The Influence of Ultra-Violet Radiation Upon Some Plant Cells

During the summer and early autumn of 1926, experiments were carried out (Delf, Ritson and Westbrook, 1927) to examine the effects of ultra-violet radiation upon plants. In the course of these experiments a number of widely differing plants (e.g. Trifolium subterraneum, Voandzeia subterranea, Pelargonium, etc.) were exposed for short periods daily to the radiations from a quartz mercury vapour lamp, and observations made upon their growth and structure. It was found that the irradiated plants were stunted, and there was a general tendency for the leaves to become rolled up and distorted; frequently the upper surfaces of the leaves were very shiny, and in many plants the exposed leaf and stem surfaces showed a brown coloration. Anatomical investigation of these plants showed that the epidermal cells had collapsed wherever they had been exposed to rays from the lamp, and the firm dead layer formed by these collapsed cells caused rolling and distortion of the leaves during their subsequent growth (Fig. 1c). In Voandzeia subterranea, in particular, browning of the leaf surface was shown to result after three daily exposures of 2 minutes at 3 feet, and to be correlated with collapse and death of the epidermal cells in those regions. Experiments with a variety of plants showed that this epidermal collapse was a general phenomenon, and the subject seemed worthy of further investigation.

Previous workers have noticed this result of exposure to ultra-violet radiation, but in most of their experiments a single long exposure of several hours has been given. Maquenne and Demoussy (1909) and Kluyver (1911) worked with quartz mercury vapour lamps and gave single exposures of several hours to a number of leaves. They found that with leaves such as Aucuba, Ficus, etc., no effect was visible at the end of the exposure, but some hours later a brown or black coloration appeared on the exposed surface. The epidermal cells of such a leaf were seen to be filled with a black-brown mass; by plasmolytic methods it was shown that these cells had been killed, but the palisade cells immediately below were uninjured. This effect is attributed to rays of short wave-length, since no blackening was produced when a glass screen 2 mm. thick was interpolated. By protecting the epidermis with strips of epidermis peeled from another Aucuba leaf Kluyver showed that the epidermal cells could protect from injury, presumably by absorbing these “harmful” rays; much longer exposures were needed to affect the layer of cells below. Both these authors emphasise the fact that the blackening of the leaves is not a specific reaction to radiations from the lamp, but is a result of enzyme action following on death of the protoplast, and can be produced if the cell is killed by other agents, such as heat or chloroform.

Raybaud (1913) grew cress seedlings under continuous irradiation from a quartz mercury vapour lamp, and found that the epidermal cells collapsed and formed a dead layer over the surface of the seedling, probably protecting the inner tissues from the harmful action of the rays.

Ursprung and Blum (1917), using plasmolytic methods for detecting injury have examined the effect of irradiation for short periods (30 sec. to 2 or 3 hours) on a number of plant cells, and have compared their resistance to irradiation. They found that the thickness of the epidermal wall had a great protective influence; young cells and organs were more sensitive than old, and (in Vicia) cells of green leaves were more resistant than those of etiolated leaves.

It appears, then, that this epidermal effect is a widespread phenomenon among plants, and that the more recent experiments have shown that a very short time of irradiation is sufficient to produce an effect ; these shorter times are comparable with the time of irradiation necessary to produce the reddening or “erythema” and subsequent browning induced in the human skin by exposure to ultra-violet radiation. Erythema does not make itself evident until a number of hours after the actual irradiation, there being a latent period of variable duration (generally from 3 to 24 hours) ; the experiments of Maquenne and Demoussy and Kluyver mentioned above indicate that in plants there is a similar interval of time between the cessation of irradiation and the appearance of reaction. Previous workers with plants have paid no attention to the question of the temperature relations of the illumination and subsequent periods, but from results obtained in other fields it would seem that these are not unimportant. Hill and Eidenow have shown that sunburning of the human skin is more marked with high environmental temperatures, while there is a considerable but by no means concordant body of evidence relating to the part played by temperature in the action of ultra-violet light on bacteria, small animals, biologic reactions, etc. (Hill and Eidenow, 1924).

