DE19845883B4 - Method for determining the phytotoxicity of a test substance - Google Patents

Abstract

method for the determination of phytotoxicity a test substance, wherein in a bioassay at least one biological Sample of the test substance exposed and the phytotoxicity by Comparison of data of the sample and reference data of at least one neutral reference sample is determined, taking the data and the reference data recorded by a digital camera and by means of an image analysis system be compared, the image analysis system by measuring a unit body known body geometry and color is automatically calibrated and where the biological Measure samples before the start of the experiment using the image analysis system and inhomogeneous samples are sorted out.

Description

The The invention relates to a method for determining the phytotoxicity of a Test substance. Biotests are known. Thus, e.g. biological Short term tests for germination and further growth of plants especially the root and the shoot to the cotyledon stage carried out.

such Biotests, depending on the question, determine the vitality of seeds, but especially for the integral detection of phyto / plant toxic Effects of pure substances, mixtures of substances and environmental samples (water, Water concentrates, soil, sediment, soil and sediment eluates and extracts, solids).

she enable the integral assessment of being introduced into the environment or there occurring substances with regard to possible hazard potentials, e.g. by Pollutant input from the air and rain in forest areas. For potential pollutant mixtures with complex, chemically-analytically not comprehensively analyzable Composition - e.g. Sewage, landfill leachate streams, soil extracts especially from contaminated sites and their biological remediation products - combustion products (Exhaust gases) from the chemical industry, the energy producer and the Road traffic are Integral effect tests for ecotoxicological evaluation indispensable.

bioassays can especially with modern pesticides (high potency at low application rate) in addition and competition to often difficult chemical analysis.

The Growth of terrestrial or aquatic vascular plants is the vital foundation for any animal and human life on earth. Nevertheless, the Use of vascular plants to track down of pollutants so far strong against the use of unicellular Algae neglected. vascular plants however, e.g. across from au xinanalogen herbicides are much more sensitive than unicellular algae, because the latter is not over have an auxin controlled growth regulation. The assessment of air pollutants is only possible by the quantitative recording of higher plants.

Of the reason for the rarer use of vascular plants is among other things the higher one Variety of shapes and complexity of vascular plants, which previously a manual optical and therefore time and cost intensive and never completely objectifiable evaluation required.

Of the Plant germination root length test e.g. Uses the high sensitivity of plants to toxic Fabrics during the germination phase to the formation of the cotyledons.

• – DIN Entwurf 38412 "Bestimmung der Wirkung von Abwasser auf das Wurzellängenwachstum der Gartenkresse". Dieser sieht ein dreitägiges beleuchtetes Wachstum von Kressesamen auf Edelstahlnetzen in eine Verdünnungsreihe von Abwasser vor. Als Meßgrößen dient die manuell gemessene Wurzellänge. Außerdem sollen Erscheinungen wie die Ausbildung der Wurzelhaarzone und Entwicklungszustand von Hypokotyl und Blättern protokolliert werden.
• – Der Entwurf DIN 350 11 269 Teil 1 beschreibt die Prüfung des Wachstums vorgekeimter Gerstesamen in Bodenproben. Auch hier sollen nach 5 bis 7 Tagen und Entfernung der Erde die Länge der längsten Wurzel, gegebenenfalls auch die des Triebes gemessen werden.
• – Die EPA-Richtlinie 600-3-88-029 sieht ein fünftägiges „Protocols for short term toxicity screening of hazardous waste sites" Wachstum z.B. von Salatsamen im hell-/dunkel-Zyklus in Petrischalen in Gegenwart der Testlösung vor. Mögliche Meßgrößen sind Keimung, Wurzellänge und Trockengewicht.

