Planta 9 Springer-Verlag 1990

Infrared thermography of Arum lily infloreseenees Hanna Skubatz 1., Timothy A. Nelson 1, Arthur M. Dong 3, Bastiaan J.D. Meeuse 1, and Arnold J. Bendich 1'2 Departments of 1Botany and 2Genetics, and 3Network and Distributed Computing Center, University of Washington, Seattle, WA 98195, USA Received 26 January; accepted 5 June 1990

Abstract. The infrared radiation emitted from the surface o f inflorescences of 12 aroid species was monitored with an infrared camera, capable of 0.1 ~ C resolution, and the data were converted to temperature values by means of temperature reference standards. Images representing surface temperatures were obtained for Amorphophallus bulbifer Blume, A. eampanulatus Blume, A. forbesii Engl. et Gehrm., A. rivieri Dur., Philodendron selloum Koch, Monstera deliciosa Liebm., Dracunculus vulgaris Schott, Arum italicum Mill., A. dioscoridis Sibth., A. creticum Boiss et Heldr., Caladium sp., and Remusatia vivipara Schott. These images were different among species with respect to temperature, duration of detectable heat development, and organ type (male and female flowers, spathe and appendix) found to be thermogenic. All these species, however, exhibited three c o m m o n characteristics: 1) production of heat by the male flowers; 2) pollen-shedding immediately after heat production had ceased; and 3) when male flowers were some distance away from female flowers along the spadix, heat was not detected in female flowers. Heat emission was associated with the alternative, cyanide-insensitive pathway that was fully operative. Key words: Araceae - Respiration - Temperature (measurement by infrared) - Thermogenicity

Introduction To date, thermogenicity has been monitored either in the field or in the greenhouse with thermometers, thermocouples and thermistors (Brattstrom 1972; Seymour et al. 1983; Tang 1987). Since this technique is not very sensitive, weakly thermogenic organs and/or species have not been identified. Heat production has been doc* To whom correspondence should be addressed Abbreviation: SHAM = salicylhydroxamic acid

umented in only a few aroid species (James and Beevers 1950; Meeuse 1975; Meeuse and Raskin 1988). In order to gain a better understanding of how widely distributed thermogenicity is in the Araceae, the pattern of surface temperature was analyzed in 12 aroid species by using a sensitive infrared camera. Heat production has been shown to be associated with a highly active cyanideinsensitive respiration in some aroid species such as Sauromatum guttatum Schott and Arum italicum Mill. (Meeuse and Raskin 1988). We, therefore, have also assayed the cyanide-insensitive respiration in thermogenic tissues obtained from 7 of the 12 aroid species studied.

Material and methods Plant material. Amorphophallus bulbifer Blume, A. campanulatus Blume, A. forbesii Engl. et Gehrm., A. rivieri Dur., Philodendron selloum Koch, Monstera deficiosa Liebm., Arum creticum Boiss. & Heldr., Caladium bicolor sp., and Remusatia vivipara Schott were grown in a greenhouse. Dracunculus vulgaris Schott, Arum italicum Mill. and Arum dioscoridis Sibth were grown in an outdoor garden. Thermal image processing system. A thermographic system was used to obtain inflorescence temperatures. An infrared camera (Thermovision 780; AGA Infrared Systems, Lidingo, Sweden) was connected to a digital recorder (Oscar; AGA) to record the data on magnetic tape. Emission of wavelengths from 2 to 5.6 gm was detected by the camera. It required 1.5 s to obtain an image composed of 112 x 128 picture elements. Each element was in the form of one 8-bit value. The image data were analyzed with a collection of computer programs entitled "Sofia" that was provided by the AGA company. The original FORTRAN code was modified to run on a VAX 6000-440 computer operating under VMS 5.3. (4 central processors; 128 Mbytes physical memory ; 17 Gbytes disk storage space, of which 5 Mbytes was available). The images were calibrated with respect to two temperature standards in the field of view. False-color images were produced by assigning colors to temperatures. Although the calibration data had a precision of 0.1 ~ C, only 13 colors were used to display the temperature range from 0~ to 42~ C, for clarity of presentation. The images were then printed on a Versatec Electrostatic Plotter (model C3436). Imaging data were collected in a darkroom at 5rain intervals when surface temperature reached a maximum, and at longer intervals during the decline of the surface temperature.

