were hardened under low and high light and freeze
tested. Under low light a dominant effect, causing death in,
the freeze test, was seen in the cross of Durani with the
hardy parent Dictoo. A 13:3 (deadzalive) ratio in the F2.
1:1 in BC; and 1:0 (deadzalive) in BC: implied that the
effect was the result of a simply inherited trait controlled
by two genes with epistatic action. Carbohydrate data for
the parents and F1 progeny at low light suggest that the

trait may be related to supply or demand of sugar or both.

TO

Maria, Jesse and Rachel

ii

ACKNOWLEDGEMENTS

To Dr. Freed my major professor without whom I could not
have completed this study without losing my family as well
as my sanity. I express heartfelt thanks.

I thank Dr. Everson for the use of his growth chambers
and his valuable advice in topics related not just to this
study but to life as a plant breeder.

I sincerely thank Dr. Kindel for being a patient and
uncritical sounding board as well as for the unconditional
use of his lab.

For the many pleasant (albeit rigorous) walks through
the woods which brought true meaning to the term
"scientist". I thank Dr. Olien. May I be able to take him
on a walk some day.

Thanks to Dr. Cress and Dr. Tom Isleib for their
invaluable help in the statistics and genetics of this
study.

Thanks to Dr. Whallon for invaluable advise in job
hunting and for cytogenetics.

Thanks to Maria, my wife and best friend, who after

staying home all day with two children was still able to
encourage and uplift me when coming home at night.

iii

TABLE OF CONTENTS

LISTOF TABLES...OOOOOOOOOOOOOOOOOOOOOO.
LIST OF FIGURESOOOOOIOOOOOOOOOOOOOOOOOOO
LIST OF ABBREVIATIONS AND DEFINITIONS...
INTRODUCTIONOOOOOOOOOOOOOOOOOOOOO0......
REVIEW OF LITERATURE - CARBOHYDRATES....
Intracellular protection............
-freezing point depression......
-prevention of plasmolysis......
Membrane protection.................
-dilution of toxic compounds....

-protein stabilization..........

Exracelular/periplasmic protection..
-adhesion

REVIEW OF LITERATURE — GENETICS.........
PURPOSE 0 O O O O O O O O O O O O O O O O O O O O O 0 O O O O O O O O O O
PROCEDURE 0 O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 0

Plant materialOOOOOOO0.0.0.0.0000...

Hardening....................................
Prefreeze incubation.........................
Extraction...................................
Chromatography...............................
Freeze test..................................

RESULTS AND DISCUSSION - PART I (CARBOHYDRATES)..

Analysis of variance...........................
-light...........................................23
-light X cultivar................................23
-cultivar........................................23
-treatment.......................................23
-light X treatment...............................29

iv

Page
...vi
..vii

.viii

0.015
.0017

.17

...17
...17
...18
...20
...20

...22

00.22

TABLE OF CONTENTS cont’d

-cultivar X treatment........
Ratj-OSOIOOOOOOOOOOOOOO0.0.00.0...
Carbohydrates and freeze tests...

RESULTS " PART II (Genetica)ooooooooo
CONCLUSIONS.0.0....OOOOOOOOOOOOOOO...

APPENDIXOOO0.0.0000000000000000000000

REFERENCES.OO....0...0.0.0.0....OOOOOOOOOOOOOO

O O O O .30
O O O O .30
O O O O .36
O O O O .39
O O O O .44
O O O I .47
O O O O .49

00.0052

LIST OF TABLES

Table Page
Table 1. Degrees of freedom (DF) and mean squares

for carbohydrates in a combined AOV for six light

levels. O00.000.00.00.0.0.0.....OOOOOOOOOOIOOOOO0.000000022

Table 2. Mean (3 replications) carbohydrate
amounts and prefreeze changes for three barley
cultivars at six light levels.........................26-27

Table 3. Degrees of freedom (DF) and mean squares
for carbohydrates at six different light levels..........28

Table 4. Survival rating in a freeze-test (see

procedure) and carbohydrate amounts for three

barley cultivars and their F1 progeny hardened at

7 umol m-zs'l............................................41

Table 5. Observed and expected Fa ratios, andehi
square value for segregating plants of the

Dictoo/Durani cross after a freeze test..................43
Table 6. Mean survival rating for three barley

cultivars, their F1 progeny and midparent values,

hardened under high light................................46

vi

LIST OF FIGURES

Figure page

Figure 1. Total carbohydrate amounts after a 3
week hardening period in three barley cultivars
over a range of light 1evels..........................24-25

Figure 2. The log of simple sugar to fructan

ratio at different total carbohydrate amounts in

three barley cultivars before and after a

prefreeze treatment...................................32-33

Figure 3. Simple sugar and fructan amounts at

different total carbohydrate levels for three

barley cultivars before and after prefreeze

calculated from the equations for the lines in
figure-2..............................................34-35

Figure 4. A comparison of the three barley
cultivars for simple sugar levels from
figure 300..0.....OOOOOOOOOOOOOOO0.0.0.0....0.0.0.000037-38

Figure 5. Kill temperature (LD-50) of three

barley cultivars over a range of carbohydrate

levels after prefreeze showing convergence at ca.

130 mg carbohydrate and diverging slopes as total
carbohydrate increases...................................4O

vii

ABBREVIATIONS AND DEFINITIONS USED

AOV: Analysis of variance

 

DF: Degrees of freedom
LSD: Least significant difference
Mean squares: equivalent to variance (33)(42)
33 = Zini - (8iY)’/n
n-l

ns: Differences are not statistically significant at
specified probability level.

Simple sugar: sucrose + glucose + fructose

ss/fn: Simple sugar to fructan ratio.

viii

Introduction

Improving freezing resistance of winter cereals is an
important aspect of crop improvement, but environmental and
genetic components associated with winterhardiness all
interact to produce a complicated system. External factors
affecting plants during winter are soil heaving, smothering,
snow and ice cover, physiological drought, freeze-thaw
cycles, diseases and insects (7,17,44). Infinite levels and
unpredictable combinations of these effects make selection
of hardy lines in the field very difficult. For example, it.
is common for winter survival ranking of barley and wheat
lines to be different from one year to the next (7,23,38),
underscoring the difficulty breeders experience when
attempting to improve winterhardiness by selecting in the
field. Identifying separate freezing stresses and
resistance mechanisms, and then combining favorable
characteristics into adapted cultivars may be a more
manageable approach.

This reductional approach to freezing resistance has
led researchers to correlate almost every measurable
biochemical characteristic with the hardening and freezing

process in plants, with very little agreement between

2

authors (17.44). In many instances correlation "between
hardiness and a cellular component in either a series of
different cultivars or different species or different times
of the year has served as justification for the inclusion or
exclusion of a particular component in the cold acclimation
process......only if the particular compound were the
product of the rate-limiting step in the entire cold
acclimation process would such a correlation exist" (44).
The purpose of this study was to examine relationships

of carbohydrates to freezing survival and to investigate the

genetics of survival in barley.

REVIEW OF LITERATURE-CARBOHYDRATES

Numerous studies have shown the importance of
carbohydrates in cold acclimation and freezing resistance
but exact mechanisms are not clear. In fact, the seemingly
obscure relationship between carbohydrates and freezing
tolerance has prompted some authors to question whether one
even exists (45). Sugar cryoprotection seems to fall into
three broad categories: 1.) intracellular (protoplasmic)
protection 2.) membranes (organelle and protoplast)
protection 3.) extracellular and or periplasmic protection

(from ice adhesions).

1.)Intr§cellular protection
Intracellular protection mechanisms involving simple

sugars are: freezing point depression of cell sap and

3
prevention of plasmolysis as water is removed during
freezing.

Freezing point depression Colligative properties of
ideal solutions, depend only on the number of solute
particles in solution. One such property is freezing point
depression. The freezing point depression constant for
dilute, nonelectrolytic solutions of water is 1.86 which
means that the freezing point of a one molar solution of
water will be lowered by 1.86'C compared to pure water. It
seems reasonable than that intracellular sugar accumulation
during a hardening phase could, to a certain extent, lower
the freezing point and prevent injury by allowing the cell
to supercool. Johanssen et a1. (quoted by 53) followed the
freezing point of wheat at varying solute concentrations and
found that freezing points were lowered on a "purely
colligative basis" (53). However, Levitt (17) reports the
highest recorded vegetative plant cell-sap concentration
would lower the freezing point by only 4'0. Therefore, this
mechanism, by itself, is not viewed to be very important in
freeze protection. While other authors cite similar
relationships between solute concentration and freezing
resistance (8,17,18,19,53) few attribute protection
specifically to freezing point depression. Most simply cite
"colligative cryoprotection" as the mechanism which seems to
include freezing point depression as well as osmotic

protection.

