BIOCHEMICAL AND CHEMOTAXONOMIC
STUDIES OF PETUNIA HYBRIDA HORT.
AND SELECTED PEIUNIA SPECIES V

Thesis for the Degree of Ph. D.
MICHIGAN. STATE UNIVERSITY
NICHOLAS I. NATARELLA
1972

  
      

LIBRARY
F Michigan Scam
University

   

This is to certify that the

thesis entitled

BIOCHEMICAL AND CHEMOTAXONOMIC STUDIES OF
PETUNIA HYBRIDA HORT. AND SELECTED PETUNIA
.SPECIES

presented by

Nicholas J. Natarella

has been accepted towards fulfillmeht
of the requirements for

at

pun degree in We

 

 

M245 6 %:16 I

Meier mfm

Date July 6J 1972 ‘

0-7839

 

   
    

  
   

"human i‘v
HUAB & SUNS' N
am EIIIIJEEI' Inc, I ,
LIBRARY amr'ns H I‘
SPIllelnov .. . ». -
J I 3;: __

 

© 1972

NI CiiOLAS JOSEPH NATARELLA

ALL RIGHTS RESERVED

 

 

BIOCHE

An invest:
indoles frOm v5
Single: QQ. flc
there was any I
fIOWer 99notype

EXtractabl

Extractable fre

There was
acid Concn frcn
T

Qualitative dif_

 

ABSTRACT
BIOCHEMICAL AND CHEMOTAXONOMIC STUDIES OF

PETUNIA HYQRIDA HORT. AND SELECTED
PETUNIA SPECIES

BY

Nicholas J. Natarella

An investigation of free amino acids, phenolics and
indoles from various plant parts of double, 22 and 2d, and
single, dd, flowered petunia plants was made to detemmine if
there was any relationship between these substances and
flower genotype.

Extractable free amino acids and phenolics were analyzed
quantitatively and by thin-layer chromatography (TLC).
Extractable free indoles were analyzed by TLC.

There was no significant difference in total free amino
acid concn from shoot tips or anthers as related to genotype.
TLC analysis of amino acid extracts from shoot tips, buds
and leaves, of the three genotypes, failed to resolve any
qualitative differences using several solvent systems.
Anthers were squashed directly on thin-layer plates and
develoPed in several solvents. There were no qualitative
differences resolved between genotypes.

There were no quantitative or qualitative differnces
resolved by TLC in phenolics from shoot tips of the three

genotypes.

 

The a:
similar qua
It has
phenolics a
relationshi
A stud

hibitor 5 fr

Fresh ;
generally 9.:
destrUCtiOm

Weter~s
in nature, f
destruction
IP-JI °Xidase
PhaSe HOr th
lat“ with t:

Phenolic
c 4 Species

i

%'B.w

Nicholas J. Natarella

The analysis of indoles indicated that they are very
similar qualitatively for the genotypes investigated.

It has been found that extractable free amino acids,
phenolics and indoles, as analyzed in this study, have no
relationship to flower genotype.

A study of extractable IAA oxidase and IAA oxidase in-
hibitors from the shoot tips of the double, 22 and QQ, and
single, dd, flowered genotypes of flowering petunias was
made.

Fresh IAA oxidase preparations from the 3 genotypes
generally exhibited no lag—phase and similar rates of IAA
destruction.

Water-soluble, heat-stable fractions, probably phenolic
in nature, from the 3 genotypes, were found to inhibit the
destruction of IAA by horseradish peroxidase (HRP) and the
IAA oxidase preparations. Neither the length of the lag-
phase nor the subsequent rate of IAA destruction was corre-
lated with the genotypes investigated.

Phenolic compounds were investigated from leaf extracts
of 4 species of Petunia (Solanaceae), g. axillaris, g.
inflate, g, violacea and 3. parodii, the reciprocals of 5
interspecific crosses,'§5 axillaris x g. inflata, g. axillaris
x g. violacea, g, axillaris x g, parodii, g. violacea x g.
inflata and g. violacea x g. parodii, and 15 cultivars of
g. hydrida to determine the taxonomical relationships among

the species and the ancestry of g. hydride.

Quantita
of phenolic <

By the I
and two diff
were separat
dEZSitOIneter
3' flag a

EQLEEE an:

Fޣu13ted
”aching co
ho301090115
x 3. w
\- W
It was

Nicholas J. Natarella

Quantitative analysis revealed a broad continuous range
of phenolic concn, for the cultivars and species.

By the use of one-dimensional thin-layer chromatography
and two different solvent systems 17 distinct phenolic bands
were separated. The bands were analyzed for OD in a scanning
densitometer. There were significant correlations between

2. inflata and g. violacea, g. inflata and g, parodii, g.

 

violacea and g. parodii and g. axillaris and all of the culti-

 

vars. Of the interspecific crosses only g. axillaris x‘g.

parodii and g. axillaris x g. violacea were significantly cor—

 

related with the cultivars. g. axillaris x g. violacea was
calculated to be biochemically closer to the cultivars.
Matching coefficients, based on the presence or absence of
homologous bands, indicated a closer match between 3. axillaris
x g. violacea and the cultivars than between g. axillaris x

g. parodii and the cultivars.

It was concluded that the 4 species investigated are
properly classified as distinct species and that g. axillaris
and g. violacea are most likely the original parents of g.
hybrida based on the similarity of phenol banding patterns.

A study of the protein and peroxidase banding patterns ob-
tained by disc electrophoresis of leaf extracts from flowering
plants of g. axillaris, g, inflate, g. violacea, g. parodii and
11 cultivars of g. hybrida was made to determine taxonomical
relationships among the species and the ancestry of g. hybrida.

A total of 11 peroxidase enzymes were resolved from the

species and 2 additional enzymes from the cultivars.

E. axillaris

with 7 sites

 

bands and _P. ‘
patterns among
identical to .

The aver
all of the cu
Eéigéig,‘ .638,
were highly c

A total
SPECieS and C
sesseci identi

Sand 12' We:

P‘ ifif

\ N +
% “Ta

all

of the C11
The 4 S-
i~

closely relat

: .
‘5 distinct c.
ibcl

\‘ ata h

that P

\.%

D

Nicholas J. Natarella

g. axillaris and g, inflata had identical banding patterns

with 7 sites of enzyme activity resolved. g. violacea had 10

 

bands and g. parodii had 7 bands. The peroxidase banding
patterns among the cultivars were generally similar but not
identical to any of the species'.

The average correlations of the individual species with
all of the cultivars were significant at the 5% level:

2. axillaris, .74; g. inflata, .73; g. violacea, .52; and g.

 

 

parodii,'.68. All of the grandiflora and multiflora cultivars
were highly correlated.

A total of 48 protein bands were resolved from the
species and cultivars. No 2 species and/or cultivars pos—
sessed identical banding patterns. Three protein bands, 2,

8 and 12, were common to all species and cultivars. Only
g. inflata was significantly correlated, on the average, with
all of the cultivars.

The 4 species investigated have been determined to be
closely related phylogenetically although properly classified
as distinct species. The results also indicate that g.
inflata has been involved in the synthesis of g. hybrida but
that g, axillaris and, to a lesser degree, g. violacea and

g, parodii may have also contributed to its development.

 

BIOCH

 

BIOCHEMICAL AND CHEMOTAXONOMIC STUDIES OF

PETUNIA HYBRIDA HORT. AND SELECTED

 

PETUNIA SPECIES

BY

Nicholas J. Natarella

A THESIS

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

DO CTOR OF PHILOSOPHY

'Department of Horticulture

1972

ACKNOWLEDGMENTS

I wish to thank the members of my guidance committee,
Dr. David R. Dilley, Dr. Paul H. Rasmussen, Dr. Larry R.
Baker, Dr. Leo W. Mericle and especially Dr. James W.
Hanover, for their help while I conducted this research and
in the final preparation of this thesis. I am particularly
grateful to Dr. Kenneth C. Sink, my advisor, for the incalcu—
able assistance he has provided for the last four years.

I would also like to thank the graduate students,
especially Murray E. Hopping, Joerg Schoenherr and Zemadu
Worku, for their willingness to always discuss with me
problems encountered during this research.

Of course, the understanding and encouragement from
my wife, Peggy, and my son, Nicky, cannot go without mention.
To them, this and all subsequent accomplishments in my life,

are dedicated.

ii

 

 

"I m m ' I
LAMI‘AT
CF F

zinc!)UCIIOS

METHODS Am '

AminO A
IndoleS
phenols
Me
Me
:zsL'LTS
Amino A
IndOIes
enOls
35338103-
Amino A
Indoles
Phenols
SHERRY
LITEPATURE C
IA
Wt- o
«RDNDUCTICI‘
it
Prepa

TABLE OF CONTENTS

Page

QUANTITATIVE ANALYSIS AND THIN-LAYER CHROMATOGRAPHY
OF FREE AMINO ACIDS, INDOLES AND PHENOLS
OF DOUBLE FLOWERED PETUNIAS

H

INTRODUCTION. . . . . . . . . . . . . . . . . . . . .
METHODS AND MATERIALS . . . . . . . . . . . . . . . .

Amino Acids. . . . . . . . . . . . . . . . . . .
Indoles. . . . . . . . . . . . . . . . . . . . .
Phenols. . . . . . . . . . . . . . . . . . . . .
Method I. . . . . . . . . . . . . . . . . .
Method II . . . . . . . . . . . . . . . . .

\Jothbw w

\0

RESULTS . . . . . . . . . . . . . . . . . . . . . .
Amino Acids. . . . . . . . . . . . . . . . . . . 9
Indoles. . . . . . . . . . . . . . . . . . . . . 15
Phenols. . . . . . . . . . . . . . . . . . . . . 24

DISCUSSION. . . . . . . . . . . . . . . . . . . . . . 31
Amino Acids. . . . . . . . . . . . . . . . . . . 31
Indoles. . . . . . . . . . . . . . . . . . . . . 33
Phenols. . . . . . . . . . . . . . . . . . . . . 35

SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 37

LITERATURE CITED. . . . . . . . . . . . . . . . . . . 39

IAA OXIDASE AND INHIBITORS FROM NORMAL
AND DOUBLE FLOWERING PETUNIAS

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . 43
METHODS AND MATERIALS . . . . . . . . . . . . . . . . 44

Preparation of IAA Oxidase . . . . . . . . . . . 44

iii

' “B

 

4'." 1cl") ‘. I‘.

 

 

 

TABLE OF CO!

1AA Oxi
Assay i
Phenol

REEITS .

SUNIRY

LZTERAT'URE c

A (1334039:

0F Peru
\

RRCD'JCT 10:;

v3“ IAN
‘HWS AND,

TABLE OF CONTENTS--C0ntinued Page

IAA Oxidase Inhibitors . . . . . . . . . . . 45
Assay for IAA Oxidase and IAA Oxidase Inhibitors 45
Phenol Estimation. . . . . . . . . . . . . . . . 46
RESULTS . . . . . . . . . . . . . . . . . . . . . . . 47
IAA Oxidase. . . . . . . . . . . . . . . . 47
IAA Oxidase Inhibitors . . . . . . . . . . . . . 47
DISCUSSION. . . . . . . . . . . . . . . . . . . . . . 55
SUWMARY . . . . . . . . . . . . . . . . . . . . . . . 57
LITERATURE CITED. . . . . . . . . . . . . . . . . . . 58

A CHEMOTAXONGMIC STUDY OF THE PHENOLIC CONSTITUENTS
OF PETUNIA SPECIES, INTERSPECIFIC HYBRIDS AND
PETUNIA HYBRIDA HORT. CULTIVARS

 

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . 6O
METHODS AND.MATERIALS . . . . . . . . . . . . . . . . 63
Statistical Analysis . . . . . . . . . . . . . . 67

RESULTS . . . . . . . . . . . . . . . . . . . . . . . 69

Quantitative Analysis. . . . . . 69
TLC Analysis . . . . . . . . . . . . . . . . . . 71

DISCUSSION. . . . . . . . . . . . . . . . . . . . . . 76
SUMMARY . . . . . . . . . . . . . . . . . . . . . . . 80

LITERATURE CITED. . . . . . . . . . . . . . . . . . . 82

PROTEIN AND PEROXIDASE ELECTROPHORETIC ANALYSIS
OF SELECTED PETUNIA SPECIES AND CULTIVARS

INTRODUCTION. . . . . . . . . . . . . . . . . . . . . 85

IMETHODS AND MATERIALS . . . . . . . . . . . . . . . . 87

iv

 

TABLE OF CO!

RES ULTS
DISCUSSION .
Silvi‘viARY .

LITERATURE c

TABLE OF CONTENTS--Continued Page

RESULTS . . . . . . . . . . . . . . . . . . . . . . . 92
DISCUSSION. . . . . . . . . . . . . . . . . . . . . . 107
SUWHARY . ...... . . . . . ...... . . . . . . . . . 111

LITERATURE CITED. . . . . . . . . . . . . . . . . . . 113

 

Tsw-

 

nmEE

035331” ITAT IV

1.

I."
a

CF FR

Coach of
Parts of
zygous d.
.Q, PEtu.
EXper lite

IAA

C Total phi

acid per
dOuble a:
pQIUnia

LIST OF TABLES

TABLE

QUANTITATIVE ANALYSIS AND THIN-LAYER CHROMATOGRAPHY

l.

A CHEMOTAXONOMIC STUDY OF THE PHENOLIC CONSTITUENTS

OF FREE AMINO ACIDS, INDOLES AND PHENOLS
OF DOUBLE FLOWERED PETUNIAS

Concn of free amino acids from various plant

parts of homozygous double flowered, 22, hetero-
zygous double flowered, Dd, and single flowered,

dd, petunias. Observations are means of 3
experiments in ug/g fresh wt. . . . . . . . .

IAA OXIDASE AND INHIBITORS FROM NORMAL
AND DOUBLE FLOWERING PETUNIAS

Total phenol concn (ug equivalents of ferulic
acid per g fresh wt) from shoot tips of the
double and single flowering genotypes of
petunia . . . . . . . . . . . . . . . . . . .

OF PETUNIA SPECIES, INTERSPECIFIC HYBRIDS AND
PETUNIA HYBRIDA HORT. CULTIVARS

Source of Petunia species and cultivars . . .

Interspecific hybrids of Petunia species ana-
lyzed for their phenolic prOperties . . . . .

Total phenol content from leaves of species,
amphiploids and cultivars of Petunia (average
of 3 extraction experiments). . . . ... . . .

TLC patterns of phenols of Petunia species,
amphiploids and cultivars . . . . . . . . . .

Correlation coefficients and biochemical dis-

tance (in parentheses) of species and multi-
flora and grandiflora cultivars . . . . . . .

vi

Page

10

54

64

65

7O

72

74

LIST OF TABI.
TABLE

& Average
cal dis
with mu

7.Matchin
(in par

2. paro
all cul

PROTEIN
OF S

 

l.Source

'13tersp
lyzed f

‘Peroxid
and Cul

UV

bandin
CultiVa

'PQIOXid
SPeCifi

Cerel
Pe+ h.

LIST OF TABLES--Continued
TABLE

6. Average correlation coefficients and biochemi-
cal distances (in parentheses) of amphiploids
with multiflora and grandiflora cultivars. . . .

7. Matching coefficients and biochemical distances
(in parentheses) of amphiploids g. axillaris x
P. parodii and g. axillaris x‘g. violacea with
all cultivars. . . . . . . . . . . . . . . . . .

PROTEIN AND PEROXIDASE ELECTROPHORETIC ANALYSIS
OF SELECTED PETUNIA SPECIES AND CULTIVARS

1. Source of Petunia species and cultivars. . . . .

2. Interspecific hybrids of Petunia species ana-
vlyzed for protein and peroxidase banding '
patterns . . . . . . . . . . . . . . . . . . . .

3. Peroxidase banding patterns of Petunia species
and cultivars. . . . . . . . . . . . . . . . . .

4. Protein banding patterns of Petunia species and
cultivars. . . . . . . . . . . . . . . . . . . .

5. Correlation coefficients from the comparison of
banding patterns of Petunia species and
cultivars. . . . . . . .y. . . . . . . . . . . .

6. Peroxidase banding patterns of Petunia inter—
specific hybrids . . . . . . . . . . . . . . . .

7. Correlation coefficients of banding patterns of

Petunia species and interspecific crosses for
perox idase O O O O O I O O O O O O O O O O O O 0

vii

Page

75

79

88

89

93

97

103

104

106

 

FIGURES

CUANTITATI

.Ninhydr-
- Ninhydri
' Ninhydri

° Niflhycirj,

' Thin-

OF F

smears c
each ger

fme pet

 

EXtracts

' 2‘Dimens:

and homo;

'Thin‘laye

ShOQt ti;

ldyG
(Method I

i

LIST OF FIGURES

FIGURES

QUANTITATIVE ANALYSIS AND THIN-LAYER CHROMATOGRAPHY

OF FREE AMINO ACIDS, INDOLES AND PHENOLS
OF DOUBLE FLOWERED PETUNIAS

Ninhydrin-positive spots from 1, 2 and 3 anther
smears of each genotype. . . . . . . . . . . .

Ninhydrin—positive spots from anther smears of
each genotype. . . . . . . . . . . . . . . . . .

Ninhydrin-positive spots of amino acid extracts
from petals of each genotype . . . . . . . .

Ninhydrin-positive spots from amino acid
extracts from petals of each genotype. . . . .

2-Dimensional chromatograms of amino acid ex-
tracts from petals of heterozygous double, Dd,
and homozygous double, 22, genotypes . . . . . .

Thin-layer chromatograms of indole extract from
shoot tips of each genotype. . . . . . . . .

