Citrus canker (Xanthomonas axonopodis pv. citri) is one of the most important citrus diseases and its control is made with the pathogen exclusion and with eradication measures. The detection chances of diseased plants are reduced in groves with low incidence of the disease. Considering the difficulties of citrus canker visual diagnosis and the advantages of a technology that can improve the detection process of this disease in the field, the researchers worked to identify the unique spectroscopic characteristics of plants with this disease. The objectives of this study was to identify the spectroscopic characteristics of the leaves with citrus canker and correlate them to the changes in photosynthetic efficiency of the diseased foliar tissue, as well as to relate the severity of the disease to its photosynthetic efficiency. Spectroscopic profiles and photosynthetic efficiency of plants of the Citrus sinensis ('Hamlin'), Citrus reticulata ('Ponkan') and Citrus limonia ('Cravo') species were assessed before and after the inoculation with Xanthomonas axonopodis pv. citri. The model y = (1 - x)ß, where y represents the relative net assimilation of CO₂ (photosynthesis of diseased plants in relation to controls) and x the severity of the disease, was adjusted to the data by non-linear regression. The monomolecular model was adjusted to the relationship between relative net assimilation of CO₂ and spectroscopic reasons of the region of the red, also by non-linear regression. The conclusion was that the disease affects the photosynthetic efficiency of 'Ponkan' and 'Hamlin'. In 'Hamlin', the disease not only reduces the foliar area, but it also reduces the remaining green tissue photosynthesis. There is a positive relationship between the spectroscopic ratios of the red region and the photosynthesis in the tested varieties.

The Xanthomonas axonopodis pv. citri bacterium (Haase) (VAUTERIN et al., 1995) is the causal agent of Asiatic citrus canker, also known as cancrosis A, one of the greatest threats to citrus culture (STALL; SEYMOUR, 1983). It is one of the diseases that cause the largest economic damage, affecting world citriculture, and it is the reason to apply quarantine laws, and of eradication in many countries (BRUNINGS; GABRIEL, 2003).

In Brazil, especially in the State of São Paulo, because it is a quarantinable disease, its control demands the adoption of exclusion and eradication measures of the pathogen. Since the State of São Paulo is responsible for over 90% of the Brazilian citrus production, and over 95% of the volume is exported (as FCOJ or concentrates, and fresh fruit), the laws and regulations in force imply the partial or total eradication of groves after contaminated plants are detected.

The incidence of the disease is used as an eradication criterion. The eradication campaign - although it did not totally eliminated the bacterium in the State of São Paulo - can be considered successful (RODRIGUES NETO et al., 2004), since it maintains the low levels of incidence of the disease in the noble area of culture (BARBOSA et al., 2001). Sampling survey performed annually revealed that the contamination rate by citrus canker was 0,19% in the citrus planting fields checked in 2006, i.e., 99.8% of the citrus planting fields of the State of São Paulo and south of the Triângulo Mineiro are free from the disease (MASSARI; BELASQUE JÚNIOR, 2006).

To inspect the groves routinely is one of the most important measures to prevent citrus canker. This procedure is performed by trained inspectors of the Fundo de Defesa da Citricultura (Fundecitrus), which inspect all the groves streets and roads looking for diseased plants, through visual diagnosis of the citrus canker symptoms. However, considering the difficulties to inspect citrus canker and the potential economic impact of a technology to improve the detection process of this disease in the field, studies are being carried out since 2002, in an attempt to characterize spectroscopically the plants contaminated by citrus canker, in order to identify possible unique spectroscopic characteristics of a plant with this disease.

Therefore, the purpose of this study was to identify some of the biological phenomena involved in the responses obtained by healthy vegetal tissues and tissues with citrus canker when excited by laser. Specifically, in plants of the Citrus sinensis (L.) Osbeck, Citrus reticulata Blanco and Citrus limonia (L.) Osbeck species, the researchers: (i) checked the spectroscopic characteristics of the leaves with citrus canker and correlated them to the changes in the photosynthetic efficiency of the diseased foliar tissue; (ii) related the severity of the disease to the leaves photosynthesis efficiency.

The experiments were carried out in a greenhouse located in the Instituto Biológico, in the city of Campinas, State of Sao Paulo.