4

Prevention of plasmolysis: Williams (53) and Levitt
(l7, and cited by 53) describe a form of stress in plants
related to minimum cell volume. Williams defines the term
"minimum volume" as:

"some volume below which an individual cell cannot be

plasmolyzed without injury which may develop whether

during the plasmolysis or during subsequent

deplasmolysis and this limit is essentially independent

of the concentration of any solute, intracellular or

extracellular."(53)
He states that reduced cell volume beyond a critical value
"is the principle if not the sole cause of freezing injury"
(53). By determining kill temperatures of shrinking cells
in solutions varying in solute concentration he found a
minimum volume to which cells were reduced before being
killed. A study of the relationship of cell volume to
freeze damage in wheat not only confirmed the minimum volume
hypothesis but showed that wheat changed its lethal cell
volume size by altering membrane properties which in turn
lowered its kill temperature (53).
Role of carbohydrates Plant cells can use sugars,
sequestered or synthesized during hardening, to resist cell
volume reduction by changing osmotic pressure. Osmotic
pressure (x) is "the pressure that must be applied to a
solution to increase the chemical potential of the solvent
to the value of its pure liquid at 1 atmosphere"(4). An
equation describing its relationship to solutes (i.e.
sugars) is:

x = -aRT

where a is the activity (equal to molal concentration in

O

dilute solutions), R is the gas constant (liters atm mole'l
°K'1) and T is temperature in degrees Kelvin.

To maintain equilibrium in a system where water is the
solvent, net movement of pure water will be from regions of
high to those of low water potential (4). Water potential,
9, is defined by:

i = a + i. + in
where x is osmotic pressure (defined above, a negative
quantity), i. is pressure potential and i. is matric
potential. If a semipermeable membrane separates two
compartments differing only in solute concentration
(temperatures are constant) then only solvent (i.e. water)
will move from the less to the more concentrated solution.
This will cause pressure to increase in the high solute
compartment until water potentials are equal in both
compartments. At this point net flow of water will cease.
In plant cells, with high solute levels in the cytoplasm,
differentially permeable membranes, and relatively rigid
cell walls, net water movement would be to the cell
interior. This would produce a pressure which could resist

plasmolysis.

2.)Membrane protection

The difficulty of studying a single component of
freezing injury in vivo has prompted many researchers to
isolate cell constituents, such as membranes, and study

their behavior in vitro (12,20,40,45,46). Photosynthetic

6

phosphorylation associated with chloroplast membranes is a
process particularly susceptible to freeze injury and is
conveniently measured in vitro. Santairius (40) suggests
that one form of freezing injury is an uncoupling of
phosphorylation from the electron transport system in
thylakoid membranes, preventing photosynthesis. He and
other researchers (9,12,45,46) report a protective effect of
sugars (sucrose, glucose and raffinose) on phosphorylation.
While exact mechanisms are not known they propose that
protection is either by reduction of the total number of
potentially toxic compounds which come in contact with
membranes (dilution) or by direct protein stabilization or
both.

Qilption of togic compounds One commonly recognized,
membrane-toxic compound (at elevated concentrations) is
NaCl. NaCl caused permanent damage to thylakoids even at
temperatures above freezing (12). But in the presence of
0.4 M sucrose, cyclic photophosphorylation was protected
100% when thylakoids were frozen at -25'C in 0.3 M NaCl
(12), while in the absence of sucrose, phosphorylation was
completely inhibited at NaCl concentrations below 0.1 M.
Other authors (9,45,46) report similar protective effects
with glucose, raffinose and glycerol. Other potentially
toxic compounds (at elevated concentrations) were Mg“,
Ca**, halogenides, nitrates, sulfates, and certain amino
acids (12). Precise mechanisms of injury to phosphorylation

are not known but Kamienietsky (quoted by 20) was reported

7

to have obtained subunits of Coupling Factor-1 (CFl) when
thylakiods were treated with NaBr. Steponkus reportedly
found CFl released in thylakoid membranes after freezing
with no salt treatment (43). He also reported an
inactivation of ATPase activity when thylakoids were exposed

to elevated NaCl (44) and higher proton uptake in sucrose

protected membranes (44).

There is some debate as to whether the protective
effect is simple colligative dilution by sugars, preventing,
or at least limiting, contact of toxic compounds with
thylakoid membranes or whether a more elaborate protection
mechanism such as protein stabilization is important
(1,12,20,40,44,45,46). Probably a combination of effects
occurs depending on freezing processes and concentration of
cytoplasmic contents.

Protein stabilization Membranes reportedly are made up
of from one-half to two-thirds protein molecules which are
imbedded in a lipid bilayer (39). As cytoplasmic water is
removed during freezing (see section entitled "Role of
carbohydrates"), water of hydration is eventually removed
from protein molecules causing permanent alteration of
tertiary structure (denaturation) (1,12,20,40,44,45,46).
Some authors feel that removal of water alone is enough to
cause denaturation (1). Levitt (17) reports that as water
is removed sulfhydryl groups lose their hydrogens and will
randomly form disulfide bonds. He describes the reaction as

shown:

8

RS-H + RS-H + 40. -¢ RSSR + H20 (17).
As water is removed one can see that this reaction is
favored. The formation of such disulfide bonds may be
irreversible and if so will cause severe protein disruption
as freezing (or thawing) continues (1?). The reaction was
apparently inhibited when sucrose was added to the system
(17). He has also shown that cells which were hardened have
a lower RSSR content in extracted protein when frozen but
admits the absolute levels were initially exaggerated due to
the extraction procedure (17). Little mention is made by
other authors of this theory of injury.

Probably the clearest description of the accepted
theory for how sugars stabilize proteins is given by Alden
et al. (1).

"Hydrogen bonds between proteins and water...are broken

upon removal of water in the formation of ice. Sugars

provide hydrogen bonding between the functional water
of the lipoproteins or directly to the sensitive
lipoproteins of the membrane systems. Sugars, unlike
water, are not frozen out and the stabilizing bonds are
not ruptured when the plant is subjected to subfreezing
temperatures. When water becomes available again in
excess, it replaces the bonds provided by sugar because

hydrogen bonds are easily reversed." (1)

In support of this theory, the ability of proteins to
bind sugar has been demonstrated (12,46) as well as
increased sugar levels in chloroplast proteins during
hardening in the tea plant (46) and in ivy (quoted by 12).
Data are unavailable for sugar binding in winter cereal

chloroplast protein but the molecules are similar so a

similar sugar binding would probably occur.

9

3ngxtracellular(periplasmicgprotection:

Olien describes a form of stress called adhesion
(25,26,27,29,30,31,32) which is a result of equilibrium
freezing [a slow rate of freezing such that very small
displacements from equilibrium occur (28)] processes in
plants and occurs at temperatures between -10'C and -30'C
(28). The advancing ice lattice upon reaching the vicinity
of the cell wall competes with it for the intervening liquid
which causes adhesion between ice and the wall or wall and
plasmalemma. As the protoplast shrinks during freezing,
adhesions to it can cause considerable damage (29).

The compliance of freezing processes to clearly defined
physical laws has led Olien to describe freezing stress
using thermodynamic and kinetic principles (28). An
explanation of adhesive stress from a thermodynamic
viewpoint is shown below.

Figure A illustrates the translational kinetic energy
of water molecules at an ice-liquid interface.

100—
90-1 ,
70-1
GO—l
50-1
40—J
30—1 ‘

B \
20-1 i\\

\

16-1 :— H fl:
0 I

-—‘..—-—..‘~
/
/
I

Frequency x 10—5

 

0
f1

 

 

 

 

0 E 10'00 E2'0‘00 3000 40b0 5:}
Figure A. . . 00(adapted from 28)
Energy m calories mole"1

10

The region "A" represents ice molecules with enough energy
to escape the lattice (melt) and "B" represents liquid
molecules with energy so low that they will add to the
lattice (freeze). As equilibrium freezing progresses the
value (energy) of the two integrals represented by regions
and B will be equal. The vertical line at "Eh" is the
minimum energy ice molecules must-have in order to escape
the lattice (activation energy of melting) and "E1" is the
maximum energy a liquid molecule can have and still add to
the lattice (activation energy of freezing). The distance

between El and Eh is the energy which must be acquired by a

molecule to escape or the amount of energy it gives up as it

becomes part of the lattice (latent heat - H).

 

 

 

100—
904 .«l
w- 70-4
X 60-( \
>4
2 50-1l \.
B 4o—J
; _. \.
E :50 B \.\\
LL 20-1 \x
1"- “. — - ‘
u -1 4—H—9 A x“ 0.2: C
n y --
" a l Eh) l T l
0 1000 2000 3000 4000 5000

Figure B. . .
Energy an colones mole"

(adapted from 28)

When the temperature is decreased the curve shifts

(28), as shown by the solid line in figure B, reducing the

11

number of molecules with enough energy to leave the lattice
(A) and increasing the number that can freeze (B) and
freezing continues. Because this is an equilibrium process
(A and B are of equal energy) some modification must occur
in the system to allow A and B to become equal. One

possibility is to decrease B by reducing the number of

liquid molecules at the interface through dilution of
interfacial water molecules with solutes. A second way for
A and B to reequilibrate is to shift the activation energy
for melting downward, which increases A. If that occurs one
should see a corresponding shift in latent heat - E’ as in
figure C. Olien has observed this shift calorimetrically in

model cellulose systems (30).

100—1
90%

/\

80 1 " \
vo-J \

l
.. s \
50 -1 '\
40 —4 \
30-4 V
20-+

Frequency x 10:5

N.
._9' ‘x

10-1 "Al—H“ \Ax‘- — -25~c

El l Err Ehl l l

0 1000 2000 3000 4000 5000

 

 

 

 

 

 

Figure C. . .
Energy m calories mole"
(adapted from 28)

He attributes the latent-heat shift to matric
interaction of ice with hydrophylic polymers as they compete

for intervening liquid. This can result in highly

12

destructive adhesions if competing polymers are part of

membranes.