Thin-layer chromatograms of phenol extract
(Method II) from shoot tips of each genotype

Thin-layer chromatograms of phenol extract
(Method II) from shoot tips for each genotype
and 16 standards . . . . . . . . . . . . . .

Thin-layer chromatograms of phenol extract
(Method II) from shoot tips for each genotype
and l6 standards . . . . . . . . . . . . .

IAA OXIDASE AND INHIBITORS FROM NORMAL
AND DOUBLE FLOWERING PETUNIAS

Destruction of IAA by HRP (0.25 ng) and IAA
oxidase preparations (500 pg of total protein
from the shoot tips of double, DD and 2d, and
single, dd, flowered genotypes of petunia. .

viii

Page

12

14

17

19

21

23

26

28

30

49

 

LIST OF FIGU.

FIGURE

2.

16~30.

Lag-ph
inhibi
shoot '
flower.
tion 0,

- Lag-ph;

ration:
double:
of petI
(0.25)
shoot 1

PROTEIN

LIST OF FIGURES--Continued

FIGURE

2.

Lag—phase induction by ferulic acid (10 pg) and
inhibitor preparations (42 ug phenols) from the
shoot tips of double, DD and Dd, and single, dd,
flowered genotypes of petunia on the destruc-

tion of IAA by HRP (concn 0.25 pg). . . . . . .

Lag-phase induction by the inhibition prepa—
rations (42 pg phenols) from shoot tips of
double, DD and single, dd, flowered genotypes
of petunia on the destruction of IAA by HRP
(0.25‘pg) or IAA oxidase preparations from
shoot tips of the DD or dd genotypes. . . . . .

PROTEIN AND PEROXIDASE ELECTROPHORETIC ANALYSIS
OF SELECTED PETUNIA SPECIES AND CULTIVARS

1—15. Protein electrOphoretic patterns illustrating

16-30.

the range of variability observed among 4
species and 11 cultivars of Petunia . . . . . .

Peroxidase electrophoretic patterns illustrat-

ing the range of variability observed among 4
species and 11 cultivars of Petunia . . . . . .

ix

Page

51

53

95

100

 

QUANT I TA:
0?

QUANTITATIVE ANALYSIS AND THIN-LAYER CHROMATOGRAPHY
OF FREE AMINO ACIDS, INDOLES AND PHENOLS
OF DOUBLE FLOWERED PETUNIAS

 

An impc
understand 9
amino acid 5
differences
lar enzyme 0
organelle or
110w is game
the normal a
recognized a
be meaningfu
535 cellular

The ent
Phenotype,
EialysQS of

:ifferenCES

INTRODUCTION

An flmportant problem in higher plant genetics is to
understand gene action. Genes provide information for the
amino acid sequence of polypeptide chains. Single gene
differences determine the function and activity of a particu—
lar enzyme or protein which is a structural component of an
organelle or membrane (42). The question often posed is:
how is gene action mediated at the molecular level to produce
the normal and mutant phenotypes? Since the phenotype is
recognized as the secondary effect of gene action it should
be meaningful if the primary function at the biochemical
and cellular level could be determined.

The entire sequence of a gene's action, from DNA to
phenotype, is not known for a higher plant. Biochemical
analyses of single gene mutant forms have shown significant
differences from the normal character in amino acid metab-
olism (35,39,47) and phenolic acids (12). Unfortunately,
single gene mutants having a simple but obvious effect upon
the morphology of an organism have been used sparingly as an
aid toward understanding the biochemical aspects of morpho-
genesis (6,14,25,26,42).

The double flowering character in Petunia hybrida Hort.

is controlled by a single dominant gene, D, which is allelic

 

"ll

 

 

to single f
number of p
sterile bec
nalformed,
flO‘w'EIEO ty:
production.
hxozygo;s i
SiIsle flow:
with the dm
Fhemtypes c
CfOsses.

The Ont
investigated
“Future fr
:orphogeneti
are initiate

This in
t“Whip. if

PITEno ' .
11c 3C1

1
if p..G' W.

. ‘~‘dn' '
QOQDler la W
. S .

to single flowers, d.1 The double flower has an increased
number of petals and stamens and is essentially female
sterile because of either the absence or the formation of a
malformed, non-functional pistal (28). Hence, the double
flowered type can only be used as the male parent for seed
production. In commercial breeding practices, only the
homozygous genotype, DD, may be crossed to a female fertile
single flowered type, dd, for the production of hybrid seed
with the double flowering genotype. Presently, double flowered
phenotypes can be identified genotypically only by test
crosses.

The ontogeny of the double flowered mutant has been
investigated (28), and in spite of its radical phenotypic
departure from singleness, the D gene alters the normal
morphogenetic events only at the true petal and sepal primordia
are initiated.

This investigation was conducted to determine the rela-
tionship, if any, between free amino acids or indoles or

phenolic acids and floral genotypes in petunia.

 

1G. W. Scott, 1937. A genetical and cytological study
Of Petunia with special reference to the inheritance of
doubleness. Ph. D. Thesis, University of California, p. 42.

 

The 9
colored mu
ation. Thl
the hetero;
Eeneticallj
zygous rec:
in the gree

‘H
1‘...

ino Acids
were made f

in awe: .

METHODS AND MATERIALS

The genetic lines used in this study were salmon or rose
colored multiflora, sister lines, inbred to the fourth gener—
ation. The homozygous double line, designated MSU-499 and
the heterozygous double line, designated MSU—503, were
genetically determined by test crossing to the single, homo-
zygous recessive sister line, MSU-SOO. All plants were grown
in the greenhouse using standard cultural practices and

disease and insect preventative procedures.

Amino Acids

Free amino acid extractions for quantitative analysis
were made from tissues of plants vegetatively propagated and
in flower. The plant parts were 1) shoot tips: approx 10
mm long with all flower buds exhibiting anthocyanin develOp-
‘ment removed; 2) flower buds: less than 10 mm long with no
color development; 3) petals: from flowers on the day of
anthesis: 4) anthers: from flowers on day of anthesis but
4-6 hr prior to dehiscence. The tissues were removed from
the plants, sealed in individual plastic bags and stored in
powdered dry ice until extraction.

The extractions were carried out according to the pro-

cedure described by Pataki.(32),.with several modifications.

Fresh tiSSI
of 80% etha
were placed
16,000 xg f
l_ m at
813% ethanol
was determi:
method With

Only t}

were anaJ-YZE

 

3 anthers we
rod. The Ch
silica gel p
:‘L‘rmagram 5;
used and are
The cmmatog
light (LWLV)

hydrin Soluti
The freshly S

Fresh tissue (500 mg) was repeatedly homogenized with 10 m1
of 80% ethanol in an all glass tissue grinder. The tubes
were placed in boiling water for 10 min and centrifuged at
16,000 xg for 20 min. The supernatant solution was dried
ifl.!§£EQ at 36°C and the residue redissolved in 5 ml of

80% ethanol. The total free amino acid concn of each extract
was determined according to Rosen (33) using the ninhydrin
method with NH4Cl as the standard.

Only the amino acid extracts from the buds and petals
were analyzed by thin—layer chromatography (TLC). One, 2 or
3 anthers were squashed directly on TLC plates with a glass
rod. The chromatograms were run on 20 x 20 cm prepared
silica gel plates with a fluorescent indicator (Eastman
Chromagram Sheets-6061 F). Several solvent systems were
used and are designated in the apprOpriate figure legends.
The chromatograms were observed under long-wave ultraviolet
light (LWUV) before and after spraying with a 0.3% nin-
hydrin solution in n—butanol with 3.0% glacial acetic acid.
The freshly sprayed plates were heated at 60°C for 10 min

for Optimum color develOpment (41).

Indoles

The method for shoot tip collection was identical to
that for amino acids prior to extraction. Indoles were
extracted for qualitative analysis from shoot tips with a

‘modified procedure according to Srivastava (40) in subdued

 

light at 2'
10 ml of CI
The homogei
upernatant
The superne
was removed
0f a 5% aqu
tracted 2 t
"'35 used to
basic and n
an“‘lysiS. (
fraction CO:
discarded.

With Lo N i
ethyl ether
The ether 13‘:
resume was

Silica gel E

light at 2-4OC. Fresh tissue, 3.0 g, was homogenized with

10 ml of cold 90% methanol in an all glass tissue grinder.
The homogenate was centrifuged at 2000 xg for 15 min. The
supernatant solution decanted and the residue re-extracted.
The supernatant solutions were combined and the methanol

was removed $2.!éSEQ- The residue was dissolved in 100 ml

of a 5% aqueous solution of NaHCO3. The solution was ex-
tracted 2 times with 50 m1 of cold ethyl ether. The ether
was used to remove most of the remaining chlorophyll and the
basic and neutral indoles then stored at 2°C for further
analysis. Chromatographic analysis indicated that this ether
fraction contained only chlorOphyll residues and could be
discarded. The aqueous phase was then adjusted to pH of 3.5
with 1.0 N HCl and extracted with two 50 ml aliquots of cold
ethyl ether to separate out the acidic indole compounds.

The ether phase was evaporated to dryness dd £3222 and the
residue was dissolved in 2 m1 of ethyl ether. Prepared
silica gel plates were also used for separation of the
extracted indoles. The solvent system used is designated in
the apprOpriate figure legend. After separation, the chroma-
tograms were observed under LWUV and SWUV. The chromatograms
were also sprayed with the salkowski or Prochozka reagents

(41) for further identification.

Phenols
Two different extraction techniques were used. Both

fresh and lyOphilized plant material were analyzed.

 

Lyophil
Fresh shoot
exhibiting <
Tuna suffi
shoot tips y
petunia tiss
pulverized a
the 1Y°phili
3:011:10 to a
the N2 the E
bottles in a

iv-Ethod I

One hu:

were attract

°"'ernisht. i‘

Lyophilized plant material was prepared as follows:
Fresh shoot tips, approximately 10 mm.long, with all buds
exhibiting color removed, were immersed into liquid N2.

When a sufficient sample (ca. 10 9) had been collected the
shoot tips were 1y0philized. Because of the oily nature of
petunia tissues, the freeze-dried material could not be
pulverized at room temp so 50 m1 of liquid N; was poured over
the 1y0philized plant material in a mortar and then easily
ground to a fine powder with a pestle. After evaporation of
the N2 the powder was stored in lightly capped plastic
bottles in a dessicator dd 32222 over anhydrous CaClz. Under

these conditions the powders were stable up to 6 months.

Method I

One hundred mg of the powdered, 1y0philized shoot tips
were extracted with 5 m1 of a T% HCl solution in 80%.methanol
overnight, in the dark at room temp. The suspension was
centrifuged at 2000 xg for 30 min. The supernatant solution
was dried ig,ydgdd at 38°C. The residue was then dissolved
in 5 ml of dilute HCl, pH 5.0 and passed through a 1 cm x
1 cm dry packed column of silica gel G. The phenols were
eluted with 15 ml dilute HCIZ The eluate was dried dg.ydggd
and redissolved in 1.5'ml'1% HCl in 80% methanol. Appro-
priate aliquots were used fOr quantitative and chromato—

graphic analysis.

 

Method II
This s:
to Tomaszews
were extract
extracts we:
aglass fibe
used as a r1;
2 N HCl and
TLC indicate
ghl’ll breakc‘
tIacted 2 ti
aqueous Phas
to a anall ‘
Eth‘act Was
24°C for 1;
Total E
Eillis (43) '
extract frOn‘
:‘butanol in

Method II

This extraction method was a modified procedure according
to Tomaszewski (45). One hundred mg of freeze—dried powder
were extracted in 100 ml of boiling water for 20 min. The
extracts were then cooIed to 60-70°C and filtered through
a glass fiber filter with approx 50 ml additional hot water
used as a rinse. The filtrate was adjusted to pH 2.0 with
2 N HCl and extracted 3 times with ethyl ether. Preliminary
TLC indicated that the ether extract contained only chloro—
phyll breakdown products. The aqueous phase was then ex-
tracted 2 times with 50 ml aliquots of n-butanol and the
aqueous phase was discarded. The organic phase was reduced
to a small vol (1-2 ml) $3 22222 at 38°C. The n-butanol
extract was brought to a final vol of 15 m1 and stored at
2-4°C for 12-24 hr before analysis.

Total phenols were determined according to Swain and
Hillis (43), using ferulic acid as the standard. Only the
extract from Method I was quantitatively analyzed since the
nebutanol in Method II was not miscible with the aqueous
solvents used for quantitative analysis.

TLC was run on prepared silica gel chromagrams similar
to those used for the separation of amino acids and indoles.
The solvent systems used for TLC are defined in the appro-
priate figure legends. The chromatograms were observed under
iUWUV and SWUV before and after exposure to NH40H fumes.

The chromatograms were also sprayed with the Folin—Ciocalteu

.reagent according to Stahl (41), except commercial Folin-

 

Ciocalteu
wfler (to
II. The ;

Photc

h.)
(1'

To pi
of4, a 45
CClOR; SW1
exposure a
“Ch Visit
laps with

835. an f

Ciocalteu reagent was diluted 7 times its vol with
water (to approx 0.25 N) and substituted for spray reagent
II. The phenols were fully resolved in 5 min.

Photographs were recorded on'Kodachrome II film, ASA
25. To photograph under LWUV light required an f setting
of 4, a 45 sec exposure and'3 gelatin filters, 2E, 85C and
CClOR; SWUV light required an f setting of 4, a 25 sec
exposure and 2 gelatin filters, 2E and CC4OR. Photographs
with visible light were taken with two 32000 K photoflood
lamps with reflectors set at camera level, a gelatin filter,

80A, an f stOp of 5 and a 1/125 sec exposure.

 

All of

at least 3 ;

9:21:10 Ac ids
\

There
and concn
snot tips
buds of the
than the 13

identical.

RESULTS

All of the quantitative data presented are a result of

at least 3 separate experiments.

Amino Acids

There was no significant difference in total free amino
acid concn among the 3 genotypes assayed from anthers or
shoot tips (Table l). The total amino acid content in the
buds of the homozygous double, DD, was significantly lower
than the Dd and dd genotypes. The latter two were almost
identical. The total amino acid content of the flower petals
of the heterozygous double, Dd, was significantly higher
than the other two genotypes (Table l).

The TLC analysis of amino acids from petals and anthers
revealed no qualitative or quantitative differences. The
difficulty in collecting anthers of identical physiological
age from the double genotypes made replication of results
infeasible. Often, the same genotype exhibited greater
variability than that between genotypes. Anther smears,
'utilizing 3 different solvent systems, proved to be qualita-
tively similar (Figures 1 and 2) after spraying with nin-

hydrin and when observed under UV light. Tailing of all

 

Table 1

II

Tissue

Shoot t;

Flower 1:

"It.
“Inhers

*Q' .
“19mm

10

Table l. ‘Concn of free amino acids from various plant
parts of homozygous double flowered, DD, hetero-
zygous double flowered, Dd, and single flowered,
dd, petunias. Observations are means of 3 ex-
periments in ug/g fresh wt.

 

 

Free Amino Acid

 

Tissue Genotype Content
DD 1032.11
Shoot tips Dd 988.82
dd 966.66
DD 459.50*
Buds Dd 764.19
dd 760.31
DD 932.94
Flower Petals Dd 1264.l9*
dd 998.56
DD 543.01
Anthers Dd 524.02
dd 677.11

 

*Significantly different at the 5% level.

11

Aucoum uco>aom

um “camfluo "mv .mmucmuomwwo mmm Hmoemoa
IOMmmnm mo omomuon Sanmnoum ma zoflnz .momhuocom
cwnufl3 so>m .huwawnmwnm> oemuuxm oDoz .Hmums
ooaafluuwo "usm>aom .ommuocmm game no mucosa
Hogucm m can N .H scum muomm o>auemomIcHHowncHz

 

.H musmam

12

aQIDO":

 

H ouomflm

mar—.0730
10 10 in an

n w —

 

 

 

;
. a
J

 

 

 

 

l3

.Amuav noun:
.Hocmmoumlc "uco>aom any .Amumumv Hosm3 .Uaum
Deacon deflomam .Hocmusnlc upco>aom Am» .mcaa
Hum ponmmsvu ouo3 massage mouse .omhuocom game
no mucosa Hosucm scum muomm m>auflmomlcdup>ncaz

 

.N ounmflm

 

 

l4

 

dd

Dd

dd

Dd

GENOTYPE

 

spots resol

 

the number
have been r
The ch
trated in F
with either
2 dimension
the heteroz
differences

first solve

AlthOu
Moe in So
highly Vari
tion with u
under LWUV

A

two f0: any

Oatailied wa.

15

spots resolved was common with the anther smears, but with
the number of solvent systems used any differences should
have been resolved.

The chromatographéd amino acids from petals are illus-
trated in Figures 3 and 4. Identical patterns were obtained
with either spot or streak application methods. A series of
2 dimensional separations comparing the homozygote, DD and
the heterozygote, Dd, also failed to resolve any qualitative
differences (Figure 5). Tailing was a problem with the

first solvent system used in this separation.

Indoles

Although the indole extractions were carried out at
2-4°C in subdued light, the resulting chromatograms were
highly variable. Figure 6 illustrates a particular separa—
tion with uniform patterns for the acidic indole fraction
under LWUV and SWUV. Frequently, one genotype appeared quite
different quantitatively and/or qualitatively from the other
two for any one separation. Unfortunately, the pattern
obtained was not consistent for the same genotype. .Mann and
Jawarski (23) and Sagi (34) have reported on the instability
of auxins at several stages of various extraction techniques.
Serious quantitative loses can occur while reducing an ex-
tract in vol 12.22232: Mann and Jawarski (23) have also
shown quantitative losses occurring from the time an extract

is spotted to the end of a chromatographic run. They have

16

Figure 3. Ninhydrin-positive spots of amino acid
extracts from petals of each genotype.
Solvent: water.