In the present investigation, leaves from a number of different plants have been exposed to short irradiations from a quartz mercury vapour lamp, and the epidermal effect has been examined. Observations have been made upon the differing sensitiveness of the leaves used, and the course of the reactions involved has been followed more closely in leaves of Pulmonaria officinalis. Especial attention has been paid to the question of the lapse of time noticed by earlier workers between the end of exposure and the appearance of a reaction—i.e. the “latent period” and its relation to the exposure given. In view of the work of Hill and Eidenow upon the human epidermis and other workers upon animals, etc., an investigation has been made of the temperature relations of the reactions involved with the plant epidermis.

The source of light in these experiments was a Hewittic “Ulviarc” quartz mercury vapour lamp, the spectrum of which furnished discontinuous bands down to 2230 Å. The lamp was always operated at 150 volts across the arc, with a current of 3·5 amps.; further details have been given elsewhere (Delf, Ritson and Westbrook, 1927).

In order to have a basis for comparison in all the experiments the doses of light were measured in arbitrary “lithopone units’ (L.U.). Crude lithopone paint, a biologically standardised sample of which was obtained from America through the courtesy of Dr Janet Clark of Baltimore, U.S.A., consists principally of zinc oxide and zinc and barium sulphides and darkens under the influence of light of wave-length 2300-3200 Å. (Pfund, 1923). From the graph (ibid. Fig. 6) showing sensitivity at different wave-lengths it appears that the maximum sensitiveness is at 2900 A. At the shortest wave-length used (2300 Å.) the sensitiveness was reduced by only 3–4 per cent., but in the ultra-violet nearer the visible spectrum there is an abrupt fall showing a reduction of sensitiveness at 3200 Å. to less than one-fifth of the maximum value, and diminishing to zero in the region of visible light (4000 Å.). For the present purposes the sensitiveness is thus practically uniform from 2300– 3000 Å.

For measurements, a little lithopone powder is placed on a glass slide, mixed to a paste with distilled water and covered with a quartz slide. It is then exposed directly below the source of light and allowed to darken to a standard tint (the standard was kindly supplied by Dr Clark), which corresponds to a “reflection factor of 50 per cent.” (Clark, 1924). The time of darkening is noted; this is, then, the time taken for the production of one lithopone unit of ultra-violet light at the distance used (1 L.U.). At 3 feet the lamp gave 1 L.u. in about 63 seconds, while at other distances up to 5 feet the time of darkening was proportional to the square of the distance. The ultra-violet output deteriorates with ageing of the lamp; lithopone measurements were therefore taken from time to time so that the exposures could be adjusted where necessary.

Unless otherwise stated, the temperature of irradiation was normal (varying from about 19 to 22° C.), and the specimens were not exposed close enough to the burner or for a sufficient time for there to be any appreciable heating effect. Pot plants and cut shoots were placed directly under the lamp, so that the rays were as nearly as possible normal to the upper surface of their leaves. Excised leaves had the cut end of the petiole wrapped in damp cotton wool, while the lamina was kept flat on a sheet of cork by means of drawing pins. They could then be exposed at definite distances from the lamp. After irradiation the leaves were kept in a damp chamber at laboratory temperature, unless otherwise specified. Progressive changes in the collapse of the epidermal cells were followed by examining fresh at intervals after irradiation, and by preserving portions of the leaves in formalin-acetic-alcohol at similar intervals for examination later.

It has been seen above (p. 293) that the effect of a short daily irradiation upon leaf surfaces is to produce collapse and death of the exposed epidermal cells. In some of the leaves examined—e.g. in Aucuba, Pulmonaria, Privet, Ivy, Elder—this effect was marked by a browning of the epidermal cells ; this was easily detected in surface view and formed a convenient indication of the death of the cells. In transverse section the killed epidermal cells were seen to be filled with brownish contents. As emphasised by Maquenne and Demoussy, this browning is not a specific reaction to ultra-violet radiation but is an enzyme action following on the death of the cells, and is produced in a similar manner when the epidermal cells are killed by means of heat or chloroform¹.