Adhesive stress may be relieved by releasing

solutes outside the protoplast producing a fluid barrier to

adhesions.

This causes a change in the number of liquid

molecules in the interface without initially affecting ice.

Once again A and B are out of equilibrium in figure D.

Frequency x 10'5

Figure D.

Molecules

melting occurs.

 

  
 

 

 

Am ‘.

/ v

I

I \\

(B \x
I 4-H'—9 -; -25°c

— -25°C after solute release

I El T Eh' I
0 1 000 2000 3000 4000 5000

Energy in calories mole"
(adapted from 28)

with energy to escape are larger in number and

For equilibrium to be reestablished the

activation energy must again change. A second shift in

latent heat [H”(which has been observed experimentally], as

shown in figure B, would indicate that this does occur (28)

13

100 —
90
80

70_

 

Frequency x ll)’5
‘3
L

 

. \
.._UB M—H'—§.\
'G—HH—E'b A

 

‘4

l Eh’Eh“l l I l
1000 2000 3000 4000 5000

Energy in calories mole"
Figure E.

(adapted from 28)

The ice lattice is now some distance from the
protoplast with an intervening liquid of higher osmotic
pressure than before. This establishes an osmotic gradient
between the protoplast and ice crystals. As freezing
progresses ice crystals will grow at the expense of
protoplasmic liquid causing desiccation.

Once the hypothesis was demonstrated in a model system,
Olien showed that this type of protection does occur in
winter cereals (26,27). A dilute solute concentration in
the extracellular spaces in winter cereals helps prevent the
growth of disease-causing microorganisms (29). This allows
ice crystals to grow up to hydrophylic cell walls during the
freezing process. When frozen, the cell releases sugars

outside the protoplast which can maintain a fluid barrier

l4

necessary for prevention of adhesions (26,27). In fact, the
hardier cereals release more sugars providing greater

protection from adhesion (26,27).

GENETICS

Most genetic studies involving winter hardiness in
cereals have shown freezing resistance to be quantitatively
inherited under the control of many genes
(2,5,10,13,15,38,51,54). While some authors found freezing
resistance to be the result of recessive genes with additive
effects (13,15) others reported both additive and non-
additive effects controlled by both recessive and dominant
genes (5,38). Amirshahi et al. (2) found additive effects
with no gene interaction.

In cases where procedures were compared, different
hardening and freeze-testing methods revealed different
genetic systems for freezing tolerance. Gullord (10,11) and
Le (16) reported a difference in ranking of winter wheat
cultivars for survival depending on whether a high or a low
intensity freezing test was used. In a severe test, Rohde
(38) and Worzella (54) reported hardiness was controlled by
recessive genes while in a less severe test hardiness was
dominant. Roberts (37) found hardiness genes in wheat to be
located on chromosomes 2A, 5A and 58 when plants were
hardened in the light and on 6A, 3B, 5B, and 5D when

hardened in the dark.

15

PURPOSE

Two thoughts are important when considering
carbohydrate related freeze protection mechanisms: first,
sugar distribution in cells, not total sugar [sugar cane
contains very high sucrose levels and is not hardy (17)] is
fundamental and second, because sugar composition is
continually changing (presumably reflecting distributional
changes), even below freezing (26,47), it is distribution
immediately pgfore thg onpgt of ptrgpp that is critical to
an understanding of carbohydrate related protection
mechanisms.

Distribution: According to Wagner et al. (49,50,52)
80% of the sucrose, and 100% of the fructan, glucose and
fructose is contained in the vacuoles of barley leaves.
Whether this reflects sugar distribution in crown tissue
(the critical region for protection in winter cereals) just
before freezing is not known. But Olien has shown that only
a portion of total extractable sugars can be perfused from
crown tissue without damaging cells (26). If the majority
of sugars are sequestered in vacuoles they would presumably
not be available for protection. This makes the ability to
measure cytoplasmic and/or periplasmic sugar critical if one
is to study how sugars are related to winter hardiness.

Sugar changes: Trunova (47) noticed an increase in the
hardiness of wheat if plants were exposed to -3'C for three

days in the dark. He called this period the "second stage

16

of hardening"(47). In addition, he observed a decrease in
levels of the storage carbohydrate fructan and an increase
in fructose and sucrose during this period. He attributed
the increase in simple sugars to respiration driven
hydrolysis of fructan since photosynthesis was not occurring
(47). Olien showed not only the same conversion in rye and
barley but an increase in the amount of perfusable sugar
after the -3'C treatment (27). An increased perfusable
sugar level not only provided critical evidence for an
adhesion protection mechanism but also underscored the
importance of sugar measurements just before freeze injury.
However, difficulties in using the perfusion technique
preclude using it to screen cultivars, so the purpose of
this study was to l) to find an alternate way to measure
carbohydrates which would reflect their spatial distribution
in crown cells and 2) to determine how that distribution is

controlled genetically.

PROCEDURE

Plant material Three winter barley cultivars were used.
Dictoo: A winter hardy introduction. Parents:
unknown. Breeder selected.
Hudson: Released in 1951. Parents: Michigan winter
and Wong. Breeder selected.

Durani: Parents: unknown. Cultivar Introduction # 6316
Eight plants per 12.5 cm pot (the 3 cultivars plus
Rosen rye which was not extracted; i.e. the four lines were
planted in duplicate in each pot) were grown in washed sand

in a growth chamber at 12°C with 18 h/day of light at 175
umol m-ts'1 (80% white fluorescent, 20% incandescent).
Hardening: After five weeks, plants were transferred
to a hardening chamber at l to 2'0 and provided with'
continuous light at 3, 7, 25, 45, 80, or 113 umol m'3s'1 for
3 weeks. Specific light levels were achieved during
hardening by covering the appropriate plants with a wire
frame covered with cotton cheese cloth. A heater coil was
used to warm pots at lower light levels to bring the
temperature up to the level of plants under high light.
Plants were watered daily and fertilized 3 times weekly with
a Hoagland’s solution for the entire growth period.
Prefreeze incubation: After hardening, plants were at

the three leaf stage with 3 to 4 tillers each. They were

removed from pots, washed free of sand in ice water, and the

17

18

shoots and roots trimmed to four and two cm. respectively.
Plantlets (the trimmed plants) were placed in slots cut in
slightly damp. circular, cellulose sponges with a pipe
flange assembly through the center of each sponge to promote
thermostability. One plantlet of each cultivar was placed
in a plastic bag for immediate extraction (control) and the
other plants from the same pot were placed in sponges for
extraction following the prefreeze incubation. This was
repeated with 9 more pots per replication [i.e. each sample
(replication) contained 10 plants]. A split plot design was
used with three replications, two treatments [hardened (H)
and hardened then prefrozen (HP)], and three cultivars.

Sponges containing plants to be prefrozen were
inoculated with ice, covered with plastic to help prevent
desiccation, and placed in a freezer at -3'C for 60 h in the
dark. Thermocouples placed near crowns in the sponges were
used to monitor tissue temperature. Because it took 3 to 7
h for the water in the sponges to freeze, actual time at -
3°C was between 53 and 57 h.

After the prefreeze incubation, plantlets were either
extracted to determine carbohydrates or freeze-tested to
determine survival.

Extraction: Control plants were extracted immediately
at l'C and frozen plants at -2'C (in a cold room) after 60 h
at -3’C using the same procedure. Plantlet roots were

removed and the fibrous tissue at the stem base (zone of

19

transition between roots and shoots) was grated with a fine-
toothed, hand grater. In addition, 5 cm of the stem,
nearest the stem base, was trimmed with a razor blade,
combined with the grated tissue and all were placed in a
mortar containing 5 ml of aq. 80% (vol./vol.) ethanol. The
tissue was mascerated for approximately 1 min with a motor
driven mortar and pestle. The ground tissue was transferred
to a 100 ml beaker and the mortar and pestle rinsed with 10
ml aq. 80% ethanol which was added to the beaker. The
tissue was placed on a 70'C water bath for 10 min to
inactivate invertase and shaken for 15 min on a gyratory
shaker. The supernatant was transferred to an erlenmeyer
flask and the tissue extracted an additional two times with
double distilled water (10 ml each) at 21'C. A fourth water
extraction contained less than 5% (of total extracted) V
additional carbohydrate. The combined supernatants were
swirled by hand and a 20 ml alaquat was transferred to a 50
ml graduated centrifuge tube; 400 mg of mixed bed resin was
added and the mixture was shaken for 10 min (to eliminate
ions damaging to the separation column). Samples were
centrifuged at 680g for 10 min and 10 ml of the supernate
decanted to a preweighed round bottom flask. They were
taken to dryness under vacuum at 35‘C and placed in a vacuum
desiccator over P305 for approximately 12 h. Samples were
weighed, brought to a concentration of 10 mg dry extract per
ml double distilled water and passed through a 0.45 micron

Millipore filter. Pulp was recovered and dried at 85'C to

20

constant weight (12 h). Total dry weight consisted of dry
pulp plus dry extract.