 

 

 

18

 

 

.Amumumumubv nouns .MOemz Rm .owom
oeuoom Hmwomam .mcoumom .Hocmusnic on “Amuhv Houm3
.HocmmonmIc ADV “Amumumv Houm3 .Uflom ueumum Hmwomam

.HocmquIc Amy uuco>aom .oQMDOGmm some mo camped scum
muumuuxo page ocflam scum muomn m>auwmomIcwuoh£cwz

.a 3ng

 

 

 

 

O. D 00. mo

QOCD (no
0)@ sec

 

Idd
4

Del
GENOTYPE
Figure

 

20

.Amunv Houm3 .HocmmoumIc coauoouao com “Amnmumv
Houm3 .oaom owuoom deflomam .HocmusnIc coasoouao
umH. "muso>aom .mommuocom amm .oansop msomMN
Ioeon can .mm..oansoo msomwuoumuoa mo mamumm scum
muomuuxo oaom ocflem mo memumoumeonnu HocoflmcoEHQIN

 

.m whomflm

 

:§e’q

 

 

 

22

Figure 6. Thin—layer chromatograms of indole extract
from shoot tips of each genotype.
(a)-—Fluorescence pattern with LWUV.
(b)--Fluorescence pattern with SWUV.
Solvent: isOprOpanol, water (8:2).

 

23

 

be

366 nm

:FI‘I .
Iww~lnaiho
III
I

dd Dd DD dd Dd DD dd Dd DD

 

25 ul 50 ul 75 ul

 

 

6b
254 nm

 

dd Dd DD dd Dd DD dd Dd DD

250' 501” 750I

 

Figure 6

 

 

—a—
n...

.. t... ._.__
-.-.
‘.
v..—
-

 

also repor

ably to oxi

Phenols

There

between ger

tissue. A1

SOlvent SYS
differences
Standards c
Pound has b
This COHClu
systensl fl

reactiOn w

24

also reported that auxins are more readily lost, most prob-

ably to oxidative reactions, on silica gel than cellulose.

Phenols

There were no quantitative differences in total phenolics
between genotypes. The average concn of phenols was 1693 for
DD, 1914 for Dd and 1944 for dd, in ug/0.l g freeze-dried
tissue. Also, the two different extraction methods and two
solvent systems employed failed to resolve any qualitative
differences with TLC (Figure 7). Of a large number of phenol
standards co—chromatographed (Figures 8 and 9) only one com—
pound has been tentatively identified, chlorogenic acid.
This conclusion is based on similar Rf‘s in two solvent
systens, fluorescence patterns with UV light and the color

reaction with the Folin-Na2CO3 spray.

Figure 7.

25

Thin-layer chromatograms of phenol extract
(Method II) from shoot tips of each genotype.
(a) and (c)--Folin—positive spots;

(b) and (d)--F1uorescence pattern with LWUV.
Solvents: (a and b) n-butanol, glacial acetic
acid, water (8:2:3), organic phase; (c and d)
toluene, ethyl formate, formic acid (5:4:1).

SI

 

26

 

 

 

 

 

 

 

 

Figure 7

 

 

---. -h. w“__. ....i.

L .._ ._....._.._a- _-.

Figure 8.

27

Thin-layer chromatograms of phenol extract
(Method II) from shoot tips for each genotype
and 16 standards.

(a) Fluorescence pattern with LWUV;

(b) Fluorescence pattern with SWUV;

(c) Folin-positive spots.

Solvent: n-butanol, glacial acetic acid,

water (8:2:3), organic phase. Standards:

(1) t-cinnamic acid (lO‘ZM); (2) pyrocatechol
(lo-2M); (3) resorcinol (10‘2M ; (4) vanillin
(10'2M); (5) caffeic acid (10- M); (6) phloro-
glucinol (lO'ZM); (7) phenoxyacetic acid

(2,x lO-zM); (a) chlorogenic acid (5 x 10-3M);
(9) coumarin (lO'aM); (10) phthalic acid
(lo-2M); (11) phthalamide (lo-2M), (12) phtha1_
imide (lO'zM): (13) traumatic acid (10-2M):

(14) 2,4-dichlorophenol (lo-2M); (15) 2,4,5atri
chlorophenoxyacetic acid (lO'zM); (l6) 2,4-di-
chlor0phenoxyacetic acid (10-2M).

 

 

28

   
 
  

  

g O=“°“' um ®=ouv .= flu or "own
I

We '

  

    
  
 

'6 O=pmx ®=Iosr ®=onxlou .=vuout

WW '

 

  

s ------- . . . o a o 000000 s
I 2 3 4 5 6 7 I DD Dd dd 9 ID II I2 I) ll '5 16

     

 

 

Figure 9.

29

Thin-layer chromatograms of phenol extract
(Method II) from shoot tips for each genotype
and 16 standards.

(a) Fluorescence pattern with LWUV;

(b) Fluorescence pattern with SWUV;

(c) Folin-positive spots.

Solvent: toluene: ethyl formate, formic acid
(5:4:1).

Standards: See explanation in Figure 8.

 

 

30

'G O=nmr Qzuom uul ®=ouv game or uowu

 

®=onx no" .zwout

 

 

D ID II I2 18 H 15 ID

DISCUSSION

Amino Acids

 

Deviations in the amino acid metabolism in some floral
mutants can be directly related to the phenotype (35,39,47).
In this study, the double flowered mutant of petunia does
not deviate from the amino acid make-up of single flowered
phenotypes. Buzzati-Traverso (3) first reported identifying
heterozygotes from homozygotes as a result of qualitative
changes in ninhydrin positive spots using paper chromatog-
raphy and crude plant or insect extracts. He showed that
this method was sensitive enough to reveal genotypic dif-
ferences masked by complete dominance in tomato and muskmelon.
It is interesting that with root homogenates he was able to
distinguish the heterozygotes of a recessive muskmelon
mutant, yellow-green, which reduces the chlorOphyll content
by one-half in the homozygous recessive state while the
heterozygote contains the normal quantity of chlorOphyll.
Based on the results presented here, DD could not be identi-
fied from Dd or dd by differences in the concn of extract-
able amino acids. ’

There was no difference in total free amino acid con-
tent for the shoot tips, while it is in this plant tissue
that the D gene manifests its activity. Although the homo-

zygous double, DD, was significantly different from Dd and

31

 

dd in bud
different
types of p
ing amino
Althc
tive anal);
some biocr
sively EXC
or more pc
than 13 he
SincI
Situctura:
that the ‘
by "progr.
eariatiOn
a result ‘

tiOn' Or

the DNA.

32

‘gg in bud tissue and similarly the heterozygous, 9g, was
different from the homozygous dominant and recessive geno-
types of petals, general conclusions cannot be made concern-
ing amino acid differences in these genotypes.

Although total amino acid content, as well as qualita-
tive analysis with TLC, have not revealed genotype and hence
some biochemical effect of gene action, it does not conclu—
sively exclude them from an endogenous role. Of the 20
or more possible amino acids and their derivatives less
than 13 have been separated at maximum (Figure 5).

Since amino acids are the primary constituents of
structural proteins and enzymes, it is reasonable to suggest
that the mutant gene might affect the free amino acid pool
by "programing" a polypeptide requiring at least a single
variation in amino acid sequence. This change in sequence,
a result of a point mutation, would require either a dele-
tion, or addition or substitution in the base sequence of
the DNA. Such a change would in turn alter the amino acid
requirements for protein synthesis and change the free amino
acid pool. Whether such a change could be identified in the
composition of the pool becomes a function of several other
factors. Is the normal equilibrium, i.e., is the composition
of the free amino acid pool, disrupted or is it maintained
by another gene or set of genes? Here feed-back inhibition
and induction are of significance. If a disparity in the

free amino acid pool exists, is it then of a magnitude that

 

 

present I
easily irc
single am
seems to

pleiotrOp
and Sink

morphogen
Patal and
that prev;
are not a;
of the nun
activity,
Catalases

(6.19) , an

33

present research techniques can resolve? This situation is
easily imagined should the mutant polypeptide have only a
single amino acid change and be tissue and time specific as
seems to be the case in doubleness. There have been no
pleiotrOpic effects correlated with this mutant. Natarella
and Sink (28) have reported that the departure from normal
morphogenetic events occurs in the double at the time of
petal and stamen primordia initiation on the floral apex and
that previous and subsequent differentiation and deve10pment
are not affected. This interpretation is plausible in view
of the numerous reports of tissue-specific enzymes and enzyme
activity, particularly for peroxidases (6,10,15,19,36,37,38),
catalases (36,37), esterases (15,36,37), polyphenol oxidases

(6,19), and cellulase (7,24).

Indoles

Indoles were investigated since the apex is believed to
be the site of auxin synthesis and auxins are growth regu—
lators affecting cellular expansion (2,4,5,44) cell division
with cytokinins (27,29) and sex expression (11,17,18,30).
A disruption in any one or all of these effects could con—
ceivably cause an alteration in the morphogenetic events
leading to the development of double flowers in petunia.

In this study the indole concn for the genotypes investi-
gated appears very similar. This lack of variation implies
that if auxins or indoles are factors in the gene action of

doubleness it is probably in the kinetic activities of

 

 

_-. -
-—.-....- o-
-.
- ...._~

transport
sort of a
(10) have
tively re
the immed
was reduc
the apex
gradient .
gation by
in the hex
CatiOns 01
inhibitor
warrants a
°f action
of this III:
The n
be at the
numerOuS I

24.31) . E

tiVE analy
in the ape

An et
10954231 re

\

3

H .
effe . B.
BortftsMon

34

transport and/or metabolism, possibly in turn affecting some
sort of a gradient within the apex and stem. Galston gg_al.
(10) have demonstrated that 3-indoleacetic acid (IAA) effec-
tively represses the formation of a peroxidase isoenzyme in
the immediate vicinity of the apex but that as the IAA concn
was reduced this isoenzyme appeared at some distance from
the apex in the pith. The existence of a similar and active
gradient can be hypothesized from the results of an investi-
gation by Sui.3 She was able to partially reduce doubleness
in the heterozygous genotype of petunia with exogenous appli-
cations of TIBA (2,3,5-triiodobenzoic acid) which is a potent
inhibitor of auxin transport (l6,20,21,22). This evidence
warrants additional investigations into the presence and mode
of action of auxins and an auxin gradient at the shoot apex
of this mutant.

The mechanism of auxin action has been hypothesized to
be at the transcriptional level (Armstrong, 1) and
numerous references give credence to this Opinion (7,8,10,
24,31). Further research should therefore include a qualita-
tive analysis of enzymes, their repression and derepression,
in the apex during floral morphogenesis.

An effective method of IAA application to cause a physio—

logical response in petunia, in situ, is still lacking.

 

3H. N. Sui, 1969. Environmental and growth regulator
effects on the morphology of double flowering Petunia hybrida
Hort. (M. S. Thesis, Michigan State University, p. 58.

 

 

 

It is sug
attempted
IAA-inhib
genesis.

tion of t
could be

genesis t‘,
quent his
colld the]

Chemical (

Phenc
tiOn in t}
Previousl}
mess in P6
latory COf
orthmdih}
tion Jfeact
results 1

among the

35

It is suggested that apical meristem culture, igbyitrg, be
attempted to gather information as to the effect of IAA and
IAA-inhibitors at various points during floral morpho-
genesis. A method utilizing constant microsc0pic observa-
tion of the exposed apical meristem during morphogenesis

could be effective in determining the exact moments in morpho-
genesis that various treatment effects are manifest. Subse-
quent histological and micro-electrophoretic techniques

could then be correlated with the morphological and bio-

chemical consequences of the treatments.

Phenols

Phenols were investigated because of their participa-
tion in the IAA—oxidase systan (9,12,13,38,45,46) and the
previously mentioned implication of auxin action and double-
ness in petunia. Monophenols have been found to be stimu-
latory cofactors and polyphenols, especially those with an
ortho-dihydroxy configuration, inhibitory to the IAA oxida-
tion reaction with the appropriate peroxidase enzyme. TLC
results indicate no qualitative or quantitative differences
among the genotypes investigated. Nevertheless, an altera-
tion in the phenol pool may be too small to detect with the
Inethods employed, as previously discussed for the amino acids;
.Also the condition ig,§i£g.Where compartmentalization and
the vacuoles play a particular role in phenol concentration

twust be considered. Hence, while there may be a profound

 

active con

ation may

 

.: (I’vll‘ui! I... ,‘-,’Lfillui'y"1li ‘

36

active concn different for each genotype in vivo, the situ—

ation may not be evident in vitro after extraction.

 

 

.. ._. -. . ,‘fl‘-—*"”
w. -—-¢— -..4

An :
indoles 1
single, g
if there
flower ge

Extr
quantitat
EXtractab

The:
acid COnc
FIOWer bu.
heterogyg‘

total 31111}

S [MMARY

An investigation of free amino acids, phenolics and
indoles from various plant parts of double, 22 and QQ, and
single, dd, flowered petunia plants was made to determine
if there was any relationship between these substances and
flower genotype.

Extractable free amino acids and phenolics were analyzed
quantitatively and by thin-layer chromatography (TLC).
Extractable free indoles were analyzed by TLC.

There was no significant difference in total free amino
acid concn from shoot tips or anthers as related to genotype.
Flower buds with the homozygous, 2Q, genotype and leaves with
heterozygous, 2g, genotype were significantly different in
total amino acid concn from similar tissues of the other geno-
types. TLC analysis of amino acid extracts from shoot tips,
buds and leaves, of the three genotypes, failed to resolve
any qualitative differences using several solvent systems
and subsequently observed under long wave ultraviolet light
and after spraying with the ninhydrin reagent. Anthers were
also squashed directly on thin-layer plates and develOped in
several solvents. There were no qualitative differences
resolved.between genotypes.

There were no quantitative or qualitative differences
resolved by TLC in phenolics from shoot tips of the three

genotypes .

37

38

The analysis of indoles indicated that they are very
similar qualitatively for the genotypes investigated.

It has been found that extractable free amino acids,
phenolics and indoles, as analyzed in this study, have no

relationship to flower genotype.

 

(v .I: I'll.- lt- I’-I'.“cl.. '9‘: .

9.

10.

LITERATURE CITED

Armstrong, D. J. 1966. Hypothesis concerning the mechan—
ism of auxin action. Proc. Nat. Acad. Sci. 56, 64-66.

Bryan, H. and E. H. Newcomb. 1954. Stimulation of pectin
methylesterase activity of cultured tobacco pith by
indoleacetic acid. Physiol. Plantarum 7, 290-297.

Buzzati—Traverso, A. A. 1953. Paper chromatographic pat—
terns of genetically different tissues: A contribution
to the biochemical study of individuality. ProcgqNat.
Acad. Sci. 39, 376-391.

Cleland, R. 1968. Wall extensibility and the mechanism
of auxin-induced cell.elongation. lg: Biochemistrv and
thsioloqv of plant growth substances. F. Wightman and
G. Setterfield, ed. Runge Press Ltd., Ottawa, Canada.
Pp. 613-624.

and J. Bonner. 1956. The residual effect of
auxin in the cell wall. Plant Physiol. 31, 350—354.

 

Evans, J. J. and N. A. AIIdridge. 1965. The distribution
of peroxidases in extreme dwarf and normal tomato
(Lycopersicon esculentum Mill.). Phytochemistry 4, 499-
503.

. Fan, D. F. and G. A Maclachlan. 1966. Control of cellu—

lase activity by indoleacetic acid. Can. J. Bot. 44,
1025—1034.

and . 1967. Massive synthesis of

 

ribonucleic acid and cellulase in the pea epicotyl in
response to indoleacetic acid, with and without concur-
rent cell division. ”Plant Physiol. 42, 1114-1121.

Fox, L. R. and.W. K. Purves.‘ 1968. .Mechanism of enhance-
*ment of IAA oxidation by 2,4-dichlor0phenol.. Plant
PhXBiOl..43, 454-456.

Galston, A. W., S. Lavee and B. Z. Siegel. 1968. The
induction and repression of peroxidase isozymes by
3-indoleacetic acid. 13: Biochemistry and physiology
pf plant growth substances. F. Wightman and G. Setter-
field, ed. Runge Press Ltd., Ottawa, Canada. Pp. 455-472.

39

ll.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.

23.

4O

Galum, E. 1962. Culture and sex modification of male
cucumber buds in vitro. Nature 194, 596-598.

Gelinas, D. and S. N. Postlethwait. 1969. IAA oxidase
inhibitors from normal and mutant maize plants. Plant
Physiol. 44, 1553-1559.

Gorther, W. A., M. J. Kent and G. K. Sutherland. 1958.
Ferulic and p-coumaric acids in pineapple tissue as
modifiers of pineapple indoleacetic acid oxidase.
Nature 181, 630—631.

Gupta, V. and G. L. Stebbins. 1969. Peroxidase activity
in hooded and awned barley at successive stages of
development. Proc. Nat. Acad. Sci. 64, 15-24.

Hall, T. C., B. H..McCown, S. Desborough, R. C. McLeester
and G. E. Beck. 1969. A comparative investigation of
isoenzyme fractions separated from plant tissues.
Phytochemistryq8,'385-391.

Hertel, R. and A. C.Le0pold. 1963. Versuche zur analyse
des Auxin-transports in der KoleOptile von Zea mays
L. Planta 59, 535-562.