In the experiments, 15 of each citrus nursery trees acquired from a reliable nursery, of the Citrus sinensis (L.) Osbeck (sweet orange, Hamlin variety, grafted on Citrus limonia), Citrus reticulata Blanco (tangerine, of the Ponkan variety, grafted on Citrus limonia) and Citrus limonia (L.) Osbeck (limão Cravo) were used, totaling forty-five nursery trees.

Before the experiment was installed, the nursery trees were transplanted to twelve-liter plastic vases filled In with soil mixed with a substract in the proportion of 50% for each component of the mix. The plants were pruned approximately two months before the experiments were installed for the leaves that were used in the appraisal would be the same age in the inoculation date.

The five youngest leaves of each plant were inoculated with a suspension of the isolate IBSBF 1421 of Xanthomonas axonopodis pv. citri. Two repetitions of the experiment were performed in time (experiment 1 and 2), both using the same methodology and with the same assessment criteria.

The inoculations were by wounds with hypodermic needles, previously embedded in a bacterial suspension of 10⁷ ufc/mL and introduced in the foliar limbo until it crossed its whole thickness.

Treatments performed were: (i) control, that corresponded to 2 perforations with a needle embedded in a PBS tampon (Phosphate Saline tampon); (ii) average severity which corresponded to two perforations with a needle embedded in the bacterium suspension of 10⁷ ufc/mL; (iii) high severity, which corresponded to 4 perforations with a needle embedded in the bacterium suspension of 10⁷ ufc/mL. To perform the perforations they were delimited by a frame of 6 cm² in area in order for the wounds to fit in the area that would be within the photosynthesis analysis camera. Five plants were used as repetition in each treatment, in the experimental delineation.

Immediately after the treatment applications, the plants were covered with humidified plastic bags and maintained in a humid camera condition during one night, to favor the inoculation effectiveness. After that period, the plastics were removed from each plant and the experiments were carried out, complying with all the phytosanitary safety rules applicable to citrus canker.

AAfter inoculation, the plants were assessed every day to determine the incubation period. On each assessment date, after the appearance of the symptoms, the 6 cm² area delimited by the original frame was photographed by a digital camera. The scanned images were transferred to a computer and, subsequently processed with a disease quantification program QUANT (v.1.0) (VALE; FERNANDES FILHO; LIBERATO, 2003) to evaluate the severity of their symptoms. The yellow halo that was frequently found on leaves was considered part of the lesion, i.e., considered in the final result of the severity.

The assessments of the physiological parameters, the net rate of photosynthesis with light saturation, transpiration rate and breathing rate in the dark, were performed with a portable infrared gas analyzer (IRGA) model LI-6400 (Li-Cor, USA ) in inoculated plants and in control plants, and one leaf per plant was evaluated (Figure 1).

To evaluate the photosynthesis and the transpiration, leaves remained in the analysis chamber for some minutes until the reading stabilized. To evaluate Breathing in the Dark, the light source was turned off and the result collected after reading stabilization.

In experiment 1, the first assessment was performed one day before the inoculation (1 ai), situation in which all the leaves were healthy and with no wounds; a new assessment was performed one day after inoculation (1 dai). The subsequent assessments were made weekly, at 7 and 13 days after inoculation (7 dai and 13 dai, respectively). From this point onwards, the assessments were made every 15 days with reading made on the 28, 44, 58 and 72 days after the inoculation (28 dai, 44 dai, 58 dai and 72 dai, respectively), finalizing this experiment.

In experiment 2, there was also an assessment one day before the inoculation (1 ai) and the experiment performed at 1 dai was repeated. However, different from what happened during the first experiment, the readings, until the end of the experiment were at 9, 16, 21 e 42 dai.

During the assessments some conditions were standardized so that the readings would not suffer any type of changes caused by factors that were not related to the disease. Therefore, certain measures were taken such as the use of a mixer to inject the CO₂, maintaining the CO₂ concentration constant in the air that enters in the chamber at approximately 380 ppm, and, standardizing the airflow by a camera at 350µmol s^-1. In order to guarantee the photosynthetic flow of photons constant during the assessments, an artificial light source of the own equipment was used regulated for 1000 µmol.m^-2.s^-1, hence guaranteeing the luminic saturation point for the three varieties under assessment.