Chromatography Carbohydrates were separated by high
pressure liquid chromatography (HPLC) with a Bio-Rad Aminex
HPX-87P column at 85°C. The mobile phase consisted of
degassed HPLC grade water (Baker Chemical Co.) and had a
flow rate of 0.4 ml/min. A Micromeritics 771 refractometer
was used to detect carbohydrates which were quantified by
co-chromatography with external standards. Fructan from
Hudson barley, collected from the column and freeze-dried,
was used as the fructan standard for all cultivars. Because
of a consistent linear relationship to absolute amount of
standards, peak-height, measured by hand, was used to
determine sugar amounts. Retention times in minutes were:
fructan 9.2, sucrose 14.8, glucose 18.4, and fructose 25.8..
Sugars are reported in mg per gram dry weight of tissue.
Simple sugars are sucrose plus glucose plus fructose and
total carbohydrate is fructan plus simple sugars.

Freeze test In addition to the three cultivars used in
the sugar extractions, F1, F3, F3 families in the F3
generation and backcrosses of the cultivars were freeze-
tested. Plantlets were treated as above until after the
prefreeze treatment at -3°C. The freezer temperature was
lowered 1°C per hour to the particular test temperature
which resulted in 50% survival (LD-50) of the cultivar being
tested. This temperature varied between -22°C and -7°C

depending on light levels during hardening (under low light

21

plants were more tender and were frozen at a higher
temperature). A minimum of three freeze tests per cultivar
at each light level were used to determine LD-503. Graphs
of temperature verses survival were drawn and the
temperature corresponding to 50% survival was taken as the
LD-50. Plantlets were held at the test temperature for 19

hours and then raised 4 to 5°C per hr to 35°C and placed in
a cold room.

The following morning plantlets were removed from
sponges, trimmed of roots and placed in flats of sand in a
growth chamber at 20°C where they were fertilized and
watered. After 2 weeks, plants were visually rated on the
basis of shoot and root regrowth on a scale of 0 (dead) to 5

(undamaged).

RESULTS and DISCUSSION - CARBOHYDRATES

Data from the carbohydrate analyses are in three
sections: 1) analysis of variances (AOV) for crown sugars at
six light levels, 2) the relationship between total
carbohydrate and simple sugar to fructan ratio, and 3) the
relationship of total carbohydrate to kill temperature.

Four variables were evaluated in the carbohydrate
study: 1) fructan, 2) simple sugar 3) total carbohydrate and

4) simple sugar to fructan (ss/fn) ratio.

1) Analysis of variance: In the overall analysis (Table 1),
highly significant differences were found for all factors
(light, cultivar and treatment) and variables (fructan,

Table 1. Degrees of freedom (DF) and mean squares for carbohydrates in
a combined AOV for six light levels.

 
     

 

Source DF fructangfn) ssgfn
Light (L) 5 450857** 509467¥t ‘500.8**
Replication 12 4068 592 2864 ‘ 11.6
Cultivar (C) 2 17350:! 38944*t 7863#* 39.43:
L X C 10 2564** 1547*: 4524** 9.9!:
Error (a) 24 81 59 138 1.4
Treatment (T) 1 36008** 24001** 1213 204.7:*
L X T 5 2883*: 2978*! 1067*: 61.3**
C X T 2 209 455 ‘ 48 10.1:
L X C X T 10 110 125 i 273 3.1
Error (b) 36 137 158 i 312 2.2

 

 

 

 

**,* Significant at 0.01 and 0.05 levels of probability respectively.

22

23
simple sugar, total carbohydrate, and ss/fn) and for some
interactions (light x cultivar, light x treatment, cultivar
x treatment).

Ligpp: Carbohydrate accumulation occurs during
hardening in winter cereals when demand is reduced due to
slower growth from lower temperatures (8,17,18,45,47). It
is logical then, if other factors are held constant, that
higher light levels will increase carbohydrate supply and
cause higher overall carbohydrate levels. Studies on
shading effects (76,14,24,36,48) as well as Figure 1 confirm
this.

‘ light 3 cpltivgp: Figure 1 also illustrates the
significant interaction between carbohydrate and cultivars
(Table 1). At high light Durani has more total carbohydrate
than the other two cultivars (not a statistically
significant difference, see Table 2) but under low light (7
and 3 umol m-3sec-1) Durani has significantly less.

Cultivar: Because an LSD calculated from the overall
AOV will bias the mean separation at extreme values, LSD’s
in Table 2 are calculated from individual experiment AOV’s
(Table 3). Cultivar differences are discussed in the
"ratios" section.

Treatment: At light levels of 113, 80, 45 and 25 umol

m-zsec-1 for every factor (except total carbohydrate)

24

Figure 1. Total carbohydrate amounts after a 3 week hardening
period in three barley cultivars over a range of light levels.
Each point represents the mean of 3 replications. Bars above the
lines represent LSD values at P=.05.

25

 

 

 

600*
J .
+3 5004
3
.>\
" ‘l
U
0‘
x 4004
C?
(’-
C
.l
E
Q 300-1
U
L.
"O .l
>\
.5:
O
.0 200-1
a
D
U .l
E
.4.)
)2 100-4
7 e—e Dictoo
B—El Hudson
0 i e—s Durani
l 'W . V7 l . l . l . l
0 2 40 60 80 100 120

/

0
Light level in ,I.L.mOi/'rn2 sec

26

Table 2. Mean (3 replications) carbohydrate amounts and prefreeze
changes for three barley cultivars at six light levels.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

wei t
simp e
Light¥cultivar tmt# fructangfn) s are as carbohydrate ssgfn ratio
Dictoo H 400b0§ 1130 513a 0.280
HP 340e 165a 504a 0.49s
Change -60 +52 +0.20
Hudson H 415ab 1070 522a 0.260d
113 HP 350d 164a 514a 0.47s
Change -65 +57 +0.21
Durani H 436a 95d 531a 0.22d
HP 3800d 139b 519a 0.37b
Chgpgg -56 +44 +0.15
LSD(.05) gpppge 26.8 7.5 ns 0.04
Dictoo H 3000d 126b 426b 0.430
HP 242a 181a 423b 0.75s
Change -58 +55 +0.33
Hudson H 354b 1000 454a 0.29d
HP 290d 168a 457a 0.58b
80 Change -64 +68 +0.29
Durani H 385a 880 473a 0.23d
HP 3180 140b 458a 0.440
Change -67 +52 +0.21
LSD(.05) Chgpgg 27.4 6.3 ns 0.04
Dictoo H 2300 182b 413ab 0.86b
HP 158d 217a 3750 1.5a
Change -72 +35 +0.64
Hudson H 277b 1530 430a 0.59b0
HP 2190 178b 397b 0.85b
45 Change -58 +25 +0.26
Durani H 304a 106d 410b 0.360
HP 272a 1370 409b 0.52bc
e -32 +31 +0.16
LSD(.05) Chgpgg 33.2 24.2 19.1 0.15

 

 

¥ Light level in umol nr28'1 during 3 weeks of hardening.

# Treatment, H=hardened in continuous light, HP=hardened then prefrozen
at -3°C for 60 hours in dark.

0 sucrose+glucose+fructose

5 Means within a column and light level followed by the same letter are
not significantly different from each other at P=0.05.

27

Table 2 cont’d
Mggg ggx weight

 

 

 

 

 

 

 

 

 

 

 

simple total
Light*cultivar tmt# fructangfn) spgan@(ss) carboh te ssgfn ratio
Dictoo H 92de§ 216b0 308ab 2.4b
HP 72e 261a 333a 3.7a
Change —20 +45 +1.3
Hudson H 118bc 1840d 302ab 1.6bc
25 HP 980d 226bc 324a 2.3b
Change -20 +42 +0.7
Durani H 145a 130s 275b 0.910
HP 128ab 164de 293ab 1.30
Chgpgg —17 +34 +0.39
LSD(.05) Changgi 22.9 44.7 ns 0.69
Dictoo H 25a 129b 154a 5.7b
HP 10b 153ab 163a 15.5a
Change -15 +24 +9.8
Hudson H 32a 1388b 170a 4.4b
HP 12b 157a 169a 14.9a
7 Change -20 +19 +10.5
Durani H 23a 790 102b 3.8b
HP 10b 620 710 8.6b
Chgpgg -13 -17 +4.8
LSD(.05) thpge 10.3 pg ns 6.0
Dictoo H 11b 134a 145a 12.3bc
HP 6b 122ab 128b 19.8a
Change -5 -12 +7.5
Hudson H 20a ll6b l35ab 6.7d
3 HP 7b 114b 122b 15.4b
Change —13 -2 +8.7
Durani H 7b 560 620 9.30d
HP 4b 43d 460 12.8b0
Cpgpgg -3 -13 +3.5
LSD(.05) ghgpge 8.9 12.8 21.8 4.0

* Light level in umol Eras-l during 3 weeks of hardening.

 

# Treatment, H=hardened in continuous light, HP=hardened then prefrozen

at -3°C for 60 hours in dark.
0 sucrose+glucose+fructose
§ Means within a cohmmiamd light level followed by the same letter are

not significantly different from each other at P=.05.

Table 3.

six different light levels.