Heslop—Harrison, J. 1956. Auxin and sexuality in
Canabis sativa. Physiol. Plantarum 9, 588-597.

. 1963. The control of flower differentiation
and sex expression. In: Regulateurs de la croissance
veaetale. Pp. 649-664.

 

Hess, D. 1967. Mutiple Formen der Phenolase und Peroxy-
dase in reinen Linen von Petunia hybrida. g.
Pflanzenphysiol. 56,295—298.

Keitt, W. J. and R. A. Baker. 1966. Auxin activity of
substituted benzoic acids and their effect on polar
auxin transport. 'Plant Physiol. 41, 1561-1569.

and . 1967. AcrOpetal movement of
auxin: dependence on temperature. Science 156,
1380-1381.

 

Kuse, G. 1953. Effect of 2,,3,5-triodobensoic acid on
the growth of lateral bud and on tropism of petiole.
Memoirs Coll. Sci. Univer. Kyoto. Series B. 20, 3.

Mann, J. D. and E. G. Jawarski. 1970. Minimizing loss
of indoleacetic acid during purification of plant
extracts. Planta 92, 285-291.

24.

25.

26.

27.

28.

29.

30.

31.

32.

33.

34.

35.

41

Maclachlan, G. A., E. Davies and D. F. Fan. 1968.
Induction of cellulase by 3-indoleacetic acid. In:
Biochemistry and physiology of plant growth substances.
F. Wightman and G. Setterfield, ed. Runge‘Press Ltd.
Ottawa, Canada, pp. 443—453.

Mathan, D. S. and R. D. Cole. 1964. Comparative bio—
chemical study of two allelic forms of a gene affecting
leaf—shape in the tomato. Amer. J. Bot. 51, 560-566.

McComb. A. J. and J. A. McComb. 1970. Growth substances
and the relation between phenotype and genotype in
Pisum sativum. Planta 91, 235—245.

 

Miller, C. O. 1968. Naturally-occurring cytokinins.
‘lg; Biochemistry and physiolggy_9f_p1ant growth sub—
stances. F. Wightman and G. Setterfield, ed. Runge
Press Ltd., Ottawa, Canada, pp. 33-45.

Natarella, N. J. and K. C. Sink. 1971. Morphogenesis
of double flowering in Petunia hybrida Hort. J. Amer.
Soc. Hort. Sci. 96, 600—602.

 

Nitsch, J. P., E. Durtz. J. Livermann and F. W. Went.
1952. The development of sex expression in cucurbit
flowers. Amer. J. Bot. 10, 946-966.

. 1968. Studies on the mode of action of auxins,
cytokinins and gibberellins at the subcellular level.
In; Biochemistry and physiology of plant growth
substances. F. Wightman and G. Setterfield, ed. Runge
Press Ltd., Ottawa, Canada, p. 563-580.

 

Ockerse, R., B. Z. Siegel and A. W. Galston, 1966.
Hormone-induced repression of a peroxidase isoenzyme
in plant tissue. Science 151, 452-453.

Pataki, G. 1968. Techniques of thin-layer chromatography
in amino acids and peptides chemistry. Ann Arbor
Science Publisher, Inc., p. 61.

Rosen, H. 1956. A modified ninhydrin colorimetric
analysis for amino acids. Archives Biochan. BiOphys.
67, 10-15. -

Sagi, F. 1969. Silica gel or cellulose for the thin-

layer chromatography of indole-3-acetic acid.
J. Chromatog..39, 334-335.

Sarkissian, I., S. S. Shah and G. L. Stebbins. 1962.
Differences in free amino acid content of seedlings of
awned and hooded barley, and their alteratidn by'
chloramphenicol treatment. Proc. Nat. Acad. Sci. 48,
1513-1519.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

42

Scandalios, J. G. 1964. Tissue-specific isoenzyme vari—
ations in maize. J. Hered. 55, 281—285.

. 1969. Genetic control of multiple molecular
forms of enzymes in plants: A review. Biochem. Genetics
3' 37-390

 

 

Sheen, S. J. 1969. The destruction of polyphenols,
chlorogenic acid Oxidase and peroxidase in different
plant parts of tobacco, Nicotiana tobacum L.
Phytochemistry 8, 1839-1847.

Shortess, D. K., W. D. Bell and J. E. Wright. 1968.
The lutescent mutant in maize. II. Nitrogen metabolism.
Genetics 58, 237-248.

Srivastava, B. I. S. 1963. Ether-soluble and ether
insoluble auxins from immature corn kernels. Plant
Physiol. 38, 473—478.

Stahl, E. ed. 1969.'~ Thin-layer chromatography, 2nd
Ed. Springer-Verlag, N. Y., p. 1041.

Stebbins. G. L. and E. Yagil. 1966. The morphogenetic
effects of the hooded gene in barley. I. The course
of development in hooded and awned genotypes. Genetics
54. 727-741.

Swain, T. and W. E. Hillis. 1959. The phenol constituents
of Prunus domestica. I. The quantitative analysis of
constituents. J. Sci. Food Agric. 10, 63-68.

Thimann. K. V. and E. W. Samuel. 1955. The permeability
of potato tissue to warer. Proc. Nat. Acad. Sci. 41,
(1029-1033. ‘

Tomaszewski, M. 1964. The mechanism of synergistic ef-
fects between auxin and some natural phenolic substances.
lg: Regulateurs naturels de la croissance vegetale,
pp. 335-351.

and K. V. Thimann. 1966. Interactions of

phenolic acids, metalic ions and chelating agents on
auxin—induced growth. “Plant Physiol. 41, 1443—1454.

 

Wijewantha, R. T. and G. L. Stebbins. 1964. Development
and biochemical effects of the agropyroides mutation
in barley. Genetics 50, 65-80.

IAA OXIDASE AND INHIBITORS FROM NORMAL
AND DOUBLE FLOWERING PETUNIAS

INTRODUCTION

The double flowering character in Petunia hybrida

 

Hort. is conditioned by a single gene, 2, and is completely
dominant to g, the allele for the normal, single flowered
phenotype.1 The morphological ontogeny of this mutant has
been reported (13).

In other studies peroxidase and IAA oxidase activity
have been correlated with morphological mutants (2,3,5,8,9,
15). The mutants exhibit an increase in peroxidative
activity as compared to the normal. Studies in this labora-
tory have shown that triiodobenzoic acid, an inhibitor of
auxin transport, when applied"exogenously, can alter the
double phenotype in the direction of the single flowered
genotype.2

This study was undertaken to determine the IAA oxidase
activity and the role of extractable inhibitors in double
and single flowering genotypes of petunia during flower

bud development.

 

1G. W. Scott, 1937. A genetical and cytological study
of Petunia with special reference to the inheritance of
doubleness. Ph. D. Thesis, University of California, p. 42.

2H. N. Sui, 1969. Environmental and growth regulator
effects on the morphology of double flowering Petunia hybrida
Hort. .M. S. Thesis, Michigan State University, p. 58.

 

43

METHODS AND MATERIALS

The genetic lines used in this study were salmon
flowered, multiflora, sister lines inbred to the fourth
generation. The homozygous double line, 22, designated
MSU-499 and the heterozygous double line, Dd, designated
MSU-503 were identified genotypically by test—crossing to
the single, homozygous recessive, dd, sister line MSU-SOO.
All plants were grown in the greenhouse using standard
cultural practices and disease and insect preventative

procedures.

Preparation of IAA Oxidase

Shoot tips, 10 mm long, with buds showing any antho-
cyanin development removed, were collected from flowering
plants and stored in sealed polyethylene bags in powdered
dry ice. The tissue was homogenized within 30 min after
collection. All weighing and extraction procedures were
carried out at 2—4°C.

The extraction procedure was modified from that of
Gelinas and Postlethwait (3). The frozen shoot tips were
hOmOQenized in cold 0.1 M sodium phosphate buffer, pH 6.0
(5 m1.per g fresh wt of tissue) in an all glass tissue

grinder. The homogenate was filtered through 4 layers of

44

45

cheesecloth, and centrifuged at O-ZOC at 16,000 xg for 20
min. Ten ml portions of the supernatant solution were
filtered once through a 1.0 x 5.0 cm column of polyvinyl-
pyrrolidone powder (Polyclar AT) (6). A positive pressure
was applied to the column and the eluate collected. This
eluate was used to determine IAA oxidase activity. Protein
content was determined from 80%»ethanolic precipitates by

the method of Lowry et a1. (7).

IAA Oxidase Inhibitors

Aliquots (10 m1) of the supernatant after the initial
centrifugation of the IAA oxidase extract were heated for
10 min in boiling water. The extract was cleared by centrié
fugation at 16,000 xg for 20 min (3). This clear supernatant
was used as the source of IAA oxidase inhibitor and was

further diluted 1:20 with H20.

Assay for IAA Oxidase and IAA Oxidase
Inhibitors

The reaction mixture contained 0.2 mM IAA, 0.1 M sodium
phosphate buffer, pH 6.0, 0.1 mM dichlorOphenol (DCP), 0.1
mM MnCl; and an aliquot containing 500 pg total protein of
the IAA oxidase enzyme preparation in a final vblume of 20
ml. .The same reaction mixture was used to measure IAA oxi-
dase inhibitor activity except that horseradiSh peroxidase
(HRP) (5 mg, Sigma type VI, RZ 3.4) was substituted for the
IAA oxidase preparation, and an aliquot containing 42 lug of

phenols of the diluted inhibitor preparation was added.

46

The destruction of IAA was followed at various time
intervals during the reaction by adding 1.0 ml portions of
the reaction mixture to 2.0 m1 of Salper's reagent (4).

The tubes were allowed to stand in the dark, at room temp
for 1 hr before the OD was recorded at 525 nm. The results

are plotted as pg IAA destroyed with time.

PhengliEstimation

Total phenol concn of the IAA oxidase inhibitor prepara-
tions was estimated by the method of Swain and Hillis (16).
An apprOpriate aliquot of‘the undiluted inhibitor preparation
was mixed with 0.5 ml of 0.25 N Folin phenol reagent. After
incubation at room temp for 3 min, 1.0 ml of 1N Na2CO3 was
added and the total vol brought to 4 ml with H20. After a
second incubation at room temp for 60 min the OD was recorded
at 725 nm and converted to pg equivalents of ferulic acid
(4—hydroxy—3amethoxycinnamic acid). It was assumed that most
of the enzymes in the inhibitor extract were precipitated by
heating and that the phenol determination represented only

non-protein phenols.

RESULTS

IAA Oxidase

IAA oxidase activity was demonstrated in the petunia
extracts (Figure 1). Fresh extracts generally exhibited
no lag-phase and similar rates of activity were observed

with all genotypes.

IAA Oxidase Inhibitors
Quantitative analysis of the inhibitor extracts indi-
cated there was no correlation of phenol concn with genotype
(Table 1). For this reason all inhibition studies were run
with equivalent pg aliquots of the inhibitor extracts to
determine if any qualitative differences existed. Figure 2
illustrates that the shoot tips of petunia contained a water
soluble moiety capable of inhibiting the destruction of
IAA by HRP. Equivalent amounts of the inhibitors induced
a 10 to 20 min lag-phase which did not appear to affect the
rate of degradation once the initial inhibition was overcome.
Ten pg of ferulic acid produced a lag-phase similar to the
inhibitor extracts. NEither the length of the lag-phase nor'
the rate of destruction of IAA was correlated with genotype.
The reciprocal effects of inhibitor and IAA oxidase

extracts were investigated (Figure 3). Even though a

47

48

 

Figure 1. Destruction of IAA by HRP (0.25 pg) and IAA
oxidase preparations (500 pg of total protein)
from the shoot tips of double, D_Q and pd, and
single, dd, flowered genotypes of petunia.

o-.. -.o. -- .._..—___.

. “7*... .--.—

no IAA DESTIOYED

35

30

25

20

10

49

\.
\
\
1)
h

OXIDASE
SOURCE

 

.fi HRP

\.

x—x Dd

.—. dd

 

 

 

 

\\

\,

 

IO 20 30 40 50 60 80

TIME (min)

Figure 1

50

.81 m~.o 5.83 BE an 5: mo
fimwvopuumop ecu no megsumm mo nonhuogmm pmuo3oam
do .39.: ea... .8 can a .0336 no use» ooofi

on» eoum Anaogmnm m1.mcv mGOAHMHmmmHm Houwnwggw
paw Am:.oav pwom owasuom m9 goauospga ommnmlmmq

 

.N munmflm

‘1‘1..."llull! 1 ii..- y".ro\ t«'.,$‘l‘ ‘

 

51

 

 

 

 

 

N oupmflm
72:: m5..—.
cop On Go On Ov On ON or
u q q u d u
11 III
In xlx I
no 41 O

O.U( U.._3Iun Cl.
uZOZ OIO

uUIfiOn
80h...82.
O

 

 

 

\ xx.

. .\.\\ .

 

 

 

 

Op

mp

ON

.mN

On

an

GBAOHLSSO VVI

52

.momhaocmm mm.

Ho mm on» no and» Doors scum mGOflumummoHQ mnmpflxo
5: no 3a 2.8 Ems an .53 no 5303306 05
no magnuom mo nommuogom pmuoSOHw .mm .mamcflm ppm
.mm .oansop no one» uoonu scum Amaoconm ma.mvv
unawumummoum Houwnangfi on» an goauospcw ommnmlmmq

 

.m musmwm

53

 

 

 

 

 

 

ELEM] . .s rushinwx
m mnpmwm
.¢_E.u'¢.h
00— 01— 08— 00' on 00 on 0' 00 ON 0—
4 _
an 3.
on $1
an an
00 an
up as:
00 L81
KO».-.IZ_ um(0.x0
"so .9830.»

 

 

 

 

 

 

O

O—

on

ma

on

an

031081530 VVI

 

II : Ii 1" l-‘l!.‘|oi.',‘1"l

54

Table 1. Total phenol concn (pg equivalents of ferulic
acid per g fresh wt) from shoot tips of the double
and single flowering genotypes of petunia.

 

 

 

Genotype pg phenols/g fresh wt Percent of dd
DD 905 106 ‘ .3
905 106 . '
Dd 800 94
799 94
dd 850 i
852 4

 

 

lag-phase was induced by the inhibitor extract upon the IAA
oxidase extracts there was no qualitative difference in
inhibitors between the homozygous, DD and dd, genotypes.

The 22 and dd inhibitor extracts induced a 10 min lag-phase
with HRP and a 30 to 40 min lag-phase with the DD and dd
enzyme preparations. 'Similar results were observed when the'
heterozygous, Dd, extracts were compared to the homozygous

recessive, dd, extracts.

 

 

 

DISCUSSION

IAA oxidase activity, with or without extractable in—
hibitors, and the double flowering genotypes were completely
independent for this study. The IAA oxidase preparations
from the double and single flowered genotypes were similar
in their ability to destroy IAA. The activity of a crude
extract could not be assayed because extracts which were not
immediately passed through the PVP column soon turned brown,
even at 2-4OC. This browning was interpreted as the oxida-
tion of phenols to quinones which in turn have been reported
to oxidize proteins and inhibit enzyme activity (6).

It was assumed that while gene action is evident in
the flowering apex during the initiation of primordia (13),
the biochemical consequences of this mutant could be identi—
fied in the preflowering apex and in numerous flower buds
at all stages of develOpment prior to anthocyanin formation.

This assumption may be too broad. The results do not
preclude the possibility of differential activity of a gene—

mediated biochemical reaction in the precise area of petal

and stamen initiation. Micro-analysis of a sample consisting

of a more precisely defined portion of the shoot tip, i.e.,

the young, flowering meristems, might resolve the question

of protein or enzymatic differences as a result of the mutant

DD genotype.
55

.M.-#flw“ -Q‘n-x‘...) .-

56

The nature of the inhibitor, i_rl yi_vd, has not been
determined. El—Basyouni and Neish (1) and Mast (11,12) have
reported the existence of protein bound phenols and membrane
bound peroxidases, respectively. Should such complexes be
the active moieties 19. £29.: it is possible that they might
have been destroyed during boiling of the inhibitor extract
and removed from the enzyme extraction with filtration.
Fractionation of the enzyme extract with (NH4)4SO4 or
Sephadex gel filtration prior to analysis may further resolve
this question. A less rigorous procedure for inhibitor
purification, e.g., lead acetate precipitation, should also

be further investigated.

.5;

 

SUWMARY

A study of extractable IAA oxidase and IAA oxidase in-
hibitors from the shoot tips of the double, 22 and 2d, and
single, dd, flowered genotypes of flowering petunias was
made.

Fresh IAA oxidase preparations from the 3 genotypes
generally exhibited no lag-phase and similar rates of IAA
destruction.

Water-soluble, heat—stable fractions, probably phenolic
in nature, from the 3 genotypes, were found to inhibit the
destruction of IAA by horseradish peroxidase (HRP) and the

IAA oxidase preparations; The inhibitor fractions behaved

similar to ferulic acid which induced a 10-20 min langhase.
In all combinations of inhibitor fraction and IAA oxidase
preparation or HRP, neither the length of the lag—phase nor

the subsequent rate of IAA destruction was correlated with

the genotypes investigated.

57

 

 

l.

2.

3.

4.

10.

LI TERATURE CITED

El-Basyouni, S. Z. and A. C. Neish. 1966. Occurrence of:
metabolically-active bound forms of cinnamic acid.and
its phenolic derivatives in acetone powders of wheat
and barley plants. Phytochemistryfs, 683-691.

Evans, J. J. and N. A. Alldridge. 1965. The distribu-
tion of peroxidases in extreme dwarf and normal tomato.
(Lycopersicon esculentum Mill.) Phytochemistry 4, 499-

503.