Experiments with leaves of a number of plants (e.g. seedlings of Radish, Lettuce, Vicia, Trifolium) have shown that the time elapsing between the exposure and the subsequent epidermal collapse is dependent upon the dose given. Cotyledons of Radish given an exposure of 39 L.u. showed epidermal collapse after a lapse of about 1 hour, while those of a similar group of seedlings given 18 L.u. reached the same stage of collapse after $212$ hours. It seemed probable that there was a direct relation between the dose given and the lapse of time before the effect appeared, and this was investigated in detail with leaves of Pulmonaria (see pp. 300, 301).

It was noticed that the sensitiveness of different leaves to doses of the same magnitude differed considerably. The younger leaves of a plant were found to be more sensitive than older ones ; a shoot of Aucuba was given an exposure of 39 L.U.—after 24 hours no change was visible in the mature leaves, but the two youngest leaves at the shoot apex showed brown patches on the upper epidermis. Similarly, with shoots of Privet given 39 L.u., brown areas appeared on the youngest leaves after 8 hours, but were not visible on the older ones until 30 hours after the end of the exposure. The thickness of the outer epidermal wall has probably a great influence on the sensitiveness of the epidermal cells. In Symphytum, which has a thin-walled epidermis, browning occurred in 2 minutes after an exposure of 15 L.u., while in Aucuba, where the outer epidermal wall is thick and cuticularised, with an exposure of double this value, browning did not appear until 24 hours later, and then only in the youngest leaves where the cuticle is thinnest (Fig. 2). Schultze (1910) found that the epidermis of Ficus and Impatiens was quite opaque to the magnesium line at 2800 Å while Kohler (1904) has shown that cuticularised, suberised and lignified walls are probably impermeable to light of wave-length 2750 Å. and less. Thus the thickness of the cuticle in leaves of Aucuba would probably explain their greater resistance to ultra-violet light.

The strong absorption of the “harmful” rays by epidermal cells was shown (after the manner of Kluyver, 1911) by stripping the epidermis from various leaves (Pulmonaria, Saxifraga, Helleborus)¹, and laying the pieces on the upper surface of a Pulmonaria leaf. An exposure of 9 L.U. was then given; browning appeared on the exposed regions of the leaf after 1 hour, but the areas protected by adhering epidermis remained green and unharmed.

The changes which result after irradiation have been examined in a variety of leaves, but they have been followed more closely in a herbaceous type—Pulmonaria officinalis (Boraginaceae). Leaves were taken at about the same stage of development, turgid and without brown spots. Owing to the presence of large upstanding hairs they have a rough surface, and in the particular plants used are a rather dark green with large irregular whitish patches. The upper epidermis consists of large somewhat papillate cells with hyaline contents and a thin cuticle. There are numbers of stomata, typically mesophytic in construction; the guard cells are level with their neighbours and contain a number of small chloroplasts. In the green areas there is one layer of loosely packed palisade cells and a spongy mesophyll of irregular cells with large intercellular spaces (Fig. 3 a). In the white areas the usual palisade is replaced by a layer of cells little differentiated from the mesophyll below.

Leaves were exposed to light from the lamp for short times (1–18 L.U.), and then observed at intervals. After a length of time, varying with the intensity of irradiation, the exposed areas began to brown, this being most easily detected in the colourless areas, but taking place equally in the green. Microscopically examined, the contents of the epidermal cells normally are colourless and transparent, while the nucleus is almost invisible. As usual, lethal changes are first shown by the latter, which coagulates so that both reticulum and nucleolus are clearly seen. A little later it begins to darken, and this proceeds until it forms a homogeneous brown mass lying against the inner wall of the cell, while the cytoplasm becomes finely granular and then rapidly darkens too. During these changes the leaf surface is very shiny; then apparently water is lost from the upper surface, the cell contents contract and the upper wall collapses in the middle. In surface view wrinkles and folds can be seen in the walls, and may explain the fact that the irradiated areas now revert to their former dull appearance.

