The influence of ammonia-induced cellulardiscoloration within cotton leaves (Gossypium hirsutum L.) on light reflectance, transmittance, and absorptance

The Influence of Ammonia-Induced Cellular Discoloration within Cotton Leaves (Gossypium hirsutum L.) on Light Reflectance, Transmittance, and Absorptance' R. CARDENAS, H. W. G A U S M A N , W. A. ALLEN, and M A R C I A SCHUPP U.S. Department of Agriculture. Weslaco. Texas

Abstract Cellular discoloration within cotton leaves, caused by anamonia treatment, reduced spectrophotometrically measured near-infrared light reflectance over the wavelength interval 750-900 mlz, where rounding of the plateau occurred. Results with ammonia-treated leaves are compared with other studies in which reduced reflectance occurred over the 750-900 m.u wavelength interval. It is possible that this rounding of the plateau caused by internal cellular discoloration may become useful in remote sensing for identifying certain nonvisual symptoms of plant leaf stresses caused by cellular discoloration. Introduction Spectrophotometric light reflectance, transmittance, and absorptance of individual plant leaves has been intensively studied in the wavelength interval 400-2500 m#. From 400-750 m~, plant leaf reflectance is relatively low, with a peak of approximately 10°, at 550 mF in the green region. The reflectance of a leaf increases to about 5000 in the near-infrared region and is relatively constant over the wavelength interval 7501350 m/~, a spectral interval for plant materials commonly called the plateau region. A transmittance spectrum has the same shape and approximately the same magnitude as a reflectance spectrum. Absorptance is high in the visible region because of leaf pigments, and in the infrared beyond 1350 mF because of water, but absorptance is at the most I or 2",, in the 750-1350 m/z plateau region (Gausman, Allen, and Cardenas, 1969). Colwell (1956), while working primarily with cereal rusts, suggested the use of infrared film for recording any disease that interfered with the internal reflection of light within leaves. Keegan et al. (I 956), with Colwell's assistance in collecting leaf specimens, did extensive research on effects of stem rust (Ptwchria graminis tritici) and leaf rust (Puccinia tritichta or Pucchtia rubigo-vera tritic) of wheat on light reflectance. Keegan's data showed that severe compared with low rust infestation caused a rounding of the shoulder of the plateau or a decrease in reflectance from 1000 to 750 rap.. The same response in reflectance was noted by Gausman and Cardenas (1968) after hair removal on upper leaf surfaces of the velvet plant (Gynttra attrontiaca). Allen and Richardson (unpublished data) noted that removal of leaf hairs had little effect on the refractive index of the velvet plant's upper leaf surfaces, but leaf absorptance was greatly increased within the wavelength interval 750-1000 mF- It was theorized that oxidation of polyphenols caused a brownish discoloration of the exudate from " s t u m p s " after hair removal, thus increasing leaf opaqueness with a subsequent increase in absorptance and decrease in reflectance ( G a u s m a n and Cardenas, 1969). The research summarized here considers: the influence of cellular discoloration within leaves on near-infrared light reflectance; and the tenet that cellular discoloration within leaves may be useful in detecting nonvisual symptoms of plant