Gelinas, D. and S. N. Postlethwait. 1969. IAA oxidase
inhibitors from normal and mutant maize plants.
Plant Physiol. 44, 1553-1559.

Gordon, S. A. and R. P. Weber. 1951. Colorimetric esti-
mation of indoleacetic acid. Plant Physiol. 26, 192-

195.

Gupta, V. and G. L. Stebbins. 1969. Peroxidase activity
in hooded and awned barley at successive stages of
development. Biochem. Genet. 3, 15-24.

Loomis, W. D. and J. Battalie. 1966. Plant phenolic
compounds and the isolation of plant enzymes.
Phytochemistry 5, 423-438.

Lowry, O. H., N. J. Rosebrough, A. L. Farr and R. J.
Randall. 1951. Protein measurement with the folin
phenol reagent. .JOur. Biol. Chem. 193, 265—275.

Mathan, D. S. and R. D. Cole. 1964. Comparative bio—
chemical study of two allelic forms of a gene affect-
ing leaf-shape in the tomato. Amer. J. Bot. 51, 560—

566.

. 1965. Phenylboric acid, a chemical agent
snmulating the effect of the lanceolate gene in the
tomato. Amer. J. Bot. 52, 185- 192.

 

Mast, C. A. Van Der. 1970. Isoelectric focusing of
indolacetic acid degrading enzymes from pea roots.
Acta Bot. Neerl. 19, 363-372.

58

11.

12.

13.

14.

15.

16.

59

Mast, C. A. Van Der. 1970. The presence of membrane-
bound IAA degrading protein complexes in homogenates of
pea roots and the manner of attachment to these mem—
branes. Acta Bot. Neerl. 19, 546-559.

1970. The influence of membrane-bound peroxi—
dages on the degradation of IAA by the specific IAA
degrading protein complexes in homogenates of pea roots.
Acta Bot. Neerl. 19, 727-736.

 

Natarella, N. J. and K. C. Sink. 1971. The moropho-
genesis of double flowering in Petunia hybrida Hort.
J. Amer. Soc. Hort. Sci. 96, 600—602.

Ockerse, R., B. Z. Siegel and A. W. Galston. 1966.
Hormone-induced repression of a peroxidase isozyme in
plant tissue. Science 151, 452-453.

Stebbins, G. L. and V. K. Gupta. 1969. The relation
between peroxidase activity and the morphological
expression of the hooded gene in barley. Proc. Nat.
Acad. Sci. 64, 50-56.

Swain, T. and W. E. Hillis. 1959. The phenolic constit-
uents of Prunus domestica. I. The quantitative analysis
of phenolic constituents. J. Sci. Food Agric. 10,
63-68.

 

 

 

 

A CHEWOTAXONOMIC STUDY OF THE PHENOLIC CONSTITUENTS
OF PETUNIA SPECIES, INTERSPECIFIC HYBRIDS AND
PETUNIA HYBRIDA HORT. CULTIVARS

W m-Mifimiww

a g.
\det—s.»

n:!' .23

 

1.11111 I'll. L“! u’-l'.. lg: . . l

INTRODUCTION

The genus Petunia (Solanaceae) contains at least 30
species indigenous to Central and South America, and parts
of Mexico and the southern United States with its principle

area of dispersion in Brazil (22) . Ferguson and Ottley (8)

report that g. axillaris (Lam.) B. S. P. was described and
in

introduced into Europe in 1793 and g. violacea Lindl.

It is then often stated that the numerous color vari-

 

 

1833.
ations that later appeared in several EurOpean gardens were

 

the progeny of an interspecific cross between 2. axillaris

and g. viodacea, i.e., g. hybrida Hort. (3,8,11). There have

been no morphological or biochemical investigations to

determine the validity of this assumption. In fact, the

validity of this hypothesis is questionable since 2 other

species which closely resemble the putative parents are also

widely cultivated. P. inflata R. Fries described in EurOpe

in 1911 has recently been re-classified by Smith (23) , who

based his Opinion only "on the gross morphological character-

istics. He‘has placed both 113. violacea and g. inflata under
_13. intgrifolia (Hook.)""Schinz and Tellung var. integrifolia.
Lamprecht (16) has gone farther listing both _P_. violacea and
g. inflata as synonymes of 13. axillaris. Similarly, his

60

 

,4‘\.‘:.‘ _ ”e
H
g.

61

decision was founded on the great variability of many morpho-

logical characteristics, as well as the work of Mather and

Edwardes (18) . They showed that at least two genes, both

affecting anthocyanin formation, are allelic among the 2
Steere (27) has

1
‘ V
\-

 

g-{n .-

{lam' ¥
JD

species _P. axillaris and'g. violacea.
reported from cytological studies that g. hybrida is prob—

mam-1m

ably a composite form of the .3 species.

To add to the confusion of the probable ancestry of

hybrida, Steere (26) in 1930 described a new species,

_13.
P. parodii W. C. S. which appears remarkably similar to g.
axillaris. This species does appear distinct from g. axil-

 

laris, when they are observed together. It is possible that

an earlier EurOpean introduction may have been assumed to

be merely a variant of _P. axillaris and therefore was not

described or classified.
Because of the morphological similarities between 3.

 

axillaris and g. parodii and between 3. violacea and _P_.

inflata we believe that all four species should be considered
in a study of their chemical relationships to one another

and to g. hybrida.

_13. hybrida is an important horticultural crop. Many

years of intensive practical breeding have developed it into

an extremely polymorphic species which appears quite distant

from. its presumed ancestry.
Thin-layer chromatography (TLC) of phenolic compounds

has proved to be. a useful technique in the investigation of

62

species relationships (6,14,20,21,29) and of hybrid popula—
tions and their origins (7,10,13,15,l9).

An earlier argument in favor of using phenolics in
chemotaxonomic investigations was that there was little
selective pressure exerted upon phenolics because no physio-
logical function could be readily attributed to them (1) .
This is not an acceptable argument since phenolic compounds“
are known to be cofactors or inhibitors to peroxidase
enzymes. However, the fact that there exists a large number
of phenolics with numerous possibilities for chemical sub—
stitution at different positions on their basic skeletons
remains a valid point in favor of their continued use (2).‘“

The objective of this study was to determine whether
the biochemical properties of phenolics from the species and
cultivars would add to and/or substantiate the recognized

species relationships as well as the presumed ancestry of

P. hybrida.

 

METHODS AND MATERIALS

The plants used for this study were grown from seed
obtained from a number of sources (Table 1) and from synthe-
sized amphiploids (Table 2). Two general classes of culti-
vars, multiflora and grandiflora, have been analyzed for
this study. These are allelic characters which affect
general growth characteristics and to some extent flower
size.1 All of the multiflora cultivars are homOzygous
recessive, gg, and all of the grandiflora cultivars are
heterozygous, g9. The 4 species used in this study are ap-
parently homozygous recessive for the multiflora allele, gq
(K. C. Sink, unpublished data). The seeds were germinated
on vermiculite in a growth chamber maintained at 22°:2C
with a 16 hr photoperiod and 10760 lux. Young seedlings
‘were transplanted into a freshly pasteurized soil mix (soil,
peat moss, perlite, 1:1:1, v/v/v) and grown in the greenhouse
to flowering. All cultural practices and disease and insect

jpreventative procedures were followed.

 

1L. C. Ewart, 1963. The inheritance of flower size in
grandiflora and multiflora petunias. Ph.D. Thesis. The
Pennsylvania State University, p. 73.

63

64

 

 

.ommuogmgm muoawepcmnm u 0 Ho muoameudsa u SN

 

 

 

 

.UCH ..OO meuumm smooch E moo .mmo pom.
.ogH ..OO manumm smooch 0 gmw .wcmx Mpcmx.
.ocH ..OO mennmm smooch 0 Amp .mocmmw~m.wmom.

.UGH mpowmmnueEmpHoo w «on .coozvumm>umm.

.UCH mpommnsuflEmpHow o mom .pflmsmmpfium.

.UGH mpmmm.£ufl8mpaoo 2 Non .mon whom.
uoouw pcm mwpam .> .z 0 who .Enoum3ogm.
uoouo cam macaw .> .z 2 Hum .uoaumom oumammm.
uoouw ppm madam .> .z 2 one .pom oumwmmm.
hammeou paw mumxmm .9 z Nmo ..QEH mumuufiaw.
mamdsoo ecm momxmm .e o mom .uoommz_xcem.
hammeoo pcm mumxmm .8 w mum .oapgmz_wpam.

.oo comm amoeuosaucmm 2 Hon .maamm mmcmuo.

.OO poem conflumemlcmm 2 con .mwmzwx.

.OO poem appeasemlgmm o mam .owmmz msam. V
omum>fluaso .unom avenger .M
mcmpumo amoegmuomszmom z mom mmflum .m mumamcfl .M
Empumumee .mnuegmuom msuuom z goo .Hpceq mmumaoep .M
.00 poem cmofiumadlcmm 2 hmm .m .U .3 eflpoumm .M
ngmpumw HMUdgmuom ammom 2 wow .m .m .m Agony mflumaawxm (M
HMdeumE HO moudom NU HO 2 sz.HOQBDZ GOXMB
roammmoo<

 

 

 

 

.mum>enaso paw moeoomm

I!

mflcsumm no mousse .- »-:;

65

Table 2. Interspecific hybrids of Petunia species analyzed
for their phenolic properties. '

 

 

 

 

 

 

 

 

 

Interspecific hybrids Pedigree No. MSU
(g. inflata x axillaris 195
.2. inflata x violacea 196
(g. axillaris x inflata 199
.g. axillaris x parodii 200
_g. axillaris x violacea 201
.E- parodii x axillaris 204
_g. parodii x violacea 205
.2. violacea x inflata 207
.2. violacea x axillaris 208
_P. violacea x parodii 209

 

zThe P. parodii x P. inflata cross could not be synthesized.

Samples were collected for analysis when all plants
*were in flower. .Sample tissue consisted of the two opposite
leaves subtending the three uppenmost flowers in full bloom
along all stems, and all leaves above these flowers at the
base of the buds. The leaves were immediately immersed in
liquid N; upon collection and lyophilized. Because of the
oily nature of the leaves the 1y0philized plant material was
‘placed in a mortar, covered with liquid N2 and ground to a
fine powder with a pestle. When the liquid N; evaporated the
Exmnders were placed into tightly st0ppered plastic jars.
After the jars reached room temp, the caps were loosened and
all samples were then stored in a desiccator over silica gel

under vacuum .

66

A modified procedure according to BateSSmith (4,5) was
used for extraction of phenols. Lyophilized tissue (0.15 g)
was added to 5 ml of 2 N HCI in a test tube and heated for
20 min in boiling water. The cooled extracts were filtered
through a glass fiber filter (Millipore, AP2002200) into a
narrow 10 ml test tube under vacuum. After the filtrate was
observed for color, 0.5 m1 of isOpentyl alcohol was added.
This was shaken vigorously and allowed to form 2 separate
layers. Ten p1 aliquots of the upper alcohol layer were
applied as 2 cm streaks on prepared cellulose thin-layer
plates, 20 cm x 20 cm (Avicel, 250 p thick, Analtech, Inc.).
The chromatograms were deve10ped in one direction with two
different solvent systems by the ascending method. The BEAW
solvent (12) (benzene, glacial acetic acid, water, 6:7:3,
v/v/v) was mixed and the phases allowed to equilibrate 12 hr.
The upper, organic phase was used to develop the chromato—
grams. The CAW solvent system (25) (chloroform, glacial
acetic acid, water, 50:45:5, v/v/v), was mixed immediately
prior to use. The spotted chromatograms were allowed to
equilibrate 8 to 12 hr over the CAW solvent and over the
lower, aqueous phase of the BEAW solvent in the dark at room
temp. The chromatograms were photographed under visible and
long wave ultraviolet light (LWUV) after fuming with ammonia
and then under LWUV after spraying with Benedict's reagent
(25). Photographs of the chromatograms under LWUV light

'were recorded on Kodachrome II with a combination of 2E, 85C

 

67

and CClOR filters, an f stOp of 5.6 and a 40 sec exposure.
The resulting slides were made into 4 x 5 black and white
negatives of uniform size. Each phenol separation was cut
out and taped to a glass slide and then scanned at 535 nm in

a Gilford densitometer. The print-out was then analyzed for

"i ' .7
' :lf._r (A.,-Ii

number of peaks, and the OD of each peak. Homologous peaks

.‘s '.l-‘t‘

were identified by their Rf's and color reactions.

The alcohol extract was analyzed quantitatively (28).

 

Although isoPentyl alcohol is not miscible in aqueous solu- :
tions, the small aliquots used for analysis (25-50 pl) were
completely miscible if the Folin reagent was added to the

water prior to the sample. The results are based onlpg

equivalents of ferulic acid (4-hydroxy-3-methoxycinnamic

acid).

Statistical Analysis

Many statistical methods are used to compare different
taxonomical units (24). Three of the methods most often
used because of their apparent reliability are the common
correlation coefficient, biochemical distance (15) (the
taxonomical distance of Sokal and Sneath, 24) and the matching
coefficient (24). The correlation coefficient and biochemical
distance were calculated from the percent of each peak of
the total peak OD for each separation. The formula for

ibiochemical distance is

n .. _ . 2
21:1 (A1] Aik)

68

where g_is the total number of different substances resolved.

Adi and A1; are the percentages of the total peak OD of the

bands resolved, d, in the lines 1 and k, respectively (15).
The matching coefficient was applied for only a few

particular comparisons. It was calculated according to the

.t _
h
kit?

a."

___)

formula:

m7mi :21

m + n
m + n + d '

where m_= number of positive matches or the presence of

‘E—umx-

homologous bands, 2 = number of negative matches or the ab—
sence of homologous bands in both taxa and d = number of
differences or a particular band occurring in only one of
the taxa.

The data illustrated for the amphiploids are an average

of the reciprocal crosses since they were very similar.

RESULTS

Quantitative Analysis

Table 3 illustrates the quantity of extractable phenols
from the plant material assayed in this study. There appears
to be a continuous range in quantity of phenols from a high
of 17.155 mg/g DW, 'Snowstorm', to a low of 3.424 mg/g DW,
'Glitters Imp.'. This suggests that phenol content may be a
quantitative character. All of the grandiflora lines (except
'Blue Mantel') contained a higher concn of phenolics (10.159
mg/g DW) than all of the multiflora lines (except 'Resisto
Scarlet') (6.4336 mg/g DW). 3. axillaris and the reciprocal
crosses involving P. axillaris and P, parodii1 (10.029 mg/g
DW) were quantitatively similar to the average concn of the
cultivars (10.159 mg/g DW). Results from this combination
suggest a single gene for quantity of phenols, P. axillaris
‘being dominant for approx double the phenol content. The
.2. axillaris and g. violacea crosses would indicate the
reverse situation, i.e., g. yiolacea would appear to be
dominant for approx one-half the phenol content. The recipro—
cal crosses involving‘35 axillaris and g. inflata, g, violacea

and g, inflata and P. violacea and g, parodii produced progeny

 

1Henceforth g. x P. will
(designate results of both reciprocal crosses.

69

 

Table 3. Total phenoI content from leaves of species,
amphiploids and Cultivars of Petunia (average of

3 extraction experiments).

 

 

 

 

 

 

 

Taxon M or GY mg phenols/
g dry wt
'Snowstorm' G 17.155 az
'Bridesmaid' G 11.478 b
P. axillaris 10.879 bc
3. axillaris x g. parodii 9.178 cd
'Resisto Scarlet' M 9.132 dd
'Harvest Moon' G 9.129 cd
'Kandy Kane' G 9.007 cd
'Rose Elegance' G 8.820 cd
'Blue Magic' G 7.905 de
'Orange Bells' M 7.824 de
'Resisto Red' M 7.703 de
'Blue Mantle' G 7.621 def
'Rose Joy' M 6.479 efg"
'Red Cap' M 5.868 efg
g. violacea x P. inflata 5.506 fgh
P. axillaris x g. inflata 5.083 gh
P. parodii 5.068 gh
P. violacea 4.810 gh
.2. violacea x P. parodii 4.746 gh
' ewpie' M 4.605 gh
P. inflata 4.544 gh
P. axidlaris x g. violacea 4.469 gh
'Glitters Imp.‘ M 3.424 h
Average of all grandiflora cultivars 10.029
Average of all multiflora cultivars 6.434
Average of all cultivars 8.296
.Average of all species and amphiploids 6.031

 

1M = multiflora or G = grandiflora phenotype.

*ZMeans followed by the same letter are not significantly
different at the 5% level, according to the Tukey test.

 

 

 

A .4).
“an...”

.. ‘II‘Itx'll- I‘ll!- 1.‘ I..- .‘L‘fie‘. 0. .lil- .
1

71

with a relatively low phenol content, similar to both parents.
These results require additional data from segregating gener-
ations to determine the quantitative inheritance of phenols

in Petunia.

TLC Analysis

The BEAW solvent system resolved a maximum of 10 bands
(Table 4). The number of bands found in any one taxonomical

unit ranged from 5 (g. violacea x g, inflata) to 8 ('Red Cap',

 

'Orange Bells', 'Rose Joy', 'Rose Elegance' and 'Blue Mantle’).
The modal number of bands was 7. The CAW solvent system
separated 7 bands. The number of bands present in any one
taxonomical unit was 5 or 6. The modal number of bands was

6. The total number of bands resolved in each system were
combined for statistical analysis, i.e., 17.

Of the 4 species, only 3. violacea and g. inflata
possessed identical patterns in each solvent system with a
correlation coefficient of .99, while 2. axillaris and P.
parodii resolved identical patterns in only the BEAW solvent
system. g. violacea x E. parodii and P. violacea x _13. inflata
resolved similar patterns in only the CAW solvent system.