H. Skubatz et al. : Infrared thermography of Arum lily inflorescences From 50 to 100 images were recorded for each species during the thermogenic response and images were analyzed only for periods during a change in surface temperature pattern. At least two inflorescences were examined from each species. Respiration in thermogenic tissue. Tissue slices of 0.5 mm thickness

were obtained from thermogenic parts of inflorescences of seven aroid species. Tissues were sliced with a microtome (Vibratome 1000; TPI, St. Louis, Mo., USA) during the thermogenic phase (time of day given in parentheses) of Arum italicum (19:30), Arum dioscoridis (I 3:00), Amorphophallus campanulatus (10: 00), A. rivieri (10:00 in the first day of heat-production), A. bulbifer (10:00 in the second day of heat production), and Philodendron selloum (1 d after heat production). Ten to twenty milligrams of slices were used from thermogenic parts of Arum italicum, Arum dioscoridis and Amorphophallus campanulatus, and about 50 mg from A. rivieri, A. bulbifer, and Philodendron selloum. The slices were transferred into 4 ml of fully aerated buffer (50 mM N-hydroxyethylpiperazine-N'-2-ethane sulfonic acid (Hepes), pH 6.6) at 22~ C. The oxygen content of air-saturated water was 237 gM. After 5 min of incubation in the buffer, the rate of oxygen consumption was determined by using a Clark oxygen electrode (Estabrook 1967). Oxygen consumption was determined again 10 rain after incubation with either I mM KCN, or salicylhydroxamic acid (SHAM) at the indicated concentration, or with both inhibitors.

Results Therrnogenicity patterns in different aroid species

Surface temperature of inflorescences of twelve aroid species was monitored during anthesis. In each species, except for Monstera deliciosa, the male and female flowers are separately arrayed in a sheath on the spadix such that the surface temperature of each flower type could be distinguished. The temperature level, the inflorescence organs emitting heat, and the duration of detectable temperature varied among the species. Figure 1 depicts four highly thermogenic species. The inflorescence of Philodendron selloum has a complex pattern of surface temperatures (Series 1). It is thermogenic for two successive days. On the first day of flowering, when the spathe unfolds, the male flowers are hot, reaching a peak of 40 ~ C (1A). Later in the morning, surface temperature declines selectively, faster in the lower part than in the upper part of the spadix (IB, C). After midday, surface temperature increases in the lower part and reaches a maximum of 36 ~ C (ID); later it slowly declines. At 01:00, after midnight, the spadix is still thermogenic (1E). The next morning, surface temperature increases again, but this time the lower part of the spadix is hotter than the upper part (IF, 1G). The temperature then slowly declines (1H) followed by a period of little change during the second day (1I, I J). At 20:00 only the lower part is weakly thermogenic (IK). The next morning (the third day), the spadix is colder than ambient. The female flowers are not thermogenic (data not shown). The inflorescence of Arurn dioscoridis is also highly thermogenic, some parts reaching 40 ~ C on the day of inflorescence opening (Series 2). Early in the morning, the male flowers (but not the female flowers) and the appendix are thermogenic (2A). Surface temperature of