These changes do not take place simultaneously in all the epidermal cells; the first granular appearance is seen in isolated patches of cells, and later on cells can be found in all the progressive stages. In preserved material the same features are discernible (Fig. 3, b, c) but the most striking point is the early disorganisation of the nucleus to one or more irregular masses, at first granular, later homogeneous and taking nucleolar stains uniformly (cp. also Aucuba, Fig. 2.

With a short dose (3 L.U.) the palisade cells appear uninjured (Fig. 3,b), but a more prolonged irradiation (38 L.U.) results in their collapse (Fig. 3 c), with browning of the cell contents and destruction of the plastids. In fixed material it is noticeable that the latter, before their complete disorganisation, show a number of small vacuoles.

It is evident that the reactions, which have been described above, are only the final expression of protoplasmic changes which have been initiated by the rays penetrating into and absorbed by the cells. One may, therefore, define as the latent period the time elapsing between the end of the irradiation and the culmination in visible browning of the reactions thereby induced. The term is used here in a sense comparable with its application to the appearance during development of the latent image on a photographic plate.

In order to determine the relation between the light dose and the length of the latent period, leaves of Pulmonaria were exposed to doses of varying intensity ; these were obtained by altering both the duration of exposure and the distance from the lamp. The lithopone reaction showed that, for rays of 2300 A.-3000 A. (the rays which produce browning) the law of inverse squares held well for distances of $12$ to 5 feet from the burner ; one lithopone unit was found to be the equivalent of 1 minute exposure at 3 feet. Hence, for each leaf exposed, one could calculate the exposure equivalent in terms of lithopone units.

Leaves were exposed at distances of 1–3 feet from the lamp. Parallel strips across the leaf were marked off, and these were given increasing duration of irradiation by uncovering them successively from the apex downwards. Table I A shows a typical result, a leaf being exposed at $212$ feet for times varying from 1 to 12 minutes. In the second column the equivalent of each exposure in lithopone units is calculated. The third column shows the length of the latent period for browning in each case, and from these values are calculated (column 4) the “rate of development of browning of the cells,” this being taken as proportional to the reciprocal of the latent period. In Fig. 4 these values (column 4) are plotted against the exposures in L.U. (column 2). It will be seen that for small doses of 1–10 L.u. the rate is approximately proportional to the dose in L.u. However, as the dose is increased, the rate of browning rises rapidly (part of the curve from 10 to 15 L.U.)— i.e. the latent period becomes very short. The explanation of this probably lies in the difficulty of measuring the latent period accurately. The action of the rays upon the cells probably begins at the beginning of the irradiation period, and the changes set up are already, advanced to some extent at the end of irradiation ; thus the value for the latent period, which is timed from the end of irradiation, is always slightly too short. When short irradiations are given, and there is a long subsequent latent period, this error is not appreciable (first part of the curve, 1–10 L.u.), but when longer irradiations are given, with a short latent period, the error becomes evident (second part of the curve, above 10 L.u.).

From purely theoretical considerations it would be expected that the same effect would be produced by a short dose at high intensity as by a dose equivalent in lithopone units at lower intensity—i.e. at a greater distance. In order to ascertain whether this was the case, leaves were exposed in strips as before, doses of ⁠. being given at distances of feet (Table I B). The rate of development of browning was calculated in each case (see above) and is shown in Table I B, column 3 ; in Fig. 5 these rates are plotted against the exposure in L.u. for distances of feet. An examination of this figure shows that for doses of 2–7 L.u. approximately, the same effect is produced when the same dose is given at distances of feet—i.e. the curves nearly coincide. With doses larger than 7 L.u., however, equal doses seem to have a greater effect when given at shorter distance. The reason for this difference is difficult to see. It cannot be due to the absorption of the air for short wave-lengths of light, since no measurable difference was found in the radiations at 1 and 8 feet (Delf, Ritson and Westbrook, 1927). It may. of course, be a temperature effect, the slightly higher temperatures nearer the lamp causing acceleration of the reaction, but this seems improbable (cp. p. 296).