28

Degrees of freedomr(DF) and.mean squares for carbohydrates at

 

 

 

 

 

 

 

 

light Source DF fructan carbohydrate ss/fn rptio
Replication 2 762 82.0 629 0.002::
Cultivar (C) 2 2284: 833:: 418 0.014::

113 Error (a) 4 271 18.0 415 0.0004
Treatment (T) 1 16,388:: 11,667:: 400 0.158::
T x C 2 39.0 76: 7.0 0.002::
Error (91, 6 180 14.0 275 0.0004
Replication 2 3493:: 361 2637:: 0.019:
Cultivar (C) 2 9950:: 2365:: 2734:: 0.100::

80 Error (a) 4 113 91.0 153 0.003
Treatment (T) 1 17,915:: 15,025:: 127 0.347::
T x C 2 24 101:: 116 0.005::
Error (b) 6 188.1 10.0 171 0.0004
Replication 2 19,324:: 819:: 12,528:: 0.368:
Cultivar (C) 2 13,321:: 9275:: 644: 0.794::

45 Error (a) 4 14.4 41.6 69.6 0.047
Treatment (T) 1 13,162:: 4154:: 2527:: 0.511::
T x C 2 636 37 620: 0.080::
Epror (b) 6 276 146 91.6 0.006
Replication 2 582: 1813:: 342: 1.12
Cultivar (C) 2 4431:: 12,680:: 2239:: 5.75::

25 Error (a) 4 60.7 122 29.9 0.256
Treatment (T) l 1606:: 7394:: 2108 2.93::
T x C 2 7.60 55.0 26.3 0.350
Error (b) 6 132 502 910 0.120
Replication 2 197: 385: 1075: 48.9:
Cultivar (C) 2 62.0 10,844:: 12,073:: 32.3:

7 Error (a) 4 14.9 58.2 120 3.51
Treatment (T) 1 1138:: 297 272 312::
T x C 2 21.0 747 639 14.5
Error (b) 6 26.5 218 297 9.05
Replication 2 48.0 98.0 281: 19.0
Cultivar (C) 2 109: 10,689:: 12,373:: 50.0:
3 Error (a) 4 8.91 21.3 40.3 4.31

Treatment (T) 1 207: 352: 1098: 195::
T x C 2 35.0 61.0 4.8 11.0
Error (b) 6 19.8 57.9 119 4.01

 

 

 

 

 

::,: Significant at 0.01 and 0.05 levels of

probability respectively.

29
differences were found for treatment effects. This confirms
past findings (21,26,48,) and could be the result of
respiration as suggested by Tumanov (48) or mass action
effects as cell contents are concentrated due to water
removal during freezing (Olien-personal communication) or
both. No significant change in total carbohydrate during
the treatment was found except for experiments at 45 and 3
umol m-zseC'1 (Tables 2 and 3). This suggests a
carbohydrate proportional shift without preferential loss of
any one component. The two experiments with significant
changes are probably a result of procedural error because
there is no continuity between the occurrence of the two
changes and it is rare for the before and after prefreeze
.treatments not to have the same carbohydrate levels.

Light x treatment: As light levels increase so do
changes in fructan and simple sugars (Table 2). For all
three cultivars prefreeze induced changes are at a maximum
at 80 umol m-3sec'1 (Table 2). At very low light levels (7
and 3 umol m-3sec-1) no significant increase in simple sugar
was found; the significant change in Table 3 (compared to
the mean) at 3 umol m-isec'1 is a result of simple sugar
decrease in Durani. It is possible that at this light level
the demand for simple sugar has exceeded the supply by
fructan hydrolysis and sequestered simple sugar is being
consumed. However, the change is barely significant at

P=.05 so procedural error is also possible.

30
_pltivpr x treatment: While Olien (26) found
significant differences in prefreeze-induced fructan
hydrolysis between rye and barley, none were found between
the three cultivars in this study at any light level. At 80
pmol m'3se0'1, however, a highly significant interaction
between simple sugar increase and cultivars was found

(Tables 2 and 3). An LSD infers that Hudson’s simple sugar
increased to a greater extent than Dictoo or Durani. If the
source of increased simple sugars is fructan then there may
be genetic differences between cultivars for fructan
hydrolysis but experimental variability in this test is too
great to detect them.

2) Ratios: All data presented are from a total extraction
of crown tissue; therefore no information is available on
carbohydrate distribution within and around cells, the
critical factor in sugar cryoprotection. One simple way to
express carbohydrate data in the form of a distribution is
to divide simple sugars by fructan [a distribution between
types of carbohydrates (compositional distributionll.

If demand for simple sugar remains relatively constant
[respiratory pools are apparently maintained at the expense
of storage and exogenous carbohydrate ( 3)] then as
carbohydrate supply is increased the storage component
(fructan) should increase and the ratio would decrease. At
low light levels (see Table 2) the ratio for Hudson after
the prefreeze was 15.4 while at high light it was 0.47. In

fact, a linear relationship was found when plotting the log

31
of the ratio against total carbohydrate (Figure 2). For all
three cultivars the correlation coefficient was at least
0.97 and was highly significant.
From the equations for the lines, simple sugar and
fructan amounts can be calculated (thereby eliminating

experimental variability) and plotted over a range of

carbohydrate amounts. This allows visual comparison of
fructan, simple sugar and their shifts during the prefreeze
incubation (Figure 3). The plots show: 1) a simple sugar
maximum between 300 and 350 mg carbohydrate which then
begins to decrease. Possibly when incoming simple sugar
exceeds a certain level it is diverted to the storage pool.
This may be analogous to what Pollock (33) found after
transferring plants to cold temperatures. Sucrose
accumulated to ca. 10 mg/g fresh wt. after 3 days and then
began to decrease while fructan levels increased. Labhart
et al. (14) suggested a threshold level of sucrose and short
chain fructans is necessary before significant fructan
synthesis begins. 2) noticeable freeze-induced changes
begin at 100 to 150 mg carbohydrate and reach a maximum
around 400 mg carbohydrate. It is at this maximum that
statistically significant differences between cultivars were
seen for simple sugar increase (Tables 2 and 3). 3) Durani
accumulates carbohydrate in the form of fructan to a greater
extent than either Durani or Hudson (leaving less simple
sugar available for cryoprotection). This could be

advantageous for a cultivar not specifically bred to cope

Figure 2. The log of simple sugar to fructan ratio at different
total carbohydrate amounts in three barley cultivars before and
after a pref reeze treatment. Each point represents the mean of 3
replications.

 

Log simple sugar/fructan (ss/fn)

33

DlCTOO

   
    
 
 

 

 

  
    
  
 

 

 

 
 
 

   
  

 

1.2
o after prefreeze
0,8 log ss/fn - 1.87 — .0045 X cho
r :- -.97ee
e
0.4
0.0 2
before prefreeze
_._4 log ss/fn - 1.54 - .0041 x cho
r - ~37”
“-8 I I . I I I I I I I r
0 100 200 300 400 500
HUDSON
1.2
after prefreeze
0.3 log ss/fn = 1.74 - .0042 X cho
r = —,98ee
0.4
0.0
:l before prefreeze .
-.4 log ss/fn = 1.29 - .0037 X cho
r - -.98~
-'8 l ‘ I I T 1 I I I f I
0 100 200 300 400 500
DURAN!
1.2
A
after prefreeze
08 log ss/I'n - 1.15 - .0033 X cho
r = -.98+*
0.4
0.0
before prefreeze
--4 log ss/fn - 1.00 - .0034 x cho
3 r = -.98:: A
_ 8 L r . r . I . 1 . . ‘1
0 100 200 300 400 500

1g tctol carbohydrate/g dry weight (cho)

34

Figure 3. Simple sugar and fructan amounts at different total
carbohydrate levels for three barley cultivars before and after
prefreeze calculated from the equations for the lines in figure 2.

 

Mg fructan or simple sugar/g dry weight

35

4001
e—e after prefreeze D ICTOO )3
-l e—e before prefreeze '/”p

 

 

 

0 100 i 200 . 300 i 400 i 500
400

1 H after prefreeze HUDSON f

«l m—o before prefreeze /
300. //

 

 

 

0 100 ' 200 i 300 ' 400 ' 500
4001 A
H after prefreeze DU RANl ' /:
l e—a before prefreeze / /

 

 

 

I.”
13H
‘3

T Y I ' l ‘ l
100 200 300 400

Mg total carbohydrate/g dry

5.
(l)
.fii'
3'
r...

36
with a harsh winter environment. Where mild winters and
favorable conditions for fall carbohydrate storage occur
Durani may have an advantage due to high stored fructan
levels, allowing more rapid regrowth in spring.

When calculated simple sugars (from Figure 3) for the
three cultivars are plotted on the same graph, differences
between the three cultivars are apparent (Figure 4). Dictoo
has a relatively high simple sugar level both before and
after freezing. Hudson has a low simple sugar level before
freezing but because its shift is greater than Dictoo it
tends to approach Dictoo after the prefreeze. Durani has
less simple sugar before and after prefreeze than either
cultivar. If simple sugar increase during the prefreeze is
important in the cryoprotective process (26,48) then
enhancing Dictoo’s conversion may improve its freezing
tolerance.