433 the male flowers then declines while the appendix stays hot (2B, 2C). Later in the morning (09:55), the male flowers are no longer hot and the temperature of the appendix declines (2E, 2F). After midday, the inflorescence is no longer thermogenic (2G). In images 2A-C, the spathe is attached to the spadix and is colder than ambient. In images 2E-F the spathe was removed from the inflorescence before the thermographic recording was made. The inflorescence of Arum italicum is thermogenic for less than 1 h in the evening. When the spathe unfolds, heat is released from the male flowers and from the appendix (Series 3). Images 3A-C show two inflorescences whose spathes unfolded at the same time. In both inflorescences, the male flowers were thermogenic (3A) but only one of the appendices produced heat. The temperature of the male flowers declined quickly, in about 20 rain, while the appendix remained hot for about 1 h. The inflorescence of another species, Amorphophallus campanulatus, is thermogenic in the morning (Series 4) and the main source of heat is the male flower-zone. Early in the morning, heat is released from the male flowers (4A) and around 11 : 00 part of the aerenchyma (the tissue located above the male flowers) is also thermogenic (4B). Later, at noon, heat production starts to decline (4C) and around 16:00 the inflorescence is at ambient temperature (image not shown). Close examination showed that the eight other aroid species studied were weakly thermogenic, reaching less than 30 ~ C. Amorphophallus bulbifer is thermogenic for 3 d (Series 1 of Fig. 2). On the first day of flowering, when the spathe unfolds, the male flowers are weakly thermogenic in the afternoon: about 1~ C above ambient (IA, B). Temperature increases somewhat on the second day (1C-E) and on the third morning, the temperature of the male flowers increases further (1G) before declining (1H-K). The temperature of most of the male flowers reaches 28 ~ C, but some of them are even warmer (orange spots in the male fiower-region in 1G). It is only on the third day that the appendix becomes thermogenic (1G-K). Another species examined is Monstera deliciosa (Series 2). Its inflorescence is faintly thermogenic during two successive mornings. The first period of heat production occurs when the spathe is only slightly unfolded; on the next morning, when the spathe is completely unfolded, the second period of heat production takes place. The spadix temperature rises to 1-2 ~ C above ambient for each morning (2A, C represent one of the two mornings). In Amorphophallusforbesii, the male flowers and the appendix are slightly thermogenic in the afternoon, between 15:00 and 17:00, for three successive days (Series 3). In Dracunculus vulgaris, the male flowers are thermogenic for 1 h in the morning (Series 4). Although not evident in this thermographic image, the appendix temperature is slightly above ambient. In Arnorphophallus rivieri heat is produced during three successive mornings. Three parts are thermogenic: the male flowers (A in Series 5), the appendix and part of the spathe (5B). Three other species were weakly thermogenic after the

434

H. Skubatz et al. : Infrared thermography of Arum lily inflorescences

DAY 1 1 09,10

DAY 2 1 09~40

08,45

19:15

11:30

15,30

09:10

09,50

01100 0 8 : 2 0

09,50

09,55

11:45

19:25

08,45

11:30:13:00

15:00

20:00

12:45

19:55

tt,15

~ N 28- 3O IR30-52 " .32-34 ~34-38 ~?~*~'~24 -25 [---138-42 25-26 9 I I 26-27 ' ',7cm ~;~:::,!;~27-28 12:05

Fig. 1. Time sequence of surface temperatures of four highly thermogenic aroid inflorescences: Philodendron selloum in Series 1, Arum dioscoridis in Series 2, Arum italicum In Series 3, and Amorphophallus campanulatus in Series 4. The time of day is indicated above each image. Data from two successive days are shown for P. selloum. A day runs from 01:00 to 24:00. Temperatures for two separate inflorescences are depicted for A. italicum (Series 3, A-C). The key on the right side shows the range of temperatures

in degrees Centigrade. Each color represents a range from greater than the first value to a temperature equal to the second value. For example, the blue color represents temperatures from above 23~ to 24~ C. The size scale is for the thermographic images only; it does not apply to the four photographs of the inflorescences arrayed vertically on the left side. Ap, appendix; M, male flowers; F, female flowers; sp, spadix; S, spathe; Ae, aerenchyma

spathe unfolded in the morning, between 8:00 and 12:00" Arurn creticurn, Caladium sp., and Rernusatia vivipara (data not shown).

chrome pathway was 0.8 gmol O 2 . m i n - l . ( g FW) -1. In the weakly thermogenic species the rate o f oxygen consumption was much lower: e.g. 0.3 lamol 02 . m i n - 1 . (g F W ) - t for Arnorphophallus forbesii. The capacity o f the cyanide-insensitive pathway, however, was high (0.6 gmol 02" min - 1. (g F W ) - 1) and its engagement was low (0.2 gmol O2" m i n - 1.(g F W ) - 1). It is therefore probable that the cyanide-insensitive pathway in these thermogenic tissues is involved in the thermogenic response. In two Amorphophallus species, A. forbesii and A. rivieri, cyanide stimulated oxygen uptake when it was added before SHAM, indicating that the capacity o f the

Tissue respiration. Cyanide-insensitive respiration was found in all the thermogenic tissues examined (Table 1). In highly thermogenic species such as Arum italieum, both total oxygen consumption and the capacity of the cyanide-insensitive respiratory pathway were high, 1.9 and 1.1 gmol O 2 . m i n - l . ( g FW) -1, respectively. The cyanide-insensitive pathway was fully engaged, 1.3 lamol O2"min - l ' ( g FW) -1, and the capacity o f the cyto-

H. Skubatz et aL : Infrared thermography of Arum lily inflorescences

DAY 1 t, r

DAY 2 t

~,5:50 09:.30 "t0:00

08:40

10:10

r

:

435

DAY .'5 t

2t,00 09:50

t0:05

1t:00

t6:30 23 ~00

12:00

09:00

11:00:

B0

-20 20-22

:24-25 ~25-26 Iml26-27 27-28 28-50 t ~7cm

Fig. 2. Tim sequence of surface temperatures for five weakly thermogenic aroid inflorescences: Amorphophallus bulbifer in Series 1, Monstera deIiciosa in Series 2, Amorphophallus forbesii in Series

3, Dracunculus vulgaris in Series 4, and Amorphophallus rivieri in Series 5. Other information as in Fig. 1

Table 1. Respiration rates in thermogenic aroid inflorescences. Total respiration in tissue slices was determined after 5 rain of incubation in buffer (Vtot,~). The capacity of the cytochrome pathway was determined after the addition first of SHAM and then of K C N (Voyt). The capacity of the alternative pathway was determined after the addition first of 1 m M K C N and then of S H A M (V~k). The engagement of the alternative pathway was determined after the addition of the first inhibitor, SHAM (V',,). No residual respiration was detected after the addition of both SHAM and KCN. SHAM was added at a concentration of: 5.5 m M for Arum italieum, 3.5 m M for A. dioscoridis and Philodendron selloum, 7.5 mM for Amorphophallus bulbifer, and 3.5 m M for A. rivieri. The slices were obtained at different times during anthesis, as described in Materials and methods. Data are in gmol 0 2 - r a i n - 1 (g F W ) - t . Each value represents the mean of one to four separate determinations • SD. n.d. = not determined Species

Organ

Vtata[

Veyt

V~lt

Vtalt

Arum italicum A. dioscoridis AmorphophalIus eampanulatus A. bulbifer A. forbesii A. rivieri Philodendron selIoum

Appendix Appendix Aerenchyma Appendix Appendix Appendix Spadix

1.9 • 0.3 0.9 _+0.3 1.5 • 0.2 0.8 • 0.01 0.3 • 0.3 0.4_ 0.1

0.8 • 0.2 0.4 • 0.1 0.8 • 0.1 n.d. 0 n.d. 0.2

1.1 1.1 • 0.2 1.1 • 0.1 0.8 • 0.0 0.6 1.2 0.3

1.3• 0.3• 1.0• n.d. 0.2•

n,d. 0.2

436 alternative pathway (Table 1, line 7), gait, exceeds Vcyt in these species (Table 1, line 7). Inhibition of the cytochrome pathway by cyanide probably diverted the electron flow to the alternative pathway. Discussion

Thermography has in a few cases been used to monitor leaf temperature (Hashimoto et al. 1984; Raskin and Ladyman 1988). We have used this method to demonstrate that in addition to the species previously studied (Meeuse and Raskin 1988; Seymour et al. 1983), eight other aroid species are thermogenic: AmorphophaIlus

bulbifer, A. forbesii, A. rivieri, Dracunculus vulgar&, Monstera deliciosa, Arum creticum, Caladium bicolor, and Remusatia vivipara. We found for each species that on the first day of anthesis it is the male flowers that first become thermogenic and that heat is produced later in other inflorescence organs (including the spathe of A. rivieri). The same pattern is also seen in Sauromatum guttatum inflorescences (data not shown). These observations indicate that a signal moves from the male flowers to other thermogenic parts, as had previously been demonstrated by van Herk (1937) and by Meeuse (1975). Salicylic acid triggers the thermogenic response in the appendix of S. guttatum (Raskin et al. 1987), and is also found in the male flowers of Sauromatum (Raskin et al. 1989), but its translocation from these flowers to the appendix has not been demonstrated. The occurrence of cyanide-insensitive respiration in these thermogenic tissues is not surprising and supports the idea that the alternative, cyanide-insensitive pathway is involved in heat production, at least in the aroids. The presence o f alternative oxidase in two species (Amorphophatlus rivieri and Arum italicum) has recently been demonstrated (Elthon and McIntosh 1987). The fact that heat production occurs in some species (Monstera deliciosa and Amorphophallusforbesii) over several days and at particular times of the day raises the question of how the flow of electrons through the cytochrome and the alternative pathways is regulated. Is the cytochrome pathway first saturated at one time of day after which the alternative pathway is used as an overflow route (Bahr and Bonner 1973)? Or, does the electron flux switch from one pathway to another, as has been demonstrated in mung-bean mitochondria (Wilson 1988)? The spatial and temporal complexity of the heatproduction process raises another question. Does salicylic acid trigger heat production in all the species examined? Since salicylic acid is only one of the two chromatographically separable compounds in the original preparation of "calorigen" (Chen and Meeuse 1975), more than one compound may act to trigger this complex process. Thermogenicity in aroid spadices is a highly controlled phenomenon. It occurs only in certain organs or, as exemplified by the aerenchyma of Amorphophallus campanulatus, in certain parts o f organs. It can occur at different times o f the day and also during the night (P. selloum) and evening (Arum italicum). A pattern, however, has emerged for the aroids in general. On the