A portable fluorescence spectroscopy system (Figure 2) was responsible for collecting the spectra, which were functioning during experiment 2, in which all the plants were assessed (one leaf per plant) as to the responses after the excitation with laser, with the advantage of not being a destructive technique for the samples. Green light was used as fluorescence excitation.

The first spectroscopic reading occurred one day before inoculation (1 ai), situation in which all the leaves were healthy and had no wounds; a new reading was performed one day after inoculation (1 dai), and the subsequent readings followed the same assessment days of the physiological parameters (9, 16, 21 e 42 dai).

In each reading, the spectra were collected in different points of the leaves with and without symptoms, collecting tree spectra per evaluated leaf. The spectra were collected from the same leaves in which the results obtained with IRGA, in the transition region between the yellow tissue (when that occurred in the leaf) and the green region, avoiding directing the point of the fiber to the area that was already necrosed.

To compare between the intensity of the spectra, their format had to be standardized, normalizing them by the excitation peak (532 nm) that was not blocked. Hence, all spectra were in a same scale, where the excitation peak is the same.

For comparison the ratios RF/ESF and FRF/ESF, where RF is the fluorescence at 685 nm and FRF the fluorescence at 735 nm and ESF (Elastic Scatering Fluorescence), represents the normalization of the spectrum at 532 nm. Therefore, the spectra were normalized at their excitation peak (532 nm) to eliminate possible external factors such as the laser variation power, angle between the fiber and the leaf, dust, etc.

The linear model was adjusted to the increment data of the disease severity data (dependant variable) in time (independent variable).

The Bastiaans model (1991), y = (1 - x)^ß, in which y represents the relative net assimilation of CO₂ relativa (photosynthesis of diseased plants in relation to controls) and x represents the severity of the disease, both in proportion, and ß represents a parameter of the model, was adjusted to the data obtained in the two experiments through non-linear regression by the minimum squares method, using the STATISTICA program (StatSoft, Tulsa, U.S.A.). In the significant regressions, the parameter ßwas compared to 1 by the t test.

The monomolecular model y = b1*(1-b2* exp(-b3*x)) (in which y represents the assimilation of CO₂ and x represents the ratios RF/ESF and FRF/ESF, b1 represents the value of the curve stabilization, b2 relates to the point of interception of the curve and b3 represents the coefficient of the curve) was adjusted to the data of the two experiments through non-linear regression.

The increase in the severity of inoculated plants was linear in the three varieties, in both experiments, and corresponded to the expansion of the lesions around the injuries.

In experiment 1, the period of incubation was reached around the seventh day after inoculation for the three varieties, with the exception of the average severity for Citrus limonia and Hamlin oranges, in which the appearance of symptoms occurred approximately one week after. In spite of the above, the Hamlin variety, since the first assessment, showed bigger lesions. Citrus limonia was always more resistant, in both experiments with small lesions and slow growth (0.043 to 0.084 cm/day).

In experiments 1 and 2, in addition to the necrotic lesions being more conspicuous in the Hamlin variety, the yellow halo was also larger in that variety, followed by the Ponkan variety and not found on the leaves of l Citrus limonia, which presented the lowest severities.

In experiment 2, the period of incubation for the three varieties, in both severities, was reached about 16 days after the inoculation. This delay in the appearance of symptoms when compared to the previous experiment, possibly occurred because of lower average temperatures of the days between the inoculations of the bacterium until the symptoms appeared in this experiment. The maximum severity obtained at the end of the experiment, taking into consideration the different treatments, did not exceed 5%, and was much lower to that found in experiment 1 (approximately 18%).

The model proposed by Bastiaans (1991) was adjusted to the values of the relative photosynthesis and of the disease severity obtained in experiments 1and 2, analyzing together the data of both experiments for the Ponkan and Hamlin (Figure 3) varieties. The non-linear regressions were significant (p < 0,01) for the varieties mentioned before, in spite of the low values of the determination coefficients (R²) obtained in the adjustment of the model (Table 1). For those two varieties, the data obtained can justify the model, i.e., there is a reduction in the net assimilation of CO₂ (photosynthesis) of the diseased leaves with an increase in the disease severity, and the values estimated for ß were, respectively, higher and equal to 1 (t test), for the varieties Hamlin and Ponkan at 5% of probability. In the case of the Cravo (Citrus limonia) variety, observations could not be made because it presented non-significant regression, i.e., the model did not adjust to the values obtained in the assessments.