diseases. However, previsual detection of plant diseases has had variable success. Manzer and Cooper (1967) found that late blight of potatoes could be detected by aerial photography 1 to 3 days before visual symptoms became apparent. However, tobacco ringspot virus could be detected only a b o u t 1 day before visual symptoms were evident (Burns et al., 1969). In contrast, Heller (1968) found that beetle damage could not be predetected, and Meyer (1967) reported that variability interfered with previsual detection of tree diseases. Materials and Methods Cotton leaves (Gossypium hirsrttum L., Texas Planting Seed Association I I0) were obtained from plants grown hydroponically in acid-washed, 20- to 30-mesh testing sand in 23-cm diameter, 7.6-1iter capacity, glazed crocks. The sand was rewashed with 0.001 N nitric acid to remove chloride and was leached several times with chloride-free water (silver nitrate test). The basic nutrient solution used was patterned after that of Hoagland and A r n o n (1938). Iron was added as iron-ethylenediaminetetraacetic acid (Nieman and Poulsen, 1967). The nutrient solution had a pH of 7. All plants received one-fourth strength nutrient solution for their first week of growth, and full strength nutrient solution thereafter. Copious applications of nutrient solutions were made by surface irrigation to maintain uniform matric water suction in the substratum. Cotton plants were grown with controlled environment using a 12-hr light-dark cycle. Light illuminance approximated 800 ft-c (8.6 ".~ 10-~ lumen cm-Z), 50 cm above the substratum surface. Ranges of other parameters were as follows: day temperature, 28.6-30.5°C; night temperature, 24.0-25.5°C; day relative humidity, 39-40% ; night relative humidity, 40-45°0. Five leaves of the same chronological age from the third node down from plant apexes were randomly harvested from ten uniform plants. Each leaf was divided into two sections by removing the midrib. The right leaf sections were left untreated (controls) and the left leaf sections were treated with anhydrous ammonia. Five leaf sections were placed on a screen in each of two airtight desiccators, each having a 9040 cm 3 capacity. A watersaturated sponge with a volume of 110.5 cm a had been placed on the b o t t o m of each desiccator to prevent leaves from drying out.

1Contribution from the Soil and Water Conservation Research Division, Agricultural Research Service, USDA, in cooperation with the Texas Agricultural Experiment Station, Texas A&M University. This study was supported in part by the National Aeronautics and Space Administration under Contract No. 160-75-01-07-10. Remote Sensing of Environment I (1969-1970). 1 99-202 199 Copyright ~ 1970 by American Elsevier Publishing Company. Inc.

5C

40

CONTROL ........... TREATED

....... ....... ..

]

,=., 0E Z

3(3

2C

"5oo

io'oo

moo'

zo'oo

25oo

WAVELENGTH (MILLIMICRONS) Fig. 1. Total light reflectance of upper surfaces of cotton leaf sections, control (untreated) and ammonia-treated. Each spectrum is an average of five leaf sections. The wire screen with 0.5 cm e holes was placed above each sponge to separate the leaves from the water-saturated sponges. Both desiccators had the same light and temperature (20°C) conditions. The desiccator used for treatment was injected with 89 cm a of anhydrous a m m o n i a to give a concentration of approximately 10,000 ppm after correcting for the sponge volume. The leaf sections were removed from both desiccators after 3 hr and immediately wrapped in Saran'-' to avoid water loss. Thickness of each leaf section was measured before and after treatment at two locations with a linear displacement transducer and digital voltmeter (Heilman et al., 1968). Reflectance and transmittance measurements were made on upper surfaces of leaf sections with a Beckman Model D K - 2 A spectrophotometer at 50-rap` increments within the wavelength interval 500-2500 mp`. Data have been corrected for a decrease in reflectance of the MgO reference caused by deterioration during aging (Sanders and Middleton, 1953). Analyses of variance were conducted on spectral data (Steel and Torrie, 1960). Pieces of leaf tissue were taken near the center of leaves approximately 2 cm on either side or the midrib. They were fixed in formalin-acetic acid-alcohol (Jensen, 1962), dehydrated with tertiary butyl alcohol, and embedded with paraffin (melting point about 52°C). Transverse sections were obtained with a rotary microtome. Photomicrographs were made with a Zeiss Standard Universal Photomicroscope at a magnification of 100 ×. The 4 x enlargement of a photomicrograph in Fig. 4 represents transections having a thickness of 14 p.. Results and Discussion Figure 1 shows the influence of ammonia treatments on total light reflectance of leaf sections over wavelength interval 500 to 2500 mp`. Discoloration of leaf sections became noticeable soon after placing them in the a m m o n i a atmosphere. As shown later, the discoloration occurred mainly in chloroplast-containing palisade and spongy parenchyma cells. The difference between treated and control spectral means was statistically significant, p = 0 . 0 1 . A m m o n i a treatment of leaf sections reduced