The unexpected effectiveness of irradiation at short distance was also noticed by Coblentz and Fulton (1924) when the bactericidal action of the quartz mercury vapour lamp was investigated; if the intensity of irradiation was reduced to 1/50th the duration of exposure had to be increased 75 times to give the same effect. With reduction to 1/10th of the intensity the divergence was not so great, but still the time of exposure was increased 13–14 instead of 10 times. These authors controlled the temperature by means of a water-cooled stage, so their results cannot be due to a temperature effect, but they advance no other explanation.

The temperature relations of (1) the irradiation period, and (2) the latent period, were next investigated, since it was thought possible that they might give a clue as to the nature of the reactions concerned.

(1)The temperature relations of the irradiation period were investigated by varying the temperature of irradiation and allowing browning to take place at normal temperature. Three leaves of Pulmonaria were taken and six pieces cut out of the lamina of each. These were kept at approximately constant temperature by floating them, lower surface downwards, on water, one piece from each leaf at each of six temperatures. These floating pieces were exposed to light from the lamp for 4 minutes at a distance of 2 feet (9 L.U.). After irradiation, the pieces of leaf were removed from water and placed on damp filter paper at room temperature and the duration of the latent period for browning noted.

The results obtained in three series of experiments (I, II, III) are shown in Table II (experiments I (a), II (a), III (a)). Column 2 shows the temperature of irradiation, column 4 the duration of the latent period (average of the three leaves used), and in column 5 the rate of development of browning is calculated as before. The results obtained in these three experiments agreed fairly well ; those of experiment I will serve for illustration here. In Fig. 6 the rate of development of browning in this experiment is plotted against the temperature (Curve I). From this one may read off the temperature coefficient for 10° C. rise in temperature (Q₁₀) ; values of Q₁₀ from 0° 25° C. are shown in Table III, column 2. It will be seen from these values that Q₁₀ is consistently low, averaging about 1·1 this low temperature coefficient is characteristic of physical reactions, and its significance will be discussed later.

(2) The temperature relations of the latent period were investigated by irradiating at normal temperatures and varying the sub-sequent temperature during browning. Three leaves were taken and six pieces cut from each as before. All these pieces were placed on damp filter paper and exposed to rays from the lamp for 4 minutes at 2 feet ; one piece from each leaf was then floated on water and kept at each of six temperatures, and the duration of the latent period noted. The results are shown in Table II (second horizontal column of each experiment). In Fig. 6 the results obtained in Experiment I are represented graphically (Curve II); the rate of appearance of browning is plotted against the temperature as before. From the curve obtained temperature coefficients, are read off (Table III, column 3). It will be seen that Q₁₀ here is higher, varying from 1·32 (0–10° C.) to 2·9 (15–25°C.); this higher temperature coefficient is characteristic of chemical reactions.

It appears, then, that between 2° and 25° C. the temperature of irradiation has little effect on the rate of browning, the temperature coefficient being nearly unity ; however, raising the temperature during the latent period after irradiation markedly accelerates the rate of browning, the temperature coefficient for this part of the reaction being about 2·5. Experiments carried out with young leaves of Aucuba japonica showed a similar difference in the temperature relations of the two phases of the reaction; the temperature coefficient for the first part of the reaction—i.e. the irradiation period—was low, of the order of 1·1, while that for the second part of the reaction, the latent period, varied from 1·5 to 1·8. These leaves were of a very different type from those of Pulmonaria, even the youngest leaves having a considerable thickness of cuticle.

These results were not unexpected, for the action of light on the protoplasm must be photoelectric, and for this kind of action the temperature coefficient is very low, Q₁₀ ideally being unity. Henri and Henri (1912) found that exposure to rays from the mercury vapour lamp excited the movements of small animals such as Cyclops, but that from 6° to 37° C. this effect was independent of temperature. This would give Q₁₀ as unity, but the latent period here is so short (e.g. only 0·2 second after 2 seconds’ irradiation) that it is difficult to measure any variations in it.