There were no differences between the species in what concerns the responses obtained by laser excitation nor was an evident difference observed among the different severities within the same species. Due to this equality among the species in the responses to laser, researchers opted for the joint analysis of the data, in an attempt to correlate the reading obtained by leaf excitation with laser with the photosynthesis values found. The ratios RF/ESF and FRF/ESF were related to the photosynthesis in healthy and diseased plants. The monomolecular model was adjusted to the data obtained in this experiment (Table 2 and Figures 4 and 5). The adjustments were significant (p<0,05).

No study relating severity of the disease and photosynthesis was performed for the citrus - Xanthomonas axonopodis pv. citri pathos-system. Some obstacles can be mentioned concerning the non-concretization of the studies, among which the fact that citrus canker is considered a quarantinable disease, and the high costs involved in the acquisition and maintenance of the required equipment.

Contrary to field observations, which classified the Ponkan variety as more resistant to citrus canker than the Cravo Citrus limonia) variety (FEICHTENBERGER et al., 2005), we observed the inverse. In fact, the observations of this study were limited to the growth of the lesion around the wound, which is one of the components of varietal resistance. The inoculation method used per se allows the colonization resistance assessment only, disregarding any barriers imposed y the plant against bacterium infection. Neither the infection frequency nor the number/extension of the wounds were examined under natural conditions. Probably, the large quantity of thorns together with a median foliar density characteristic of the Cravo (Citrus limonia) variety have citrus plants of this species, under field conditions presenting a large number of lesions deriving from the friction of the leaves with the thorns, resulting in less resistance of the Cravo (Citrus limonia) variety when compared to the Ponkan variety.

The term virtual lesion comprises the visual lesion area plus the surrounding area in which the photosynthetic activity is also affected by the pathogen's activity. The concept of virtual lesion proposed by Bastiaans (1991) was the base for a model that quantifies the photosynthetic capacity of diseased leaves. The relationship between the disease severity and the photosynthesis of the infected leaves is described by one single parameter ß (BASSANEZI, 2000). According to the results obtained in experiments 1 and 2, the Ponkan variety presented a value for ß equal to 1 and the Hamlin variety, a value for ß greater than 1, indicating a decrease in the photosynthetic rate of those two varieties with an increase in severity of the disease. In the Hamlin variety the decrease of the photosynthetic rate was proportionately greater than the area affected by the disease, showing the existence of virtual lesion. In the Cravo (Citrus limonia) variety, the model did not adjust to the values obtained in the assessments, since for that variety, in both experiments, the severities did not reach expressive values, therefore the Citrus limonia variety had the least lesion expansion around the wound in relation to the other evaluated varieties.

At the severity levels studied, the spectroscopic measures (RF/ESF e FRF/ESF ratios) have a significant positive relationship with the photosynthesis value. The monomolecular value adjusted itself better to the data than the linear model and, apparently, it bears a relationship with the contents of chlorophyll present in the leaves, because the fluorescence emitted in the region of the red depends mainly of the concentration of chlorophyll in the leaf, because only it emits in this region (CEROVIC et al., 1999; LINS, 2005). The lowest photosynthesis values are related to the lowest values of RF/ESF and FRF/ESF, probably due to the lower content of chlorophyll in those samples, causing a reduction of the emitted fluorescence. As the contents of chlorophyll in the leaves increase, the photosynthesis also increases and the fluorescence spectroscopy system reads those increments showing higher values for RF/ESF and FRF/ESF. The photosynthesis reaches its maximum and stabilizes, but the spectroscopy system continues with increasing readings of RF/ESF and FRF/ESF. Therefore, the portable spectroscopy system is capable of performing readings that are directly related to the leaves photosynthesis, however this relationship is only truthful up to a certain point, from which it continues picking up the intensities of higher fluorescence (greater RF/ESF and FRF/ESF), although the leaves are with their constant photosynthetic activity. It is important to stress the large variation in fluorescence spectroscopy readings, possibly explained by the variation of the evaluated leaves age, different reading days and, which, contrary to the IRGA equipment that measures the photosynthesis of a 6 cm² area, the spectroscopy system makes timely readings of the leaf.