3) Carbohzdrate and Fregge tggpp: Because these data do not
supply information on actual distribution of carbohydrate
within plant cells and interstitial spaces expectations of a
simple correlation between carbohydrate fractions and kill
temperature would be somewhat naive.

However, if total carbohydrate increases, over a range
of light levels, and demand (i.e. respiration) for sugar
remained the same (3) then the "surplus" sugar would be
available for storage or cryoprotection or both. If true,
then, while actual distributions are unknown, increasing

total carbohydrate should increase overall freezing

37

Figure 4. A comparison of the three barley cultivars for
calculated simple sugar levels from figure 3. Dictoo and Durani
are values after a prefreeze; Hudson (dotted line) is before and
after a prefreeze.

 

Mg simple sugar/g dry weight

300.0 1

200.0-

 

  

I
u 100

38

Duroni ofter prefreeze
Hudson after prefreeze
Hudson before prefreeze
Dictoo ofter prefreeze

IEII

 

 

I l I Ti I l

l l l
200 300 4 0 500

0
Mg total carbohydrate/g dry wei ht

IQ

39

tolerance. Figure 5 illustrates the relationship of total
carbohydrate to kill temperature. If, with increasing
carbohydrate levels, a portion of the additional (beyond
that needed for respiration and storage) carbohydrate is
allocated for cryoprotection then the difference in the

relationship of kill temperature to total carbohydrate

(slope) between the three cultivars, indicates a difference
in the allocation of additional carbohydrate for
cryoprotection; this suggests differences in cellular
distribution. The three cultivars also appear to converge
at about 100 to 150 mg carbohydrate. At this carbohydrate
level all three cultivars kill at about -10°C which is the
temperature where adhesive stresses (28) begin to be a
factor in cell death.

When raised under the same low light conditions,
however, Durani has significantly less carbohydrate than the
other two cultivars (see Figure 1). Therefore, after being
hardened under identical light conditions and freeze tested,
Dictoo and Hudson were killed at around the same temperature

but Durani was always killed at a much higher temperature.

RESULTS - GENETICS
At high carbohydrate levels differences between
cultivars for total carbohydrate/g dry wt. were not
statistically significant (Tables 1 and 2). Most genetic

studies of winter hardiness were performed with plants

hardened under high light (5,10,15,38,51,54) and genetic

40

. .35538

B be: 3.8 0.502093 do. new 25 0083 on 3.. dis: .3516 05 82.. 5 23

a 3.. Pen .8 be ..e. .0303 2:. 683880 use: «in 05 was! 3.8.3.. 20.. 35am 89.:
34.3 302a 385: 358 3% B 0833.. 8303 838 .33 non 3838398. 353
an enema we saw: 303 vacuum a mo 51 a Banana #50.— 3 demon—moan wound magma

 

 

382553 no 09.2 a 3.6 893:6 .823 025 no .813. 38093 as .s 8.3.—
29m; >00 o\caocc\€oc:oo 2:3 32
can oov com com cos
n“ _l P —l P _l P I—r IF F p O
owl
a u u ,
3 .7/././ mN H c ..
..emali ESE/xx 9. n B -
ill a .I/ elm. El cEldelull I- 19.:
..Bmfui zomozzim/I // I . El . _ >139. .
C elli.l.llll!lll.l l/ll/HHHWO/U H
:95..- It .23.:5 Jiliil u xii,
L lllll Ilblmb -
U l..w..é lOPl

T

 

SJFIlDJSClLUel

lll>Il

d—

'3“!

41

effects were found to be complex. Under low light, (see
Figure 5) differences in kill temperature between Durani and
the other two cultivars were more related to amount of total
carbohydrate than at high light. For example, while the
kill temperature of the three cultivars converged when
carbohydrate was restricted, under the same low light
conditions, carbohydrate levels between Durani and the other
two cultivars were significantly different (Figure 5 letters
0 and d).

Freeze test and carbohydrate data for the three parents
and their F1 progeny are shown in Table 4. Differences
between reciprocal crosses were not significant so their
results were pooled.

Tabke4. thmvrwfl.rathmzintifreemrdzet humepnxXEMre)amd

carbohydrate amounts for three barley cultivars and their F1 progeny
hardened at a light level of 7 “Incl m'3s'1 .

pgrxutpggygfiegggygrwt

I

 

 

 

 

 

 

 

 

 

 

Cuhflvar sundyabt fnmnan smgye»mqpu‘ «unfinhmgggei _—
Dictoo(Di) 2.03a6 22a 122a 144a
Hudson(HU) 1.17b 17ab 92b 109cd
DuranigDu) 0.00d 80 51e 59g
F1
Di/Hu 2.42a 16b 115a 13Gb
Midparent 1 . 60:: 19 . 5ns 107ns 135ns
Di/Du 0.4lcd 9c 85bc 95de
Midparent 1.02:: 15.0:: 87ns 101ns
Hu/Du 0.28cd 60 64d. 71fg
Midparent 0.58ns 12.5:: 72ns 84:
# mean of (:50 plants (replications); vismlly rated 0=dead to
Sandmmqed.

8 Means within a column followed by the same letter are not
significantly different from.each other at P=.05.

::,: F1 is different from the midparent at 0.01 and 0.05 levels of

probability respectively.

42

Highly significant differences from midparent values
in several crosses (Table 4) indicated the presence of
dominant effects. While Dictoo had a highly significant
dominant effect for hardiness in the Dictoo/Hudson cross, in
a Dictoo/Durani cross the effect was apparently masked by a
dominant effect for non-hpgdingpg in Durani.

Under conditions of this test, when crossed with
Hudson, Durani’s dominant effect for non-hardiness was not
observed. This could be due to either different genetic
effects in that cross or too harsh a test causing Hudson to
have a lower mean survival rating and consequently lower
midparent value (if frozen at a higher temperature Hudson
would have a higher survival rating while Durani could still
have a survival of zero; this would raise the midparent
value and possibly show a dominant effect in the cross).
Durani also showed a highly significant dominant effect for
reduced fructan levels.

Because dominance effects were greatest in the
Dictoo/Durani cross (under conditions described in Table 4)
this cross was considered the most likely to show the effect
of a simply inherited system in a freeze test under those
conditions; consequently, while some data was taken on all
possible crosses (see appendix), only this cross was
analyzed in detail. In addition, because the number of
samples needed to do an adequate genetic analysis of
carbohydrate inheritance was so large only freeze test data

was taken.

43

F3 Seeds from eighty F: families were classified
according to rachilla hair length [a simply inherited trait
(6)] and were not significantly different from a 3:1
(longzshort hair) ratio. They were allowed to self and 12
Fa-plants per F2 family (960 total plants) were freeze
tested along with 96 backcross; (b0); and 40 b0;
individuals. All plants were hardened under low light and
test frozen at -12°C under conditions where survival for
Dictoo and Durani was 100 and 0 percent, respectively.
Plants were allowed to recover for 2.5 weeks at 22°C and
were rated as either dead or alive.

A chi square was performed on observed Fa ratios.
Expected F3 values were calculated on the basis that F:
heterozygotes would be segregating in the F3 generation. So
a 13:3 (dead:alive) ratio in the F3 generation would give a
49:15 (dead:alive) ratio in the Fa. The results indicated
that observed values were significantly different (at
p=0.001) from both a 3:1 (dead:alive) F2 ratio and a 15:1
(dead:alive) F2 ratio but were not significantly different
from a 13:3 F2 ratio (Table 5).

Table 5. Observed and expected.dead:alive Fa ratios, and.Chi square
::::e for segregating plants of the Dictoo/Durani cross after a freeze

I (flmerwxl expxned

 

Dead 746 735
Alive _i £14 225
Chi square 0.64 |

44

A 13:3 Fa ratio implies that two genes control the
effect with one epistatic to the other. A model showing
possible F2 genotypes and their proportions is shown below.
"A" is a gene whose effect results in death when dominant;

it is epistatic to "B", a gene resulting in death when

 

recessive.
F2 ymemnmphz
pquutrmm enounne genetundon expnumdon
9/16 A_B_ B dominant, but overridden by A.... dead
3/16 A_bb brecessive,Adominant............dead
3/16 aaB_, a recessive, B controls expression. alive
1/16 aabb a recessive but b is also.......... dead

 

 

 

If Dictoo is the genotype aaBB and Durani AAbb then the
F1 would be AaBb and backcrosses should result in the
following combinations:

Parental F1 953199 ratio
gmmfies AB l ab (kndnflive

 

Dunno aB

AaEB
Dunnd Ab .Nflfi) b .mnm. allchod

lAb l“B
lAaBblaaBBlaaBb 1:1
AAbb AaB
A backcross with Durani resulted in 39:1 (dead:alive)
which was within the confidence limits for a frequency of
zero alive (42). A backcross with Dictoo had a ratio of

55:41 (dead:alive) and a chi square of 1.76 which is not

significantly different from a 1:1 ratio.

DISCUSSION - GENETICS
Martin (22) found lower respiration rates and higher
"conserved" carbohydrates in hardier rye and wheat
cultivars. While the present study provides no evidence to

confirm that low fructan levels are an indication of higher

45

respiration rates, if fructan accumulates when supply
exceeds demand (34,35) then low fructan suggests either
higher demand or lower supply of carbohydrates or both. If
true then Durani’s dominant factor could be a respiration
promoter. For example, dominant A would cause high
respiration resulting in little or no sugar availability for
cryoprotection. Dominant B (in Dictoo) could be a
respiration inhibitor which causes low respiration allowing
sugar accumulation. If b is recessive respiration would not
be inhibited and would result in low sugars and a higher
kill temperature (less hardy).