It is probable that the temperature relations in Pulmonaria and Aucuba correspond to a real difference in the nature of the two phases of their reactions. Bovie (1913) showed that egg albumen, which has a high absorption coefficient for light of short wave-length, is coagulated by exposures to ultra-violet light. At least two reactions are involved: (1) a change produced by the light, and (2) the production of a visible coagulum ; this latter reaction has Q₁₀ equal to 2. The very low temperature coefficient of the first part of the reaction was shown by irradiation of tubes of albumen at various temperatures from o to 50° C. On putting these into a water bath at room temperature the amounts of coagulum which developed subsequently in each were about the same as if they had been in the bath the whole time. A study of the precipitation by ultra-violet light of albumen in relation to its iso-electric point (Clark, 1922) has afforded other evidence for the existence of a primary photoelectric effect followed by chemical change.

The probable occurrence of these changes was recognised by Hill and Eidenow (1924) in their experiments with infusoria and ultra-violet light, but cannot be deduced from their results since they used continuous irradiation at constant temperatures. The temperature coefficient of 3 which they obtained for the interval of 10–20° C. therefore refers only to the aggregate of the reactions taking place.

In the irradiated Pulmonaria leaves it may be supposed that the shorter of the rays which penetrate into the cells and are there absorbed produce a physical change in the constitution of the protoplasm, which finally results in the death of the cell. As the breakdown of its finer organisation proceeds, the equilibrium of the various chemical reactions will be upset, and uncoordinated enzyme action will be permitted. If then, as in Pulmonaria, the suitable oxidases and chromogens are present, browning or blackening will result.

The browning of leaf surfaces in a variety of plants on exposure to short wavelengths of light can be induced by very brief doses (i.e. of about 1–5 L.u.) ; the effect of these doses is not evident at once, but appears only after a short latent period which varies with the intensity of exposure. In this the effect resembles that of the pigmentation of the human skin on exposure to ultra-violet radiation—the initial dose is of the same order of magnitude in the two cases, and the latent period and subsequent appearance of erythema in human skin is comparable with the phenomenon found in plants. Here, however, the resemblance ceases. In the human epidermis erythema and pigmentation are produced without the killing of cells, and with moderate doses the epidermis recovers completely after an interval of several days. Pigmentation after successive exposures is not due to a post-mortem change but to the deposition of melanin round the nuclei of the cells of the basal layer of the skin—a specific reaction to the effect of the light, the melanin crystals probably serving to protect the nuclei of this region from the harmful effects of the ultra-violet rays. In the plant epidermis, on the contrary, pigmentation is due to death of the epidermal cells and can be produced equally well by other agents— e.g. by heat, injury or chloroform. In no case is there recovery of the epidermal cells once browning has set in. Also it has been shown above that killing of the plant epidermis is due to rays of wave-length less than 2900 A. (see pp. 293, 296), while the pigmentation in the human epidermis is induced by rays of 2900–3300 Å., those of wave-length less than 2900 A. producing erythema but no pigmentation (cp. Russell and Russell, 1927, p. 181 ; Hill, 1927). Neither are the temperature relations identical. For example, Hill and Eidenow (1924) irradiated the skin of the fore-arm for a short period at three temperatures: 2°, 16° and 37° C. The subsequent erythema (developed at room temperature) was much the greatest in the last, while with the lower temperatures it was very slight. In Pulmonaria epidermis it was found, however, that between 2° and 25° C. the temperature of irradiation made little difference to the subsequent rate of development of browning. The experiments which have been here described indicate that two phases are involved in Pulmonaria; an initial photolysis caused by the direct action of the ultra-violet rays absorbed by the protoplasm and having a very low temperature coefficient, and secondary reactions thereby induced, which end in the death and browning of the cell and have a high temperature coefficient of the order of magnitude typical of chemical reactions.

The writers’ thanks are especially due to Dr E. M. Delf, under whose direction the work was done, utilising apparatus purchased with the aid of a grant made to her by the Research Committee of the British Association. One of the authors (M. T. M.) thanks the Council of Westfield College and the Department of Scientific and Industrial Research for studentships held during the course of the work.