The TLC patterns of the cultivars were identical in the
CAW’solvent, i.e., they all contained 6 bands and were miss—
ing only band 3. The BEAW solvent system resolved an inter—
esting banding pattern: the multifloras possessed band 8

and the grandifloras possessed band 7. They each lacked the

72

 

.omhuocmm muoamwpgmum H Um
.wmmuocmnm muoamwuede u 2%

.Omam mommouo Hmuoumeumn gnu monogmox

 

+-++-+-+4-+-++-+-++-+-++-+-+4—+-++-+-+

+-+4-+-++—+-++-+-++-+-++-+«r+
+

+-+ +-++-+-++-+-++-+-++-+-+
+-+4—+-++-+-++-+-+4-+-+41+

+-+-++-+

+-++-+-++-+-++—+-++-+-++-+-++—+-++—+-+
+-++-+

+-++-+-++—+-++-+-+4-+-++—+-++—+-++-+-+

+-++

+-++-+-e4—

+-++-+-++-+-++-+-++-+-++-+

+-++-+-++-+-+

+-+-++-+-++-+

+-+4«+-+4—+‘+-++-+-+4-+-++-+-++-+-+4-+

+-+4—+

+-++-+-++-+-++-+

+-++

+-++

+-++-+

+-++-+-++-+-++—+-++-+-+

+-+

+-++-+

+-++-+‘+-++-+-++-+-++-+-++-+-++—+-++-+

+-+4-+-++-+ +-++-+-++-+-+4-+-++—+-++-+

ml.
0'
0

6|

0|
0'
me I
2'
2'
SI.

SI.

.cooz_umm>umm.
.pamEnopHHm.
u .oammz_osam.
.oaugmz poem.
.Ehoum3ocm.
.wcmx hpcmx.
.mucmmoam mmom.
.hoo mmom.
.mHHom mmcmuo.
.oem3mx.
..neH muouueao.

z n .uoaumom oomamom.
z u .com oumammm.

.hZI
x moona0a>
mwmwwmw x mmmmwmww
wwmmmmm .M.x mammaawxm
mwmmwmwfl .M_x mammaaexm
xmmmawmw am x maumaaaxm
wwmmmmm

MNMMwaN

a

mammaaexm

depoumm .M
.m

 

 

 

 

.Qmo pom.

mnwndeMoquMnd

 

l‘

\O'
In

e m N H
3&0

 

OH

0 m

[x

35mm

 

swan m ugo>aom

Q'

01

N

H mvcmm

coxma

 

 

.mum>fluapo ppm mpfloameroEm .mmfiummm message no maogosm mo announce oqe

.¢ manna

73

other respective band except for a multiflora, 'Rose Joy'
and a grandiflora, 'Rose Elegance'. Each of these contained
both bands.

In no instance did any cultivar show a pattern similar
to a species or amphiploid after development in the BEAW
solvent. However, all of the cultivar patterns were identi-
cal to the g. axillaris x g. violacea interspecific cross
when the extracts were developed in the CAW solvent.

(g. axillaris was more highly correlated (.64) with the
cultivars than any of the other species (Table 5).

.g. violacea and P. inflata were very similar in their average
correlation to the multifloras and grandifloras, .41 and .40,
respectively. Although the correlation of multiflora with
grandiflora phenotypes was high (.79) it was less than
within multifloras (.94) and within grandifloras (.89).

Only 2 interspecific combinations had a correlation
coefficient greater than .50 with the cultivars, but the
correlation coefficient of the g. axillaris x g. violacea

amphiploid was slightly higher (.67) (Table 6).

 

 

y .‘llialuiibllllito 15‘- b’- (’iOLOI‘. ‘O.-al’ l
. .

 

74

113... . I!
h: {isle-2:117

Hm>oa an.o om namoamacmnm...
Ho>mH ea om osmoamacmam..

Hm>ma Rm um unmowMHcmems
.msoum poumcmemop may cenufi3 ugoeoemmooo coepmaonuou mmmuo>¢.

 

 

 

 

 

 

N

«can >
Ame Ave Anav Away Away Ame

«ssnmm. seemh. me. He. ow. sevm. muoamwpcmuw
Adv rode Amav Away Ame

«ssnvm. mm. me. ow. «sew. muoemeueaz

Ammv Agmv away .I

#«on. **NB. ma. vaoumm .m

Aeav Away .1

sssmm . .1 may. MOUMHOH> . m

\AANHV .l

we. mumamge .m

maumweaxm .M

muoamepcmuu .muoemauaoz flavoumm rm momma0w> am mumamce .M. meHmHHexm .M coxma

mum>euasu mmeummm

 

 

.nHm>wquo muoamepcmum pcm mHoHuwuHsnppcm newcomm
mo Amomogugoumm ngmmocmuuwp Hmoaemnuown pom mucowoflmmmou cowumamnuoo .m manna

75

Ni. In”:

.F

"119

Ho>oa ea om unmoamncmemss

 

 

 

«oax N

lmae Amae Amae .I I.
mm. Hm. mm. Hflpoumm .m x.mmomaoe> .m
3.: a: a: I
mm. Hm. mm. mumamgw x noan0e> .m
E at so I I
seem. ssmm. «sow. Hwfloumm .m x mflHMHHfiXM .m
roe roe Ame I, I
ssho. eshm. sshm. MOUMHOH> .m X mMHMHHflxm .m
Away Amae ulmae .IIIIIII .IIIIIIIII I
em. mm. mm. mumamcw .m x mesmeawxm .m
pmgwneou muonHpcmuw muoewwuesz coan

 

moame Hm>fluasu

 

 

.mum>wuanu muoemapgmum paw muoameuape nuw3 upeonwfimEm mo
Anemonugwumm Gav moucmwmflp.amueew300fln_pgm nucowowmwmou downwawuuou mmmuo>¢

.o OHQMB

 

|.~ .v.‘l'il€ ll...-
. .

Q.‘ I'.’|OD‘I 0| ‘5'.» I.

DISCUSSION

The total phenolic content of the lines investigated
as well as the correlation coefficients and biochemical
distances calculated from the chromatographic data suggest
that P. axillaris may be one of the putative parents of
g. hybrida. This has been suggested from cytological
studies by Steere (25) and presumed by others (3,8,11).

P. axillaris and the reciprocal crosses with g. parodii were

 

the only species and amphiploids which contained a similar
quantity of phenols to that of the cultivars. The recipro-
cal crosses of g. axillaris and P. violacea as well as

(g. axillaris and g, inflata produced progeny all relatively
low in total phenolics. Analysis of F2 populations of these
crosses would be necessary to determine the genetic control
of phenol content.

The factor(s) affecting quantitative phenol content may
be distinct from that (those) governing the qualitative
nature of endogenous phenols (Tables 3 and 4). When the
number of bands resolved in the BEAW and CAW solvent systems
are combined, 17, there was only a variation of 11 to 14
ibands resolved from all taxa. Six (BEAW: 1, 2, and 7; CAW:
4, 5, and 7) of the total 17 bands were found in every taxon

analyzed. These 6 accounted for a total of 53% of the

‘76

77

phenols as expressed in percent of total peak OD per sep—
aration.

While all correlations in this study are positive and
many are nonsignificant, it is important to view the entire
matrix since general trends of similarity are just as
important as any single camparison (24). .Mean correlation
coefficients and biochemical distances are charted for com—
parison of synthesized amphiploids to multifloras, grandi-
floras and the two combined classes (Table 6). Although
the correlation coefficient of P. axillaris x P. violacea
was the higheSt (.67), and significant at the 5% level, the
correlation coefficient of the cross involving P. axillaris
and P. parodii was also significant (.61). From these data
it appears that P. axillaris and P. parodii could be the

original parents of P1 hybrida. P. axillaris and the crosses

 

involving P. axillaris and P. parodii were the only lines

found to be similar on a gross (total phenol content) and

a more definitive quantitative basis (correlation of OD of
individual phenols) to the cultivars.

One morphological observation demands that this situa-
tion be further pursued. Both P. axillaris and P. parodii
produce only white flowers, while P. violacea, as well as
.P. inflata, produces deeply pigmented purple or magenta
flowers. Although a mutation in either of the white flowered
genomes is theoretically possible no such mutation has been

reported. As mentioned earlier, Mather and Edwardes (18)

78

concluded from a study of inheritance of flower pigments in
P, axillarig and P. violacea that the white flowered species
is recessive for lack Of pigmentation while the purple
flowered species is homozygous dominant for anthocyanin pro—
duction. Should the white species mutate to the color form-
ing allele it would be in the much less common direction of

recessive to dominant (17). This single observation requires

‘ ‘ ‘- , t-.(9'A‘.~ ‘31.”. W

further consideration Of the P. axillaris x P, violacea

 

 

amphiploid. A qualitative analysis Of the banding patterns

 

by the matching coefficient (Table 7) for these amphiploids
and the cultivars in this study shows that a greater simi-

larity exists between the P. axillaris x P. violacea amphi—

 

ploids and cultivars than between the P. axillaris x P.

 

parodii amphiploids and cultivars.

P. violacea and P, inflata are not significantly differ-
ent in phenol content and their correlation coefficient is
.99. This information appears to give some validity to the
Opinions Of Smith (22,23) and in part to Lamprecht (16) who
propose that they are a single species. If this is assumed
to be correct, it allows for the possibility that P. inflata
could be one of the putative parents Of P. hybrida as pro-
posed by Steere (27) from cytological studies. The evidence~
presented in this paper argues against this possibility since
P. inflata alone or combined with the other 3 species exhibited

low correlation coefficients with all Of the cultivars.

79

 

 

 

 

 

 

 

 

 

 

 

 

 

he--. .i LI.
Noam n
31 E E E E 13 18 8: 3e 5 SC at E E 31 mango? .MI
mm. mm. mm. mm. mm. mm. mm. an. mm. om. mm. mm. mm. mm. om. x meandaflxm .m
av 8v 5 2: E S E 8: 8: E 8: r: 8: ENE Seaman .mI
mm. as. He. an. on. mm. as. mm. mm. on. an. an. an. an. on. x manoeaexm .m
ON I I I I I I I I I I I I I I
.. emmmeaw unseen...” 2......
1.2 1 r. n n O u s s e M r. s s p. uemwuommuoucH
«+u A P a a M p. a a u .d 1. r. I.
r. a a s .A 5 r. 1 s s O
Inn s m N W 1 n. D a a a 3 4 e
2.: 1 e e O X TI O I J O O .o
1.4 e .n u m e e .A 8 s I
sun W I. I. n u .o I a S H
O P o I a e .L I o a
e O I I a I u T. m e P
T. u I D B d I I
I I a I . I
I I a
A
mHOHMflpgmuo muoamwuasz
num>wuaoo
.mum>euapu Ham
suw3 moOMHOM> .M x nflumaawxm .M.pcm proumm .M.x mwumaawxm .M_mpeOHmw£osm
mo Aoomonugoumm gee moocmumeo Hmuwsvnooen pom nugoeowmmmoo.mGenOumz. .h manna

S IMMARY

Phenolic compounds were investigated from leaf extracts

of 4 species Of Petunia (Splanacea), P. axillaris, P, inflata,

 

P. violacea and P. parodii, the reciprocals Of 5 interspecific

crosses, P, axillaris, x P, inflata, P, axillaris x P.

 

violacea, P. axillaris x P. parodii, P. violacea x P. inflata
and P. violacea x P. parodii, and-15 cultivars Of 2. hybrida
tO determine the taxonomical relationships of the species and
the ancestry of P. hybrida.

Quantitative analysis revealed a broad continuous range

Of phenolic concn, with most of the cultivars, P. axillaris

 

and the reciprocal crosses between P. axillaris and P. parodii

 

containing 2—3 times more phenols than the other species and
interspecific hybrids.

By the use Of one—dimensional thin-layer chromatography
and two different solvent systems 17 distinct bands were
separated. Photographic negatives of the developed chroma-
tographs under long-wave ultraviolet light were analyzed for
OD in a scanning densitometer. Each band's peak OD was
expressed as a percentage of the total OD of all the peaks'in
the spectrum. There were significant correlations between P,
inflata and P. violacea, P, inflata and P. parodii, P.

violacea and P, parodii and P, axillaris and all Of the

80

 

81

cultivars. Of the interspecific crosses only P. axillaris x

 

_P. parodii and P, axillaris x P. violacea were signifi-

cantly correlated with the cultivars. P. axillaris x P.

 

violacea was calculated to be biochemically closer to the
cultivars. Matching coefficients, based on the presence or
absence Of homologous bands, indicated a closer match between
P. axillaris x P. violacea and the cultivars than between

P. axillaris x P. parodii and the cultivars.

It was concluded that the 4 species investigated are

properly classified as distinct species and that P, axillaris

 

and P, violacea are most likely the original parents Of

P. hybrida.

wr—

 

10.

LITERATURE CITED

. Alston, R. E. and B. L. Turner. 1963. Biosystematics.

Prentice—Hall, Englewood Cliffs, N. J , p. 319.

, Mabry, T. J. and B. L. Turner. 1963.
Perspectives in chemotaxonomy. Science 142, 545—552.

 

Bailey, L. H. and E; Z. Bailey. 1941. Hortus second.
The Macmillan CO'., New York, p. 778.

Bate—Smith, E. C. 1964. LeucO-anthocyanins. I.
Detection and identification Of anthocyanidins formed
from leuCOranthocyanins in plant tissues. Biochem.”J.
58, 122—125. ' '

 

and N. H. Lerner. 1964. Leuco—anthocyanins. II.
Systematic distribution of leuco-anthocyanins in leaves.
Pdochem. J. 58, 1264I32.

 

Brehm, B. G. and R. E. Alston. 1964. A chemotaxonomic
.study Of Baptisia leucophaea var. laevicaulis
(Leguminosae). Amer. J. Bot. 51, 644-650.

 

Dedio, W,, P. J. Kaltsikes and E. N. Larter. 1969.
A thin-layer chromatographic study Of the phenolics Of
Triticale and its parental species. Can. J. Bot° 47,
1589—1593. ‘

 

Ferguson, M. C. and A. M. Ottley. 1932. Studies in
Petunia. III. A rediscription and additional discus—
sion of certain species Of Petunia. Amer. J. Bot.

19, 385—403.

Fries, R. E. 1911. Die Arten der Gattung Petunia.
k. Sv. Vet. Akad. Handl. 46, 25-40.

Garber, E. D. 1965. “The genus Collinsia. XXVII.
A paper chromatographic and disk electrOphoretdc study'
Of leaf extracts from 14 species and progeny from 5
interspecific hybrids. Can. J. Genet. CytOP. 7, 551—
558.

82

 

 

83

ll. Gleason, H. A. and A. Cronquist. 1963. Manual of vascu—
lar plants Of northeastern United States and adjacent
Canada. D. Van Nostrand CO., Inc., Princeton, N. J.,
p. 810.

 

12. Griffiths, L. A. 1952. Separation and identification
of aromatic acids in plant tissues by paper chromatog-
raphy. Nature 180, 286.

13. Harney, P. M. 1966. A chromatographic study Of species
presumed ancestral to Pelargonium x hortorum Bailey.
Can. J. Genet. Cytol. 8, 780—787.

 

 

 

l4. and W. F. Grant. 1964. A chromatographic study
of the phenolics of species of Lotus closely related
to P. corniculatus and their taxonomic significance.
Amer. J. Bot. 51, 621-627.

 

15. Jaworska, H. and N. Nybom. 1967. A thin—layer chroma-
tographic study Of Saxifraga caesia, g. aixoides, and
their putative hybrid. Hereditas 57, 159-177.

 

 

 

l6. Lamprecht, H. 1953. Petunia axillaris (Lam.) B. S. P.
und ihre synonyme P. violacea Lindl. und P. inflata R.
Fries. Aqri Hortique Genetica 11, 83-107.

 

 

 

17. Paris, C. D. 1956. Genetic analysis Of color inheritance
in Petunia. Doctoral thesis. Michigan State University.

18. Mather, K. and P. M. J. Edwardes. 1943. Specific dif-
ferences in Petunia. III. Flower colour and genetic
isolation. J. Genet. 45, 243—260°

 

19. Sheen, S. J. 1970. Polyphenol content, polyphenol-
Oxidase, and peroxidase activity in certain Nicotiana
species, varieties and interspecific hybrids. Theoret.
Appl. Genetics 40, 45—49.

20. Simon, J. P. 1967. Relationship in annual species of
Medicago KL. Analysis Of phenolics by means Of one
dimensional chromatographic techniques. Aust. J. Bot.
15, 83—93.

 

21. and D. W. Goodall. 1968. Relationship in annual
species of.MedicagO VI. Two-dimensional chromatography

of the phenolics and analysis of the results by prob-
abilistic similarily methods. Aust. J. Bot. 16, 89-100.

 

22. Smith, L. B. and R. J. Downs. 1966. Solanacea. In: .
Plora ilustrada Catarinense. P. R. Reitz, ed. Itajai,_
S. Catarina, Brazil, p. 321.

23.

24.

25.

26.

27.

28.

29.

84

Smith, L. B. 1968. Personal Communication.

Sokal, R. R. and P. H. A. Sneath. 1963. Principles of
numerical taxonomy. W. W. Freeman and CO.,
San Francisco, p. 359.

Stahl, E. 1969. Thin-layer chromatography. 2nd Ed.
Springer-Verlag, N. Y. Inc., p. 1041.