If A is a respiration promoter, which reduces sugars
available for cryoprotection, and if adequate sugars are
provided, supply would exceed demand and the effect of A
could be masked; in this case sugar cryoprotection would be
adequate and the plant would die at a lower kill temperature
(more hardy) and possibly from different stresses. In fact
when hardened under high light Durani’s carbohydrate level
is not significantly different from Dictoo or Hudson (table
2) and the dominant effect for non-hardiness is not observed
(table 6); in contrast, a dominant effect for hardiness is
seen in the Dictoo/Durani cross under these conditions
(Tables 6-see appendix for complete set of crosses at high
light). This could explain Roberts (37) finding genes for
hardiness on different chromosomes depending on whether
plants were hardened in the dark or in the light. However,

more information is needed to confirm this.

46

Table 6. Mean survival rating for three barley cultivars, their F1
progeny and midparent values, hardened at 113 nmol m"3s'l .

 

 

 

 

 

 

 

 

 

 

Parents 13
survival# survival
Dictoo 3.62a6 Di/Hu 3.45ab
Midparent 2.56::
Hudson 1.50d
Di/Du 2.98b
Durani 0.10e Du/Di 2.350
Micmarent 1.86::
Hu/Du . 1.08d
Midparent 0.8
# mean of 60 plants (replications); visually rated 0=dead to

5=undamged.
5 Means within a column followed by the same letter are not
significantly different from each other at P=.05.
:: F1 is different from the midparent at 0.01 level of probability.

CONCLUSION

Difficulties in improving winter hardiness of existing
crops have prompted many investigators to try and reduce
protective mechanisms to individual simply inherited traits.
When considering carbohydrate related protective mechanisms
sugar distribution immediately before the onset of stress is
critical to an understanding of this important category of
cryoprotection (26,47). Because current techniques are not
amenable to screening cultivars for sugar partitioning the
aim of this study was to derive a measurement from total,
extractions which would reflect carbohydrate partitioning in
plant crowns and would relate to freezing tolerance.
Specific findings were:

1) A simple sugar to fructan ratio, expressed as a log,
was found to have a linear relationship to total
carbohydrate/g dry wt.i When equations for the lines were
used to calculate simple sugar and fructan amounts (which
eliminated experimental variability) at varying total
carbohydrate levels the following differences were apparent:
a) simple sugar/g dry wt. increased to a maximum at about
300 to 350 mg total carbohydrate/g dry wt. and then began to
drop as total carbohydrate/g dry wt. increased b) the

cultivar Durani allocated more of its total carbohydrate to

47

48

the storage pool, fructan, than the other two cultivars c)
prefreeze-induced simple sugar/g dry wt. change was at a
maximum at the maximum simple sugar level (300 to 350 mg
total carbohydrate/g dry wt.) and the change during the
prefreeze was greatest in the cultivar Hudson.

2) As the level of total carbohydrate was lowered,
kill temperatures increased at different rates in the three
cultivars, (suggesting a difference in the amount of sugar
allocated for cryoprotection) and converged at a
carbohydrate level of approximately 100 mg carbohydrate/g
dry wt. At this point all the cultivars were killed at
around -10°C, the temperature at which adhesive stress was
reported to begin (27). When hardened under the gppg
conditions, however, the less hardy cultivar Durani
accumulated significantly less carbohydrate than the other
two cultivars and consequently was killed at a higher
temperature.

3) When hardened under low light and freeze tested
such that survival of Dictoo and Durani is 100 and 0
percent, respectively, chi square tests of F: and backcross
progeny from the Durani/Dictoo cross indicated that two
genes were operative. Durani had a dominant gene expressed
as freezing susceptibility and was epistatic to a dominant
gene in Dictoo expressed as freezing resistance.
Carbohydrate data suggest that the genes may be related to
supply or demand of sugars or both, but more study is

necessary to confirm this.

 

APPEND IX

 

50

Table A1. Survival rating and frequency for each rating class of three
barley cultivars and their progeny hardened under low and high light

 

 

 

 

 

 

 

 

 

 

 

 

HIGH LIGHT L113 umol brig-'1) 10W LIQHT (7 “£01 grief-1)

fmue_ncz fwncz

survival 012345 survTival 012345

PARENTS
Dictoo 3.55s 022556 2.90abcd 063416
Hudson 1.55fg 833321 2.35cde 722126
Durani 0.15l 1811000 0.00.1 2000000
F1
Di/Hu 3.05abc 033662 3.50ab 230249
Hu/Di 3.30ab 013772 3.30abc 134138
Di/Du 2.15def 3 3 7 3 3 1 1.10mi 10 4 3 1 1 1
Du/Di 1.65fg 6 3 6 2 3 0 0.80ghij 14 2 l 1 1 l
Hu/Du 1.15ghijk 10 5 1 1 2 1 1.15ghi 14 1 0 0 3 2
DuZH_u 0.60131 126 02 0 0 0.20ij 18 l 0 1 0 0
F2
Di/Hu 2.500de 332570 2.65bcd 610535
Hu/Di 2.85abcd 115751 3.30abc 240167
Di/Du 1.35ghij 6 5 6 2 1 0 1.40fg 11 1 3 0 4 1
Du/Di 1.45fghi 9 2 3 4 1 1 1.103111 13 1 2 1 1 2
Hu/Du 0.80hijkl 12 3 3 1 1 0 0.30hij 17 2 0 0 1 0
DuZH_u 0.5021 16 L0 0+2 0 0.10.1 18 20 0 0 0
BACKCROSS
Di/Hu/lDi 3.15abc 22 l 64 5 3.80s 030449
Hu/Dil/Di 3.50a 0 2 3 3 7 5 3.60s 1 1 2 4 5 7
Hu/Di/ll-iu 2.10ef 444 34 l 2.500de 900 1 3 7
Di/Hu/lHu 1.5% 7 4 4 3 l l 2.60ch 3 5 L1 54
Di/Du//Di 2.65bcde 2 2 3 8 4 1 2.20def 6 4 1 3 1 5
Du/DiI/Di 2.60bcde 1 3 5 6 4 1 1.65efg 5 5 4 4 2 0
I
Di/Du//Du 0.65jkl 14 2 2 1 1 0 0.00j 20 0 0 0 0 0
Du/Di/lDu 0.65jkl 14 2 2 1 1 0 0.10.)' 19 0 1 0 0 0
Hu/Du/lHu 0.75ijkl 8 9 3 0 0 O 0.85ghij 14 1 1 3 0 1
Du/Hul/Hu 0.60kl 13 4 1 2 0 0 1.20811 12 2 1 1 3 l
Hu/Du//Du 0.101 18 2 0 0 0 0 0.05j 19 1 0 0 0 0
Du/Hu/lDu 0.65jkl 15 l 1 2 1 0 0.00j 20 0 0 0 0 0
Total 212 85 75 85 72 31 300 55 30 39 54 82

Table A2. Estimates of standard deviation (sd) for carbohydrate factors
from Table 2. Values were calculated by taking the square root of
variance estimates (6’ l. A nonsignificant f test is assured to give an
expected sd of zero.

 

 

 

 

 

Mag” wt-
source f rufln spgar carbohydrate f n/ as ratio
Light 157 34 168 5.2
Cultivar 22 33 14 0.98
L X C 19 15 26 0.37
treatment 26 21 0 1.9
L X T 17 18 9 2.6
C_X T 0 0 0 0.66
Model and error mean am used in calculations for variability

estimates in Table A2;

Y1”: = 11 +14 +1?” +01: + (1.0)” + on” +11 + (LT)II + (C'Plnn +
(LCI‘hnu +Bnun

 

mm Mean Ms:

Source is

Light (L) 035 It 1:03. + 01:03:- + rct03n
Replication 035 + to”. + cto’r
Cultivar (C) 035 + to“. + rlte'c
LXC 03b+t02e+rtezlc
Error (a) 03b ‘1' to“. ‘
Treatment (T) 035 + r0103:

L X T 035 + r003”

C X T 03b '1’ r103“

LXCXT 035+r03nct

Error (b) 035

Legend:

1 = 6

r = 3

c = 3

t = 2

GD

3 = estimate of variation

REFERENCES

Alden, J., and R .K. Hermann. 1971. Aspects of the
cold-hardiness mechanism in plants. Botanical Review.
37: 37-142.

Amirshahi, M. C., and F. L. Patterson. 1956. Cold
resistance of parent varieties, F2 populations, and F3
lines of 20 oat crosses. Agron. J. 48: 184-188.

Collins, F. W., and K .R. Chandorkar. 1973. De novo
synthesis of fructooligosaccharides in leaf disks of
certain asterasceae. II. Certain physiological and

biochemical changes accompanying fructooligosaccharide
formation. Can. J. of Bot. 51: 1931-1937.

Chang, R. 1981. Physical Chemistry with Applicptiong
to Biological Systems. 2nd ed. Macmillan Publishing
Co., Inc. New York.