reflectance about 4.5% at the green peak of 550 rap` compared with untreated leaf sections. Between 1000 and 750 nap., a rounding of the plateau occurred, and reflectance decreased 4.1, 20.0, and 29.8% at 1000, 850, and 750 mp. respectively. A m m o n i a treatment had no statistically significant effect on leaf thickness. Over the wavelength interval 1000-2500 rap`, ammonia-treated leaf sections had a b o u t 2% higher reflectance than the controls with crossing-over (lower reflectance) occurring at 1150 mp`. We suspected that the increase in reflectance was caused by water loss because the a m m o n i a was anhydrous. There was no significant difference, however, in leaf water moisture; the tintreated and treated leaf sections contained an average of 77.8 and 77.4°;, water (oven-dry weight basis at 68°C), respectively. Further study is needed on the influence of a m m o n i a on light reflectance over the wavelength interval 1150-2500 mp`. Figure 2 illustrates the effects of a m m o n i a treatments on light transmittance of leaf sections over wavelength interval 500-2500 m/z. Ammonia, compared with the control treatment, reduced transmittance a b o u t 17.0%, at the 550 rap` green peak. R o u n d i n g of the plateau, or decrease in transmittance, occurred in the interval 1350-750 mp`. Tiffs effect was essentially linear (progressive decrease) from 1000 to 750 rap.. Approximate values were 9.9, 26.5, and 39.5% for 1000, 850, and 750 m/x, respectively. Figure 3 indicates absorptance of light over wavelength interval 500-2500 mp`, calculated as a b s o r p t a n c e = 1 0 0 - - ( % reflectance+ % transmittance). It is apparent that the reduced reflectance and transmittance revealed in Figs. 1 and 2 at the 500 m F green peak and over the wavelength interval 750-1350 mF were caused by the increased absorptance shown in Fig. 3. Absorptance progressively increased for treated leaves from 1000 to 750 mF. Values were " L1, 46.5, and 69.2% for 1000, 850, and 750 mp`, respectively. Figure 4 is a representative, unstained transection of an ammonia-treated leaf section showing a brownish discoloration within chloroplast-containing palisade and spongy parenchyma cells. Very little discoloration occurred in upper or lower epidermal cells. Microscopic examinations revealed that cell walls of discolored cells were intact. Some chloroplasts were deformed or ruptured and discolored, but many appeared to 6Of-

,

50

,

/.~'""'"'""~

, / ~

CONTROL

t n," 40 Z 30 I-ZO

Io -'-.% ~

o

500

,doo

,~oo

zo'oo

zsoo

WAVELENGTH (MILLIMICRONSI Zrrade names and company names are included for the benefit of the reader and do not imply an endorsement or preferential treatment of the product listed by the U.S. Department of Agriculture. Remote Sensing of Environment 1 (1969-1970), 1 99-202

Fig. 2. Total light transmittance of cotton leaf sections, control (untreated) and ammonia-treated. Each spectrum is an average of

five leaf sections. R. Cardenas, H. W. Gausman, W. A. Alien, and Marcia Schupp

200

i i

'°° I CONTROL

.........TREATEO

t

:

i

J

~6or uJ <2 auJ

&

$ X ~'10.-

30-

20-

iO

Q~

s6c

,&,,:~

,soo

~o'oo

2~oo

WAV~'L [ NGTH t MtLL~MICRONS}

Fig. 3. Total light absorplance of cotton leaf sections, comro] {tintreated) and ammonia treated. Each spectrum is an average ~,f five leaf sections. be unaffecled by ammonia treatment. The cell cytoplasm ,.~:as USLrally discolored and often appeared to be coagulated. Anhydrous ammonia probably reacted with s~ater in the leaf scction~ to form ammonium hydroxide (NH ~OH) which coadu~ated the cytoplasmic proteins. A leaf is normally highly Iransparcrtt to light over the wavelength interval 750-1350 mt.~. It is theorized that the brownish discoloration increased leaf opaqueness and thereby increased absorptance and reduced reflectance and transmittance over wavelength intervals 750-1350 mff (near infrared) and 500 750 nlff. Since anhydrous a m m o n i a would probably react with water in leaf sections to Form NH.;OH, the discoloration may have