Steere, W. C. 1930. Petunia parodii, a new species Of
the subgenus Pseudonicotiana from Argentina. Papers
Mich. Acad. Sci. 13, 213-215.

. 1932. Chromosome behavior Of triploid
Petunia hybrids. Amer. J. Bot. 19, 340—356.

 

Swain, T. and W. E. Hillis. 1959. The phenolic constit-
uents Of Prunus domestica I. The quantitative analysis
Of phenolic constituents. J. Sci. Food Agric. 10,
63—680

Tornes, A. M. and D. A. Levin. 1964. A chromatographic
study Of cespitose zinnias. Amer. J. Bot. 51, 639-
643.

III» v II 11‘s...- .0... I, I‘IoI'IOI [I.II I

 

 

 

PROTEIN AND PEROXIDASE ELECTROPHORETIC ANALYSIS
OF SELECTED PETUNIA SPECIES AND CULTIVARS

11:10.. - .Q.‘ 3.4.-

 

IIIIInIh P.— l‘

a.
_

INTRODUCTION

The genus Petunia (Solanaceae, containing approx 30
species, is indigenous to Central and South America reaching
as far northsas the southern parts Of the United States.

The present study is concerned with the species: P, axil—
laris (Lam) B. S. P., P. violacea Lindl., P, inflata R.
Fries, and P. parodii W. C. S. The ancestry Of P, hybrida
Hort. is speculative, but several authorities (6,9) have
suggested that it is the result Of an interspecific cross
between P. axillaris and P. violacea. Steere (23), on the
basis Of cytological studies has suggested that P. inflata,

as well as P, axillaris and P. violacea, has contributed to

 

the development Of P. hybrida.

Presently, the classification of these 3 species is
controversial. Smith (21,22) has combined both P. violacea
and P. inflata under P, integrifolia (Hook.) Schinz and
Tellung var. integgifolia. He based his Opinion on the
magnitude of variation he Observed in their distinguishing
characteristics. Lamprecht (14) has combined these 2 species
under P. axillaris.'"Other investigators have reported on
allelic similarities found among members Of the genus, e.g.,

flower color (16) and the multiflora vs grandflora character

(1).

85

 

86

The amount Of morphological variability exhibited be-
tween species gives an estimate Of the extent Of their
genetic divergence. Numerous taxonomical methods have been
used to measure these differences. Most Of these differences
are controlled by simple genetic factors comprising a small
portion Of the entire genetic constitution (10). A method
which would assess and compare a random sample Of the
genome should provide useful taxonomical information.

Protein analysis by gel electrophoresis is such a technique
which measures genetic variation at the molecular level (I1).
Many chemotaxonomical studies have made use of the results
Of protein electrOphoresis (Datura, 3; Collinsia. 8;
Triticinae, 13; Nicotiana, 17, 19, 20).

The objective of this study was to determine if varia—
tion found in the biochemical properties Of general protein
and peroxidase from leaves Of flowering plants would sub-
stantiate the recognized classification of the species and

provide evidence on the ancestry of P, hybrida.

METHODS AND MATERIALS

The plants used for this study were grown from seed
Obtained from several sources (Table 1) and from synthe-
sized amphiploids (Table 2). Two general classes of culti—
vars, multiflora and grandiflora, have been analyzedfor
this study. These are allelic characters which affect
general growth characteristics and to some extent flower
size.1 All of the multiflora cultivars are homozygous
recessive, 33, and all of the grandiflora cultivars are
heterozygous, 99- The four species used in this study are
apparently homozygous recessive for the multiflora allele,
33 (K. C. Sink, unpublished data).

The seeds were germinated on vermiculite in a growth
chamber maintained at 22°:2C, with a 16 hr photOperiOd and
10760 lux. Young seedings were transplanted into a freshly
pasteurized soil mix (soil, peat moss and perlite, 1:1:1,
Viv/v) and grown in a greenhouse to flowering. Standard
cultural practices and disease and insect preventative pro—
cedures were followed.

Sample tissue, collected from flowering plants, con-
sisted of the two Opposite leaves subtending the three

uppermost Open flowers along all stems, and leaves above

 

1L. C. Ewart. 1963. The inheritance of flower size in
grandiflora and multiflora petunias. Ph. D. Thesis. The
Pennsylvania State University, p. 73.

87

 

 

88

.omhuoconm manmepcmnm u O HO muoemeuese n 2N

 

 

 

.ocH ..oo menumm nmmmoe z moo .nmo com.
.ocH mooom eyeshadow 0 mos .eflmEmooaum.
.OCH mpoom Apeempaow 2 Non .SOO omom.

uoouo pom madam .> .z 0 who .euoum3ocm.

uOOHo pom madam .> .z 2 are .uoaumom Ounflmom.

.oo cam mumxmm .e s «we ..dsH mumuueao.

.OO pom mumxmm .9 o are .oaucmz_opam.

.OO poom cmuanoeeIcmm 2 Hon .mdaom omcmno.

.OO poom coUHHOE¢Igmm 2 con .oemzox.

.oo poem samenessucmm o moo .oamms_osam.
"mum>wpaso .uuom MWNMMNH .M
mgopnmo Hmoecmuom Hohom S mom moflum .m MMMflme am
encampmee .mpoecouom mounom z oom .Hpgwq mmwMflmwM 5M
.oo comm cmuauosaIcmm a 5mm .m .O .s.wwmmmmm .M
mnopummsHMOflGMUOm Homom .2 com .m -m .m A.eoqv maumaaexm .M
oOHdOm no HO 2 .D 2 HonEflz goxoa

godmoouoe

 

 

.mum>euedo paw moeoomm mwcsuom MO oousom .H define

89

Table 2. Interspecific hybrids of Petunia species anaPyzed
for protein and peroxidase banding patterns.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Pedigree
Interspecific hybrids Number MSU
P. inflata x P, axillaris 195
P. axillaris x P, inflata 199
P. parodii x P, axillaris 204
P. axillaris x P, parodii 200
P. violacea x P, axillaris 208
P. axillaris x P, violacea 201
P. inflata x P. violacea 196
P. violacea x P, inflata 207
P. parodii x P, violacea 205
P. violacea x P, parodii 209

 

 

2The P. parodii x P, inflata cross could not be synthesized.

90

these flowers at the base of the buds. The leaves were im-
mediately immersed into liquid N; and extracted within 48 hr.
Protein acetone powders were made according to El—Basyouni
and Neish (5). The powders remained stable up to 3 months at
-20°C. Protein extracts were prepared by dissolving 50 mg of
the acetone powder in 5.0 ml of 0.2 M tris-citrate buffer,
pH 8.3, with 0.1% L-cysteine, 10.0%,sucrose and 1.5 g poly—
vinylpyrrolidone (Polyclar AT), which had been cleared Of
fines and hydrated in tris-citrate buffer. The extraction
was carried out at 2-4°C for 12 hr with occasional stirring.
The slurry was centrifuged at 30,900 xg for 30 min and the
supernatant solution stored under N2 until used. Protein
content was determined from ethanolic precipitates (15).
Polyacrylamide gels, 4.5 x 0.5 cm, were prepared according
to Davis (4) but the running buffer was diluted to only &
strength. The loaded protein samples consisted of 200 ug of
protein, per gel. Electrophoresis was carried out at 2—4OC
with 2 m amps per tube.

The gels were stained for total protein or peroxidase
enzymes. Total protein was stained with a 0.05% solution
Of Coomassie blue in 12.5%ftrichloroacetic acid (2).
Peroxidases were stained in an ethanolic solution of
o-dianisidine (3,3'Hdimethoxybenzidine) as the hydrogen
donor (20). This stain was used since the color product is
stable in an aqueous solution for several weeks. The stained

gels were photographed and scanned for OD in a Gilford

.I‘I.I’-..')Olavll. Cl III) 1

 

 

 

91

scanning densitometer 12 to 18 hr after staining. The
peroxidases were scanned at 410 nm and the proteins at 569
nm.

Although equivalent weights of protein were placed on
the gels, band OD's were transformed into percentages of the
total OD of all the bands in a separation (13). This method
further equalized the areas under the curves and permitted
direct comparisons between taxa for OD of specific protein'

fractions. Homologous proteins were determined by equivalent

 

migration distance of band X 100),

hRf s, (migration distance Of front

i 0.01 mm, while

ts’

homologous peroxidases were determined by equivalent hRfS

(

In the latter case, the bromphenol blue tracer band completely

migration distance Of band X 100)
migration distance Of selected standard band '

disappeared in the peroxidase stain solution. For this
reason, a peroxidase-positive band which was present in all
Of the taxa was used as a standard.

The correlation coefficients between the taxa were calcu-

lated from the average Of 2—4 gels.

.- ‘ 8111-1“

 

“If“! 0’
I‘yJ

:-

V .
M..-

“' -. v 1,"

I I!!! I'.0.I 'IIIOI'."III§.

 

RESULTS

P. axillaris and P, inflata had 7 identical peroxidase
bands (Table 3). P. parodii also had 7 bands but it dif-
fered from P. axillaris and P. inflata exhibiting bands 2

and 6, but lacking bands 1 and 3. P, violacea had 10 bands,

 

of which 6 were identical to P, axillaris anle. inflata:

 

bands 3, 4, 7, 8, 9, and 10. P. violacea also exhibited all
of the bands in P. parodii: bands 2, 4, 6, 7, 8, 9, and ID.
In addition P. violacea possessed two bands at 5 and 10.

In all species 11 bands Of peroxidase activity were found
(Table 3 and Figures 16-19, on page 100).

The banding pattern Of P. axillaris, when compared to

that of the species, was most highly correlated with P.

inflata (r = .997, Table 5). It was also significantly corre—
lated with P. parodii (r = .54) but not with P. violacea

 

(r = .37). Similarly, P. inflata was correlated with P.
parodii (r = .53) and not with P. violacea (r = .36).

P. parodii was significantly correlated with P. violacea

 

(r = .65).

The peroxidase banding patterns among the cultivars were
generally similar but not identical to any Of the species
(Table 3, Figures 20-30). Five cultivars possessed identical
patterns, 'Kewpie' and 'Red Cap', multifloras: and 'Snowstorm',

'Bridesmaid', and 'Rose Elegance', grandifloras. They all

92

Moth—i- («Edam
\ .-‘_5_ I.

W’

—‘——-

4

D 'a."

‘1."

 

 

93

 

.omhuogosm muoamepgmum ON

.ommuogosm muoamwuaofi n 2%

 

+ -+ + -+ +

. + .

+ -I + -+ +
+

+ -I + -+ + -+ + -I + .I + -+ + -I +
+ -+ + -+ + -+ + -+ + -I +
+ -I + -I + -I + -+ + -I + -+ + -+ +
+ -I + -+ + -I + -I
+ -I + -I + -I + -I +

I -+ + -I + -+ + -+ +
+ -+ + -+ + -I + -+ +

OI .oocmmoam owom.
OI .peMEMOOeHm.

OI .Euoumzonm.

OI .oaucmz.osam.
.mOI .oammz.osam.
:T .meaom omgmuo.

an .98 com.

2T .MOO omom.

SI .uoaumum oumwmom.
SI .dsH magpie.
sz. .3858?
wwmmmmm .M

moumaow> .M

maul... .m

maumaaaxm .m

 

 

 

ma NH HH-OH.m m N. O m

<‘
M
N
,4

umpcmm coxma

 

.mHm>HUdSU paw mowoomu owcsuom we

announce onepgmnIOmmponuom .m manna

A‘i

 

 

I
.. _“—_—— .‘.1_..... .
. , r..._—o..—.o-.--oa-4-

Figures 1—15.

94

Protein electrOphoretic patterns illustrating
the range Of variability Observed among 4
species and 11 cultivars of Petunia.

Figure
Figure
Figure
Figure

Figures

Figure
Figure
Figure
Figure
Figure

ewww
m. I O O

5
6
7
8.

9.

-9

inflata.

. violacea.

parodii.

Grandiflora cultivars of
P. hybrida.

'Blue Magic' .

'Blue.Mantle'.

'Snowstorm'.

'Bridesmaid'.

'Rose Elegance'.

'leUMHW

Figures 10—15. Multiflora cultivars Of

Figure
Figure
Figure
Figure
Figure
Figure

10.
11.
12.
13.
14.
15.

P. hybrida.
'KewPie'.
'Glitters Imp.'.
'Resisto Scarlet'.
'Rose Joy'.

'Red Cap'.
'Orange Bells'.

   

1'2 34'
5 6‘7‘8 9
10 n 12 Is 14 15

Figures 1—15

 

 

 

“Hubby“ .

em
.

.‘ In III-.1... 11' 1".I' poi-ISII'III

96

exhibited bands 2, 3, 4, 7, 8, 9 and 10. Three cultivars
possessed 8 bands: 'Glitter Imp.'. bands 2, 4, 7, 8, 9, ll,
12 and 13; 'Rose Joy', bands 1, 2, 3, 4, 7, 8, 9 and 13;
'Blue Magic', bands 2, 3, 4, 7, 8, 9, 10 and 13. The remain-
ing 3 cultivars had 6 bands: 'Resisto Scarlet', bands 2, 4,
7, 8, 9 and 13; 'Orange Bells', bands 2, 3, 4, 6, 8 and I3;
'Blue Mantle', bands 3, 4, 6, 7, 9, and 13. There was no
distinguishing pattern between the multiflora, dd, and
grandiflora, gg, genotypes.

The average correlation coefficient for multiflora com-
pared to multiflora was .79, grandiflora compared to grandi-
flora .77 and multiflora compared to grandiflora .77. These
coefficients were all significant at the T% level.

Only bands 4 and 13 were common to all species and culti—
vars. 'Glitters Imp.’ possessed two bands, 11 and 12, which
were not found in any Of the species or the other cultivars.
All other bands found in the cultivars were also in one or
more of the species.

All species were significantly correlated with most
cultivars. P, axillaris and P, inflata had the highest
average correlation coefficients with all cultivars, .74
and .73, respectively. The average correlation Of P. parodii
‘with the cultivars was .68; P, violacea was lowest (.52).

A total of 48 protein bands were resolved from the species
and Cultivars (Table 4 and Figures 1-15, page 95). None Of
the species had identical banding patterns. P, axillaris and

.P. inflata had 25 bands. P, violacea possessed 26 bands and

‘Tfltb‘ ‘7“; Jun-n: ~. '8
| .

 

. mo....—.._._._.--.... .. .._l

, 5...... .-.—..-—v4_-—-o-.-.

Protein banding patterns Of Petunia species and cultivars.

Table 4.

 

 

97

,aoue6913 asou,
IPTENSGPIIHI
,mioasmous,
Ialluew anI8.
.Drfiew ante.
.sIIag abueio,
Idea pea.

,KOf esouI
IQSIIBDS 0191998.
I'dmr sxaanrto.
,srdmax,
TIPEEEH '3

eeoetorA 'd

 

EQPIJUI '3
sriettrxe '3
'U

c

m

m

+ +-+ +-I+
+-+-+ + +-+
++-+ + +-+
+-+ + +-+
+-++-+-+ +~+
+-+ + +-+
+4-+ +-++
+-++ + +
+-+ + +-+
+-+ + +-+
+-+ + +-+

HNMQ‘LDKOFQO‘

Ordm
HF4H

M
H

fi'
H

m
H

\O
H

l‘
,4

(no
r-lI-l

 

98

 

+++

++

++++

++

+

++++

++

+++

++++++++

+++

+++

+++

++

+++++ +

++

+++

+++

me
:V
we
m¢
in
me
was
HO.
0e.
mm
mm
mm
mm
mm
¢m
mm
mm
an
on
mm
mm
mm
mm
mm
VN
mm
am

am
om

 

 

-
A
I

I; egg-1 I

 

. _._._,-_..-.‘HA.-_-.- --._ “h..._‘._.—.__4.~ . ~

Figures 16-30.

99

Peroxidase electrophoretic patterns illus-
trating the range of variability Observed
among 4 species and 11 cultivars Of

 

Petunia.

Figure 16. P. axillaris.

Figure 17. P. inflata.

Figure 18. P. violacea.

Figure 19. P. parodii.

Figures 20-24. Grandiflora cultivars of
P. hybrids.

Figure 20. 'Blue Magic'.

Figure 21. 'Blue.Mantle'.

Figure 22. 'Snowstorm'.

Figure 23. 'Bridesmaid'.

Figure 24. 'Rose Elegance'.

Figures 25-30. Multiflora cultivars of
P. hybrida.

Figure 25. 'Kewpie'.

Figure 26. 'Glitters Imp.'.

Figure 27. 'Resisto Scarlet'.

Figure 28. 'Rose Joy'.

Figure 29. 'Red Cap'.

Figure 30. 'Orange Bells'.

100

I

II»- In

-16.- 17.113 19

HHHF'."

_292‘ 22 23 24__-.,

iii“

H-firwnqfl

25 26 27 28 29 30

Figures 16—30

 

101

P. parodii had 24 bands. The four species had 5 common
bands (2, 3, 8, 10 and 12). The following are the number
Of bands each pair of species had in common: P. axillaris

and P. inflata, 14; P. axillaris and P, violacea, 14; P.

 

axillaris and P, parodii, 11; P. inflata and P, violacea,
18; P. inflata and P, parodii, 18; P. violacea and P.
parodii, 16.

P. axillaris was not correlated with any Of the other
species on percent total OD of each band (Table 5). However,
the remaining 3 species were significantly correlated with

each other: P. inflata and P, violacea, .36: P. inflata

 

and P. parodii, .57; P, violacea and P. parodii, .29.