Eunus, A. M., L. P. V. Johnson, and R. Aksel. 1962.
Inheritance of winterhardiness in an eighteen-parent

diallel cross of barley. Can. J. Genet. Cytol. 4:
356-376.

Everson, E. H. 1952. The genetics of yield
differences associated with awn barbing in the barley
hybrid (Lion x Atlaslo) x Atlas. PhD. Thesis, Univ.
Caifornia.

Grafius, J. E. 1980. Breeding for winter hardiness.
In Analysis and Ipprovgpent of Plgnt Cold Hardiness.
CRC Press Inc. Boca Raton, Florida. p171.

Green, D. G. 1983. Soluble sugar changes occurring
during cold hardening of spring wheat, fall rye and
alfalfa. Can. J. of Plant Sci. 63: 415-420.

Grotelueschen, R.D., and D. Smith. 1968.
Carbohydrates in grasses. III. Estimatians of the
degree of polymerization of the fructosans on the stem
bases of timothy and bromegrass near seed maturity.
Crop Sci. 8: 210-212.

52

10.

ll.

12.

13.

l4.

15.

16.

17.

18.

19.

20.

21.

53

Gullord, M. 1974. Genetics of freezing hardiness in
winter wheat (Triticum aestivum L.). Ph.D Thesis.
Michigan State University.

Gullord, M., C. R. Olien, and E. H. Everson. 1975.
Evaluation of freezing hardiness in winter wheat.
Crop Sci. 15: 153-157.

Heber, U. and K. A. Santarius. 1973. Cell death by
cold and heat and resistance to extreme temperatures.
Mechanisms of hardening and dehardening. In:

Temperappge and Life. Springer-Verlag, New York.
p232-292.

Jenkins, G. 1969. Transgressive segregation for frost
resistance in hexaploid oats. (Avena spp.). J. Agric.
Sci. Camb. 73: 477-482.

Labhart, Ch., J. Nosberger, and C. J. Nelson. 1983.
Photosynthesis and degree of polymerization of fructan
during reproductive growth of meadow fescue at two
temperatures and two photon flux densities. J. of Exp.
Bot. 34: 1037-1046.

Law, C. N. and G. Jenkins. 1970. A genetic study of
cold resistance in wheat. Genet. Res. Camb. 15: 197-
208.

Le. H., D. A. Reicosky, C. R. Olien and C. E. Cress.
1986. Freezing hardiness of some accessions of
Triticum taushii and T. turgidum L. var. durum. Can.
J. Plant Sci. 66: 893-899.

Levitt, J. 1980. Responses of Plants to Environmental
Stress. 2nd Edition. Academic Press.

Levitt, J. 1978. An Overview of Freezing Injury and
Survival, and Its Interrelationships to Other Stresses.

In: Plant Cold Hardiness and Freezing Stress. Li, P.H.
and A. Sakai eds. Academic Press. p3-15.

Levitt, J. 1959. Effects of artificial increases in
sugar content on frost hardiness. Plant Physiol. 34:
401-402.

Lineberger, R. D., and P. L. Steponkus 1980.
Cryoprotection by glucose, sucrose and raffinose to
chloroplast thylakoids. Plant Physiol. 65: 298-304.

Livingston, D. P. III. 1985. A comparison of freeze-
induced carbohydrate changes in winter barley crowns.

M. S. thesis. Michigan State University. Department

of Crop and Soil Science.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

4

w

Martin. J. H. 1927. Comparative studies of winter
hardiness in wheat. J. of Agric. Res. 35:493-535.

McKersie B. D. and L. A. Hunt. 1987 Genotypic
differences in tolerance of ice encasement, low

temperature flooding and freezing in winter wheat.
Crop SCie 27: 860-863e

McMaster, G. 8., J. A. Morgan and W. O. Willis. 1987.
Effects of shading on winter wheat yield, spike

characteristics, and carbohydrate allocation. Crop
Sci. 27:967-973.

Olien, C. R. 1988. Carbohydrate distribution in
mildly frozen rye crowns. Crop Sci. in press.

Olien, C. R. and G. E. Lester, 1985. Freeze-induced
changes in soluble carbohydrates of rye. Crop Sci.
25: 288-290.

Olien, C. R. 1984. An adaptive response of rye to
freezing. Crop Sci. 24: 51-54.

Olien, C. R. 1980. Analysis of midwinter freezing
stress. In: Analysis and Improvement pf Plgnt Cold
flprdinegg. CRC Press Inc. Boca Raton, Floridp. p35-
59.

Olien, C. R. 1977. Ice adhesions in relation to
freeze stress. Plant Physiol. 60: 499-503.

Olien, C. R. 1974. Energies of freezing and frost
desiccation. Plant Physiol. 53: 764-767.

Olien, C. R. 1973. Thermodynamic components of
freezing stress. J. of Theor. Biol. 39: 201-210.

Olien, C. R. 1971. A comparison of desiccation and
freezing as stress vectors. Cryobiology. 8: 244-248.

Pollock. C. J. 1986. Environmental effects on sucrose
and fructan metabolism. In: Current Topics in Plant

Biochemistry and Physiology. 5: 32-46.

Pollock, C. J. 1984. Sucrose accumulation and the
initiation of fructan biosynthesis in Lolium
temultentum L.. The New Phytol., 96: 527-534.

Pollock, C. J. 1984. Physiology and metabolism of
sucrosyl-fructans. In: Storage Carbohydrates in
Vascular Plants Ed. D.H. Lewis. Cambridge Univ.
Press. p97-113.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

55

Ramachandra, R. A. and V. S. Rama Das. 1987.
Modulation of sucrose content by Fructose 2,6-
bisphosphate during photosynthesis in rice leaves
growing at different light intensities. J. of Exp.
Bot. 38: 828-833.

Roberts, D. W. A. 1986. Chromosomes in ’Cadet' and
'Rescue’ wheats carrying loci for cold hardiness and

vernalization response. Can. J. Genet. Cytol. 28:
991-997.

Rohde, C. R. and C. F. Pulham. 1960. Heritability
estimates of winter hardiness in winter barley
determined by the standard unit method of regression
analysis. Agron. J. 52: 584-586.

Salisbury Fb B. and Ross C. W. 1978. Plant
Phygiology. 2nd ed. Wadsworth Pub. Co. Inc. Belmont
Cal.

Santerius, K. A., 1973. The protective effect of
sugars on chloroplast membranes during temperature and
water stress and its relationship to frost, desiccation
and heat resistance. Planta. 113: 105-114.

Siminovitch, D. 1964. The quantitative estimation of
frost injury and resistance in black locust, alfalfa,
and wheat tissues by determination of amino acids and
other ninhydrin-reacting substances released after
thawing. Can. J. Bot. 42:637-649.

Steel, R. G. D. and J. H. Torrie. 1980. Principles
and Procedures of Statistics a Biometricpl Approach.
2nd ed. McGraw Hill Book Co. Pg.598-599.

Steponkus, P.L. 1978a. Responses to extreme
temperatures. Encyclopedia of Plant Physiol., New
Series. 12A: 385-402.

Steponkus, P.L. 1978b. Cold hardiness and freezing
injury of agronomic crops. Advances in Agronomy. 30:
51'98e

Steponkus, P.L. 1968. The relationship of
carbohydrates to cold acclimation of Hedera helix L.
cv. Thorndale. Pysiol. Plant. 21: 777-791.

Sugiyama, N., and T. Simura. 1967. Studies on the
varietal differentiation of frost resistance of the tea
plant. IV. The effects of sugar level combined with
protein in chloroplasts on the frost resistance.

Japan. Jor. Breed. 17: 292-296.

47.

48.

4£L

50.

51.

52.

53.

54.

56

Trunova, T. L. 1965. Light and temperature systems in
the hardening of winter wheat and the significance of
oligosaccharides for frost resistance. Fiziol. Rast.
12: 70-77.

Tumanov, I. I., T. I. Trunova, N A. Smirnova and G. N.
Zvereva. 1976. Role of light in development of frost
resistance of plants. Fiziol. Rast. 23: 109-114.

Wagner, W., F. Keller, and A. Wiemken. 1983. Fructan
metabolism in cereals: induction in leaves and
compartmentation in protoplasts and vacuoles. Z.
Pflanzenphysiol. 112: 359-372.

Wagner, W., A. Wiemken, P. Matile. 1986. Regulation
of fructan metabolism in leaves of barley (Hordeum
Vulgare L. cv. Gerbel). Plant Physiol. 81: 444-447.

Warnes, D. D., and V. A. Johnson. 1972. Crown-
freezing and natural survival comparisons of Fz-Fe bulk

populations from 25 winter barley crosses. Crop Sci.
12: 403-405.

Wiemken A., M. Frehner, F. Keller, and W. Wagner 1986.
Fructan metabolism, enzymology and compartmentation.
In: Current Topics in Plant Biochemistry and
Physiology. 5: 17-37.

Williams, R. J. 1980. Frost desiccation: An osmotic
model. In: Analysis and Improvepent of Plgnt Colg
Hardiness. CRC Press Inc. Boca Raton, Florida. p89-
115.

Worzella, W. W. 1935. Inheritance of cold resistance
in winter wheat, with preliminary studies on the
technic of artificial freezing tests. J. Agric. Res.
50: 625-635.