The relationship of the fluorescence spectroscopy reading with photosynthesis suggests that the diseases that affect the content of leaf chlorophyll cannot be made distinct based on the studied reasons, since they will present similar modifications. Actually, Marcassa et al. (2006) were unable to accurately differentiate leaves with citrus canker and citrus variegated chlorosis (CVC). However, the same authors were able to differentiate samples represented by leaves with or without citrus canker symptoms. Probably, due to the relationship that exists between the fluorescence spectroscopy, the way it was tested, and the photosynthesis, more promising results will be obtained when comparing plants with or without stress, and not plants under different types of stress.

BARBOSA, J. C.; GIMENES-FERNANDES, N., MASSARI, C. A., AYRES, A. J. Incidência e distribuição de cancro cítrico em pomares comerciais do Estado de São Paulo e sul do Triângulo Mineiro. Summa Phytopathologica, Botucatu, v. 27, p. 30-35, 2001.

BASSANEZI, R. B. Efeito de doenças foliares na eficiência fotossintética do feijoeiro (Phaseolus vulgaris L.) como contribuição na avaliação de danos. 2000. 233p. Tese (Doutorado em Fitopatologia) - Escola Superior de Agricultura "Luiz de Queiroz", Universidade de São Paulo, Piracicaba, 2000.

BASTIAANS, L. Ratio between virtual and visual lesion size as a measure to describe reduction in leaf photosynthesis of rice due to leaf blast. Phytopathology, St. Paul, v. 81, n. 6, p. 611-615, 1991.

BRUNINGS, A. M.; GABRIEL, D. W. Xanthomonas citri: breaking the surface. Molecular Plant Pathology, London, v.4, n. 3, p. 141-157, 2003.

CEROVIC, Z. G.; SAMSON, G.; MORALES, F; TREMBLAY, N.; MOYA, I. Ultraviolet-induced fluorescence for plant monitoring: present state and prospects. Agronomie, Paris, v. 19, p. 543-578, 1999.

FEICHTENBERGER, E.; BASSANEZI, R.B.; SPÓSITO, M. B.; BELASQUE JUNIOR, J. Doenças dos citros (Citrus spp.). In: KIMATI, H.; AMORIM, L.; REZENDE, J. A. M.; BERGAMIN FILHO, A.; CAMARGO, L. E. A. (Ed.). Manual de Fitopatologia: Doenças das plantas cultivadas. 4. ed. São Paulo: Ceres. 2005. v.2, p. 239-269.

LINS, E. C. C. C. Técnicas ópticas para análise da manifestação do estresse hídrico em laranjeiras. 2005. 102p. Dissertação (Mestrado em Engenharia Elétrica) - Universidade Federal de Pernambuco, Recife, 2005.

MARCASSA, L. G.; GASPAROTO, M. C. G.; BELASQUE JUNIOR, J.; LINS, E. C.; DIAS NUNES, F.; BAGNATO, V. S. Fluorescence spectroscopy applied to orange trees. Laser Physics, Moscow, v. 16, n.5, p. 884-888, 2006.

MASSARI, C. A.; BELASQUE JÚNIOR, J. A campanha de erradicação do cancro cítrico no Estado de São Paulo - situação atual e contaminação em viveiros. Laranja, Cordeirópolis, v.27, p.41-55, 2006.

RODRIGUES NETO, J.; BELASQUE JUNIOR, J.; AMORIM, L.; BERGAMIN FILHO, A. Larva-minadora aumenta a incidência do cancro cítrico. Visão Agrícola, Piracicaba, v.2, p. 14-19, jul./dez. 2004.

STALL, R.E.; SEYMOUR, C.P. Canker, a threat to citrus in the gulf-coast states. Plant Disease, St. Paul, v. 67, p. 581-585. 1983.

VALE, F. X. R.; FERNANDES FILHO, E. I.; LIBERATO, J. R. Quant. A software plant disease severity assessment. In: International Congress of Plant Pathology, 8, 2003, Christchurch. Proceedings... Christchurch: International Society of Plant Pathology, 2003. p. 105.

VAUTERIN, L., HOSTE, B., KERSTERS, K., SWINGS, J. Reclassification of Xanthomonas. International Journal of Systematic Bacteriology, Washington, v. 45, p.472-489, 1995.