There were no identical protein banding patterns among
the cultivars. The average number of bands resolved from
the multiflora and grandiflora cultivars was 27. The number
of bands Observed ranged from 23 in 'Rose Elegance', to‘3l
in 'Blue.Mantle'. Only bands 2, 8 and 12 were found in all
Of the cultivars. There was no band or series of bands which
distinguished multiflora from grandiflora cultivars. The
average correlation coefficient for multiflora compared to
multiflora was .37, grandiflora to grandiflora, .18 and
multiflora to grandiflora, .19, Only the multiflora to
multiflora comparison was significant at the 5%»level.

Protein bands 2, 8 and 12 were common to all species
and cultivars. Bands 1, 18, 25, 38 and 48 were only found
in the cultivars, whereas there were no bands unique for

the species.

 

102

P. axillaris was significantly correlated with only

 

2 cultivars, 'Orange Bells' (r = .37), and 'Snowstorm' (r =

.38). P. violacea was significantly correlated with 4

 

cultivars, 'Glitters Imp.‘ (r = .32), ‘Orange Bells' (r =
.32), 'Snowstorm' (r = .44) and 'Bridesmaid' (r = .49).

.P. parodii was significantly correlated with 2 cultivars,'Blue
Magic' (r = .49) and 'Bridemaid' (r = .48). None Of these
three species, P. axillaris, P. violacea and P, parodii were
significantly correlated, on the average with all Of the

cultivars. P. inflata was significantly correlated with 6

cultivars, 'Kewpie', r = .43, 'Resisto Scarlet', r = .36,
'Orange Bells', r = .30, 'Blue.Magic', r = .42, 'Snowstorm‘,
r = .40 and 'Bridesmaid', r = .51. The average correlation

Of P. inflata with all Of the cultivars was also significant
(r = .32). Four cultivars were not correlated with any of
the species, 'Rose Joy', 'Red Cap', 'Blue Mantle' and 'Rose
Elegance'.

' Prqtein and peroxidase patterns were Observed for the
interspecific crosses Of the four species in this study.
These data cannot be statistically analyzed with that Of the
species and cultivars presented since they were grown during
a different season.

The total.number of peroxidase bands exhibited by the“
interspecific hybrids was 11. Eight bands appear to be
homologous, based on similar hRfst's, to 8 bands exhibited
by the species and cultivars (Table 6). The 3 remaining

bands, designated A, B and C, could not be identified as

. Fin

'3“...

'
13:3"

”'3’."

 

103

., III.- \l.)

’43.. .
E. IV.¢ .\ «iriluugihi

. .Ho>oa ma.o one um osmoamacmamsa. “Ho>oa as one om ucmoemacmamaa
“Ho>oa am one no uCOOMMHomflma “ommuoconm mHOHMHpgmum n O» nommuogonm muoamauase H 2%

 

 

 

 

 

 

mum>euaso

OH. mm. mm. mm. mo. mm. mm. «a. saw mom came moaoodm
a aa a aa aa

am. am. mm. om. mo. mm. me. me. O Ham mom came monomnm
a aa a aa aa

mo. mm. mm. am. me. me. me. «5.. z Ham mom came monommm
a aa aa aa

we. mo. 0H. mm. mm. mm. ms. om. O I .oocmmmam omom.
aa a aaa aaa

me. me. am. NH. me. am. as. om. O I .Oamsmmeanm.
aaa aaa aaa aa a aaa aaa

NH. we. oe. mm. as. mm. am. am. O I .euoomzocm.
aa aa aa aa a aaa aaa

mo.I mo. me. me. «me. me. Mm. mm. O I .oancmz_osam.

me. as. me. eo. cm. as. no. mo. NO I .oammz_osam.
aaa aa a aa aa aa

Ho.I mm. on. am. On. me. no. OO. 2 I .maaom omcmuo.
a a aa a aa aa

so. 50. ON. ON. ms. mm. am. mm. s I .mmo Omm.
aa aa aaa aaa

ea. mo. mm. OH. NO. mm. om. om. . s I .moe «mom.
. a aaa aaa

mo. am. On. am. am. on. N». as. s I .uoauwom oumamom.
a aaa a. aa aa

ea. mm. am. mm. mm. ca. .sa. me. a I ..neH muouonHO.

ea. an. me. ma. ms. -mm. as. mm. as I .oamsmx.

aa aa .a aaa aaa I

am. am. mo.I mm. mm. «m. Ame aaeoomm .m

a aaa. aa a a. II

On. me. on. an. Ase moomaoa> .m

a IllllllI II

he. ham. nae someone .m

fit. I

Ame maumaaaxm .m

Ame 1>v Ase Ame Ame A>e Ase Ame coxme

maeououm HOHOCOO oompexouom

 

 

.mum>eueso.pcm mowoomm
mecsuom mo monouumm ocepgmn mo OOmflHmmEOO osu EOHm mucoeoemwooo coeumaouuoo .m manna

104

Table 6. Peroxidase banding patterns of Petunia inter-
specifiC‘hYbrids.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Taxa Bands:2 2 4 A B 6 7 8 C 9 ll 13
P. inflata x axillarig + + + + +
P. axillaris x inflata + + + + +
P. parodii x axillaris + + + +
P. axillaris x parodii + + + +
P. axillaris x violacea + + + + + + + +
P. violacea x axillaris + + + + + + +
P. inflata x violacea + + + + + +
P. violacea x inflata + + + + + + +
P. parodii x violacea + + + + + +
P. violacea xparodii + + + + + +

 

 

zBands homologous to those in Table 13, based on similar
hRfst's have retained the same number designations. New
band sites have been inserted relative to the established
bands and lettered consecutively.

 

I ICIV I!" '1‘. ..

.II. ..9‘ .‘lellIlIn'I’IIdITI .

105

having homologous bands because of their distinct hRfSt
values. Bands 2 and 13 were common to all Of the hybrids.
Only the reciprocal crosses involving P. axillaris and

P. parodii exhibited identical banding patterns.

Only the hybrids in which P. axillaris was one of the
parents exhibited band A. None of the remaining bands could
be identified with a particular species. Most of the
hybrids' banding patterns were not correlated with either-
of the parents' banding patterns (Table 7). P. inflata x
violacea and the reciprocal cross were significantly corre-
lated with P. inflata. The P. violacea x axillaris cross

was correlated with P, dxillaris.

106

.HOPOH RH mflu #m HGMUHMHGmHmaa
.HO>OH Rm onu um #:moemecmema

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

mm. ¢O.I flepoumm x mflnmeeexm (M
mm. - go.I mfluoaaflxm x proumm .M.
ma.I Ho. Hepoumm x moomaoe> .M
mm. 00. moomeoe> x wepoumm .M

mm. amp. mammaafixm x moomaoe> .M

No. mo.I moomaoe> x mwumeaexm .m

om.I «mm. mumamce x.moumaoe> .M

mm. aagm. moomaow> x.mumamgA .M

mm. 00. panama“ x maumaaexm .M

mm. o~.I mwumaaaxm x.momamca .M

wwmmmmm .M. moomaoa> .M. MMMflMmN am mammaeexm .M _nommouo owmwoommuougH

mowoomm

 

 

.oompwxouom Mom nonwouo Oemwoomm
Immune pom mowoomn nwcsuom mo ocuouumm mcepgmn mo mugoeowmwooo goeumeouuou .h manna

DISCUSSION

Similar electrophoretogram patterns of enzyme or protein
extracts have been used to provide significant clues to
phylogenetic relations among species Of the same genus (3,13,
17,18,19). Johnson and Hall (I3) have shown that protein
spectra can discriminate various degrees Of affinity. The
peroxidase banding patterns show that P. axillaris and P.
inflata possess identical peroxidase enzymes (Table 3).

Their correlation coefficient of .997 indicates that the
peroxidases are present in the same relative amounts (Table 5).
This finding supports Lamprecht's (14) classification of

P. inflata as a synonym of P; axillaris. However, his identi-
fication Of P. violacea as a synonym of P, axillaris is not
supported by these results. The peroxidase banding pattern

Of P. violacea was not significantly correlated with that of

 

.P. axillaris or P, inflata but its protein banding pattern

 

was correlated with P. inflata at the 5% level (r = .36).

The Opinion Of Smith (21,22) that P, violacea and P,
inflata are a single species,‘PT integrifolia, is not sup—
ported. The morphological similarities exhibited by
P. violacea and P, inflata, in light of these results, may be
interpreted as a consequence of convergence, since they

are found in similar locations (21).

107

108

The peroxidase banding pattern of P. parodii was sig—
nificantly correlated to the other species and with the
protein banding patterns Of PI violacea and P, inflata.
There are two possible explanations for these results.

P. parodii may be a hybrid of two or all three Of the
species. It is the most recently described species Of the
four investigated (23). The combination of morphological and
biochemical results suggest that the 4 species are extremely
close taxonomically and genetically but are most probably
separate species.

It is difficult to determine the taxonomical relation-
ship Of P. inflata and P. parodii. They have a correlation
coefficient Of .57, significant at the 0.T% level, based
on their protein banding patterns but they have proven to
be cross incompatible under greenhouse conditions. It is
possible that the cross incompatibility exhibited is con—
trolled by a small portion Of their genomes and that they
are, in fact, closely related. This interpretation is un-
acceptable on a morphological basis. Their leaf and flower
characteristics are very different. Of the 4 species con-
sidered, P. inflata produces the smallest leaves and the
shortest flowers which are'all purple. P, parodii produces
the largest leaves and the longest flowers which have no
pigmentation.

The taxonomical relationship between the cultivars and

the species based on the peroxidase banding patterns is

109

ambiguous. From these results, it can be assumed that all

4 species may have participated in the develOpment of

.P. hybrida. The close correlations among the species make
this conclusion the only one possible. While the correlations
based on general protein patterns Of the species and the cul-
tivars are even less definitive, the data warrant the con-
clusion that P. inflata has provided a significant portion Of
the heritable protein complement resolved in this study.

P. violacea has provided the least influence.

The variability exhibited among the cultivars indicates
that breeding programs may have unintentionally placed some
pressure upon the peroxidase enzymes, as well as on other
proteins and enzymes.

The peroxidase banding patterns resolved from the
interspecific crosses, like those of the cultivars, were
variable (Table 6). Except for the reciprocal crosses involv-
ing P, dxillaris and P. parodii all Of the patterns were dif—
ferent. Without further breeding data, this may be attributed
to the heterozygosity of the original parents.

In a study of this nature, genetic variation within the
species is preferable to artifically inbred species which
would not be truly representative Of the species. The varia—
tion in peroxidase banding patterns found among the cultivars
does not gO beyond that exhibited by the species (Table 3 and
Figures 16-30). Only 1 cultivar possessed 2 bands not posses-

sed by any Of the species. One Of these bands, 11, was also

110

identified in 3 Of the interspecific hybrids (Table 6).
Further genetic studies are necessary to identify the nature
of what may be a hybrid peroxidase enzyme.

The results of this investigation show that the 4 species
investigated are properly classified as 4 distinct species
in the genus Petunia. Although they are very closely related
genetically, their correlation coefficients based on percent
peak OD of proteins and peroxidases indicate that all may
have been involved in the development Of P. hybrida.

A more extensive survey Of the 4 species investigated
from areas to which they are indigenous is still necessary
to identify the existence Of any invariant proteins or

enzymes which would more accurately identify the species.

SUMMARY

A study Of the general protein and peroxidase banding
patterns Obtained by disc electrophoresis Of leaf extracts
from flowering plants Of P, axillaris, P, inflata, P.
violacea, P, parodii and 11 cultivars of P, hybrida was
made to determine taxonomical relationships among the
species and the ancestry Of P,"hybrida.

A total Of 11 peroxidase isoenzymes were resolved from
the species and 2 additional isoenzymes from the cultivars.

P. axillaris and P. inflata had identical banding patterns

 

with 7 sites Of enzyme activity resolved. P. violacea had

10 bands and P. parodii had 7 bands. The peroxidase banding

patterns among the cultivars were generally similar but not

identical to any of the species'. Five of the cultivars

possessed identical banding patterns. Only 2 bands, 4 and

13, were found to be common to all the species and cultivars.
P, axillaris and P, inflata were highly correlated.

The average correlations of the individual species with all

Of the cultivars were significant at the 5% level:

P. axillaris, .74; P, inflata, .73; P, violacea, .52; and

P. parodii, .68. All of the grandiflora and multiflora

cultivars were highly correlated.

111

112

A total Of 48 general protein bands were resolved from
the species and cultivars. NO 2 species and/or cultivars
possessed identical patterns. Only 3 general protein bands,
2, 8 and 12, were common to all species and cultivars.

P. axillaris was not significantly correlated with P, inflata,
P. violacea or P, parodii although they were significantly
correlated with each other. Only P. inflata was significantly
correlated, on the average) with all Of the cultivars. On

the average, comparisons Of multiflora to multiflora, multi—
flora to grandiflora and grandiflora to grandiflora were not
correlated.

The 4 species investigated have been determined to be
closely related phylogenetically although prOperly classified
as distinct species. The results also indicate that P.

inflata has been involved in the synthesis Of P, hybrida but

that P. axillaris and, to a lesser degree, P. violacea and

 

P. parodii may have also contributed to its develOpment.

10.

LI TERATURE CITED

Chlebowski, B. E. 1967. Genetic studies on Petunia.
III. Inheritance Of the grandiflora character in a
cross between P. hybrida grandiflora and P. axillaris.
Genetica Polonica 8, 57—73.

 

Chramback, A., R. A. Reisfeld, M. Wyckoff and J. Zoccari.
1967. A procedure for rapid and sensitive staining
Of protein fractionated by polyacrylamide gel electro—
phoresis. Anal. Biochem. 20, 150—154.

 

Conklin, M. E. and H. H. Smith. 1971. Peroxidase iso—
zymes: A measure Of molecular variation in ten
herbaceous species Of Datura. Amer. J. Bot. 58,
688—696.

Davis, B. J. 1964. Disc electrophoresis. I. Method
and application to human serum proteins. Annals N. Y.
Acad. Sci. 121, 404—427.

El-Basyouni, S. Z. and A. C. Neish. 1966. Occurrence
of metabolically-active bound forms Of cinnamic acid
and its phenolic derivatives in acetone powders of
wheat and barley plants. 'Phytochemistry 5, 683-691.

Ferguson, M. C. and A. M. Ottley. 1932. Studies in
Petunia. III. A redescription and additional discus-
sion Of certain species of Petunia. Amer. J. Bot. 19,
385-403.

IFries, R. E. 1911. Die Arten der Gattung Petunia.

K. Sv. Vet. Akad. Handl. 46, 25-40.

 

Garber, E. D. 1965. The genus Collinsia. XXVIII.
A paper chromatographic and disk electrOphoretic study
Of leaf extracts from 14 species and progeny from 5
interspecific hYbrids. Can..J. Genet Cytol. 7, 551—558.

,Gleason, H. A. and A. Cronquist; 1963. .Manual Of

vascular plants Of northeastern United States and
adjacent Canada. D. vanNOstrand CO., Inc., Princeton,
N. J., p. 810.

Gottlieb, L. D. 1971. Gel electrOphoresis: new approach
to the study of evolution. Bio. Science 21, 939-944.

113

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

21.

22.
23.

114

Hubby, J. L. and R. C. Lewontin. 1966. A molecular
approach to the study Of genic heterozygosity in
natural populations. I. The number Of alleles at
different loci in DrOSOphila pseudoobscura.
Genetics 54, 577-594.

Jaworska, H. and N. Nybom. 1967. A thin-layer chroma-
tographic study of Saxifrage caesia, g. aizoides, and
their putative hybrid. Hereditas 57, 159-177.

 

 

Johnson, B. L. and G. Hall. 1965. Analysis of phylo-
genetic affinities in the Triticinae by protein
electrophoresis. Am. J. Bot. 52, 506-513.

Lamprecht, H. 1953. Petunia axillaris (Lam.) B. S. P.
und ihre synonyme P. violacea Lindl. und P. inflata R.
Fries. Agri. Hort. Genet. 11, 83—108.

Lowry, O. H., N. J. Rosebrough, A. L. Farr and R. J.
Randall. 1951. Protein measurement with the folin
phenol reagent. Jour. Biol. Chem. 193, 265-275.

Mather, K. and P. M. J. Edwardes. 1943. Specific dif-
ferences in Petunia., III. Flower colour and genetic
isolation. J. Genet. 45, 243-260.

 

Reddy, M. M. and E. D. Garber. 1971. Genetic studies
Of variant enzymes; III. Comparative electrOphoretic
studies Of esterases and peroxidases for species,
hybrids, and amphiploids in the genus Nicotiana.
Bot. Gaz. 132, 158-166.

Scandalios, J. G. 1964. Tissue—specific isozyme varia—-

tions in maize. J. Hered. 55, 281—285.

Sheen, S. J. 1970. Polyphenol content, polyphenol-
oxidase and peroxidase activity in certain Nicotiana
species, varieties and interspecific hybrids.
Theoret. App. Genet. 40, 45-49.

Smith, H. H., D. E. Hamill, E, A. Weaver and K. H.
Thompson. 1970. .Multiple molecular forms of peroxi—
dases and esterases among”Nicotiana species and
amphiploids. J. Hered. 61, 205-212.

Smith, L. B. and R. J. Downs. 1966. Solanacea. £2:
Flora.ilustrada Catarinense. P. R. Reitz, ed.
Itajai, S. Catarina, Brazil, p. 321.

. 1972. Personal Communication.

 

Steere, W. C. 1930. Petunia parodii, a new species of
the subgenus PseudonicOtiana from Argentina.
Papers Mich. Acad. Sci. 13, 213-215.

 

 

 

4

34

IIIIII