been caused by the reaction of NH.~OH with chlorophyll. It is well documented (Goodwin, 1966) that treatmen~ of chlorophyll #t ritro with a hot alkali (saponification); usually sodium or potassium hydroxide, yields porphyrins containing four pyrrole nuclei. These are red compounds. If saponification of chlorophyll is conducted i~z vitro in methyl alcohol, a brown color is produced. All porphyrins have four absorption bands between 500 and 700 mlZ (Robinson, 1963). Figure 5 describes a second chemical reaction which produces brown pigmentation (Bonnet and Ga]ston, 1952), and which probably occurred concurrently with chlorophyll saponification. The severe treatment with anhydrous ammonia ruptured some chloroplasts and undoubtedl~ affected the permeability ur the chloroplastie membranes, thus releasing poIyphenoloxidase into the cellular cytoplasm. Oxidation and polymerization of po]yphenoloxidase to a brown pigmentation would rcsu[t. This discoloration was also noted when pubescent (hairy) leaves of G.rmtra attrcottiaea [velvet plant) were shaved with an electric razor [Gausman and Gardenas, 1968). The exudales deveJoped a brownish color on the " s t u m p s " remaining after shaving. The preceding examples and other conditions which also causcd decreased reflectances are illustrated in Fig. 6 for the wavelength interval 750-900 raft-. This spectra] range may have practical possibilities as z wavelength band for detecting nonvisua[ symptoms of plant leaf" stresses by remote sensing. As shown in Fig. 6, reflectance in this range was reduced by severe rust infection on Westar wheat leaves (Keegan et aL, 1956), benzene vapor on cotton leaves, natura[ freezing of Cr~ccoloM.~ trrifera (sea grapel leaves (Cardenas and G a u s m a n , unpublished data), a m m a n i a treatment of cotton leaves (this paper), and hair removal by shaving velvet planl leaves (Gnusman and Cardcnas, 1968). At 800 m#, decreases in rcflectances compared with experimental controls were 26.2, 10.8, 4.8, 3.6, and 2.0",, for ammonia, wheat rust, benzene, hair removal, and natural freezing sludies, respectively. Hair removal, freezing, and ammollia and benzene treatments caused discoloration which increased lear opaqueness, with resultant increased absorptance and decreased reflectance. Relative to rust damage Bawden (1933), Clark (1946), and Colweli (I 956) theorized that fungus hyphae penetrate the intercellular spaces and release byproducts that absorb inlrared radiation. The fungal hyphae, however, which penetrate plant cells and form haustoria are hyalin {translucerlt). If the cell is damaged by hyphae penetration, it seems feasible that some oxidation and polymerization of polyphenoloxidase would occur, with a resulting brown color as postulated for results of the ammonia treatment of leaf sections.

OH i HO-'

"' - R " *

PHENOL

O

POLYPHEI"4QLt~XlDASE-~

NO~

-#

O-DIPHEf'IOL

O 0=

~ - R * N,O

OXIO~,TION ~Jn~' POLYM, ERIZ'~TIO~I

>

BROWN PIGMENTS

0 = OUINONE

Fig. 5. D[lgram showing the oxidation and polyn'~erization of polyphenoloxiaase to brown pigments.

Fig. 4. Photomicrograph representing unstained, transverse sections of ammoniaqreated cott~3n leaves, 400 • .

Remote Sensing of Enwkonmem 1 (1969-1970). 1 99-202

The Influence of Ammonia-Induced Cellular Discoloration 201

30,

,~. "-.

i

"'" "'. ".

i

reflectance from the plateau shoulder caused by chlorophyll absorptance.

i

A ~ A M M O N I A (th=s paper} * ~WHEAT RUST (Keegon,1956) o : B E N Z E N E VAPOR ( u n p u b h s h e d ) e : H A I R REMOVAL (Gousman 0ncl Cordenos,1968)

"%-.

Acknowledgments Appreciation is expressed to Dr. C. L. Wiegand for his cooperation and helpful commentary: to Jean Ryan for general assistance: to Ron Bowen for photographic help; and to Guadalupe Cardona for art work.

13:NATURAL FREEZING(unpubhshed)

".A.