H. Skubatz et al. : Infrared thermography of Arum lily inflorescences first day of inflorescence opening, male flowers become thermogenic. Later the temperature rises in other organs of the inflorescence. Pollen is not shed until after the temperature of the male flowers has declined. No detectable heat was produced (0.1~ C limit of sensitivity) by female flowers for any of the species although for Monstera deliciosa the interspersion of male and female flowers on the spadix made it impossible to identify which flowers released the heat. Considering this diversity in the timing and tissue responsiveness, we suspect that more than one agent can act to trigger the thermogenic response in aroid species. We acknowledge the help of Dr. Alan H. Rowberg and Lawrence E. Gales in writing computer programs. We also thank Dr. R.D. Stevenson for his advice, and Dr. Arthur W. Guy for the use of his thermographic imaging equipment. T.A.N. was supported by a National Science Foundation Graduate Fellowship. References

Bahr, J.D., Bonner, W.D. (1973) Cyanide-insensitive respiration. II. Control of the alternative pathway. J. Biol. Chem. 248, 3446 3450 Brattstrom, B.H. (1972) Temperature changes in heat producing plants. Bull. South. Cal. Acad. Sci. 71, 54-55 Chen, J., Meeuse, B.J.D. (1975) Purification and partial characterization of two biologically active compounds from the inflorescence of Sauromatum guttatum Schott. Plant Cell Physiol. 16, 1-11 Elthon, E.T., Mclntosh, L. (1987) Identification of the alternative terminal oxidase of higher plant mitochondria. Proc. Natl. Acad. Sci. USA 84, 8399-8403 Estabrook, R.W. (1967) Mitochondrial respiratory control and the polarographic measurement of ADP:O ratios. Methods Enzytool. 10, 41-47 Hashimoto, Y., Ino, T., Kramer, P.J., Naylor, A.W., Strain, B.R. 0984) Dynamic analysis of water stress of sunflower leaves by means of a thermal image processing system. Plant Physiol. 76, 266-269 James, W.O., Beevers, H. (1950) The respiration of Arum spadix. A rapid respiration, resistant to cyanide. New Phytol. 49, 353373 Meeuse, B.J.D. (1975) Thermogenic respiration in aroids. Annu. Rev. Plant Physiol. 26, 117-126 Meeuse, B.J., Raskin, I. (1988) Sexual reproduction in the arum lily family, with emphasis on thermogenicity. Sex. Plant Reprod. 1, 3-15 Raskin, I., Ladyman, J.A.R. (1988) Isolation and characterization of a barley mutant with abscisic-acid insensitive stomata. Planta 173, 73-78 Raskin, I., Ehmann, A., Melander, W.R., Meeuse, B.J.D. (1987) Salicylic acid - a natural inducer of heat production in arum lilies. Science 237, 1601-1602 Raskin, I., Turner, I.M., Melander, W.R. (1989) Regulation of heat production in the inflorescence of an Arum lily by endogenous salicylic acid. Proc. Natl. Acad. Sci. USA 86, 2214-2218 Seymour, R.S., Bartholomew, G.A., Barnhart, M.C. (1983) Respiration and heat production by the inflorescence of Philodendron selloum Koch. Planta 157, 336-343 Tang, W. (1987) Heat production in cycad cones. Bot. Gaz. 148, 165-174 Van Herk, A.W.H. (1937) Die chemischen Vorg/inge im Sauromatum-Kolben. Rec. Tray. Bot. N6erl. 34, 69-156 Wilson, S.B. (1988) The switching of electron flux from the cyanide-insensitive oxidase to the cytochrome pathway in mung bean (Phaseolusaureus L.). Biochem. J. 249, 301-303