25

2oi

References Bawden, F. C. (1933), Infrared photography and plant virus diseases, Nature, 132, 168. Bonner, J., and A. W. Galston (1952). Principles of Plant Physiology, W. H. Freeman, San Francisco. Burns, E. E., M. J. Starzyk, and D. L. Lynch (1969), Detection of plant virus symptom with infrared photography, Trans. Illinois State Acad. Sci. 62 (I). Clark, W. (1946), Photography by hffi'ared, 2nd ed., Wiley, New York. Colwell, R. N. (1956), Determining the prevalence of certain cereal crop diseases by means of aerial photography, Hilgardia, 26, 223-286. Gausman, H. W., W. A. Allen, and R. Cardenas (1969), Reflectance of cotton leaves and their structure, Remote Sells. Environ. 1 (I),

"....

I

Z lad 0 rrLU (3tJJ 0 Z < It3 U.I / LL Ld

"%

15-

19-22.

Gausman, H. W., and R. Cardenas (1968), Effect of pubescence on reflectance of light, Proc. 5th S.vmp. Remote Sells• E f l v h ' o / L , Univ• of Michigan, Ann Arbor, pp. 291-297. Gausman, H. W., and R. Cardenas (1969), Effect of leaf pubescence of Gymlra ata'antiaea on light reflectance, Bat. Go'-., 130 (3), 158-162• Goodwin, T. W. (1966), BiochemLYto, o f Chloroplasts, Academic Press, New York. Heilman, M. D., C. L. Gonzalez, W. A. Swanson, and W. J, Rippert (1968), Adaptation of a linear transducer for measuring leaf thickness, Agron. J. 60 (5), 578-579. Heller, R. C. (1968), Previsual detection of Ponderosa Pine trees dying from bark beetle attack, Proc. 5th Syrup. Remote Sens. Era,iron., Univ. of Michigan, Ann Arbor, pp. 387-434. Hoagland, D. R., and D. I. Arnon (1938), The water culture method for growing plants without soil, Calif. Agr. Expt. Sta. Circ. 347. Jensen, W. A. (1962), Botanical Histochemisto,, W. H. Freeman, San Francisco. Keegan, H. J., J. C. Schleter, W. A. Hall, Jr., and Gladys M. Haas (1956), Spectrophotometric and calorimetric study of diseased and rust resisting cereal crops, Nat. Bar. Stds. Rept. 4591. Manzer, F. E., and G. R. Cooper (1967), Aerial photographic methods of potato disease detection, Maine Agric. Exp. Sto. Bull. 646. Meyer, M. P. (1967), No title, Proc. Workshop htfrared Color Photogrophy m life Plant Sciences, Winter Haven, Florida, Part V, pp. 5-7. Nieman, R. H., and L. L. Poulsen (1967), Interactive effects of salinity and atmospheric humidity on the growth of bean and cotton plants, Botan. Gaz. 128 (I), 69-73. Robinson, T. (1963), The Organic Constituents o f Higher Plants, Burger Publishing Co., Minneapolis, Minn. Sanders, C. L., and E. E. K. Middleton (1953), The absolute diffuse reflectance of magnesium oxide in the near infrared, J. Opt. Sac. Amer. 43 (I), 58. Steel, R. G. D., and J. H. Torrie (1960), Principles and Procedures of Statistics, McGraw-Hill, New York.

i

•

Q--

"'-,,.

° -°..

.

-, 0 L

1

zso

°

TO - f 4 0

--.E.-

aoo

a~o

~oo

WAVELENGTH IN MILLIMICRONS Fig. 6. Influence of treatment of cotton leaves with ammonia (~) and benzene gases (O), rust infection of Westar wheat leaves ('1, hair removal from the velvet plant by shaving (Q), and natural freezing of sea grape leaves ([Z]) on light reflectance from their upper surfaces over the wavelength interval 750-900 mt~. Values plotted in the figure are decreases in reflectances compared with corresponding experimental controls.

Conclusion Cellular discoloration within leaves has important practical implications. The effect on decreasing reflectance (rounding of the plateau) within the wavelength range from 1350 to 750 or 700 m/x, may become useful for detecting nonvisual symptoms of plant leaf stress by remote sensing. A wavelength band of 700-900 mfz may be best, since it contains the sharp drop in

Received August 18, 1969 Revised version received Febrtmry 12. 1970

Remote Sensing of Enwronment I (1 969-1970). 199-202

R. Cardenas. H. W. Gausman, W A. Allen, and Marcia Schupp 202