Guidelines for the test execution can be found eg in:

• - DIN Draft 38412 "Determination of the effect of wastewater on the root length growth of garden cress". This provides for a three-day illuminated growth of cress seeds on stainless steel nets in a dilution series of wastewater. The measured variables are the manually measured root length. In addition, phenomena such as the formation of the root hair zone and state of development of hypocotyl and leaves are recorded.
• - The draft DIN 350 11 269 Part 1 describes the testing of the growth of pre-germinated barley seeds in soil samples. Again, after 5 to 7 days and distance of the earth, the length of the longest root, possibly also that of the instinct should be measured.
• - The EPA guideline 600-3-88-029 provides for a five-day "Protocol for short-term toxicity screening of hazardous waste sites" growth of, for example, lettuce seeds in light / dark cycle in petri dishes in the presence of the test solution , Root length and dry weight.

all Guidelines are common that in addition to the destructive Dry mass determination only the simplest measure of growth - the length of the Root is determined, since only this length measurement of the human Eye is easily possible.

Also known are biological short-term tests for growth inhibition / promotion of duckweed (Lemnaceae):
Rooting or free-floating vascular plants are an essential component of many aquatic ecosystems in addition to algae. Duckweed is a food source for waterfowl and many small animals. They serve fish as food, cover and shade. Small invertebrates use them as carriers. However, if they grow too fast, they can suppress the supply of light and oxygen to such an extent that the fish dies.

One Test is e.g. in the ASTM guideline E 1415-91 "Standard Guide for Conducting Static Toxicity Tests With Lemna gibba G3 ", a comparable European directive does not exist yet. The test takes advantage of the fact that the floating leaves of Duckweed with sufficient temperature control, lighting and nutrient supply share very quickly. However, under toxic load, it comes to a yellowing existing floating leaves and a reduction in the division speed.

To have to before starting the test by eye in all experimental approaches "equivalent" duckweed plants sorted out of the pre-breeding and used in containers with nutrient solution. This is fundamental because there are start differences during the test further enlarge and so that pollutant effects can cover. At the same time is the optically manual selection of plants with "equivalent" number of leaves and Size too with regard to the requirements of quality assurance and documentation burdened with a high subjective factor. After the end of the normally seven days Growth phase in Phytotron, a very evenly lit climate chamber, must be based a suitable measure the promotion or inhibiting the growth of treated trials versus untreated ones control batches be determined. The most frequent and simplest method is counting the individual floating leaves independent of their size. Further Measured variables are the number of roots formed, the dry weight and the total length of all Root. The dry weight is the most objective and reproducible Endpoint, but is only possible once by destroying the duckweed. Of the Chlorophyll content is very meaningful for the vitality of the plants, but also only under destruction directly measurable.

adversely in the test methods mentioned is the high workload, the poorly documented, as subjective influencing factors and / or the object-destroying measures.

In the broader context of the invention, the disclosure DE 39 06 215 A1 and based on this building WO 90/10273 A1 discloses a method and a device for automatic optical classification of potted plants by pixel classification of images of potted plants according to predetermined color classes and based on this classification, a method and an arrangement for controlling the growth process of Pflänzlingen by regulating the ambient temperature.

The WO 95/25955 A1 discloses a method and an apparatus for Determination of toxicity a water-soluble or water-miscible substance, the rate of movement of lower animals before and after addition of the substance to the aqueous Environment by means of an image acquisition unit with object recognition is observed and measured, and from this a relative toxicity of the substance is determined.

From the DE 38 36 716 A1 Furthermore, a method is known for the evaluation of cell images, wherein the cell images are first marked interactively and then statistical characteristics of the cell images from the number and position of the brands are determined.

The DE 42 11 904 A1 discloses a method and a device for determining the ecological status of a liquid sample, wherein objects of different types are automatically optically identified and counted on the basis of reference objects, and a list of the objects is created.

It is therefore an object of the present invention, while avoiding the known from the prior art disadvantages an easy to perform, objectifiable and object-preserving possibility to carry out and Evaluation of biotests to create.

These Task is solved by a method for determining the phytotoxicity of a Test substance, wherein in a bioassay at least one biological Sample of the test substance exposed and the phytotoxicity by Comparison of data of the sample and reference data of at least one neutral reference sample is determined, taking the data and the reference data recorded by a digital camera and by means of an image analysis system be compared, the image analysis system by measuring a unit body known body geometry and color is automatically calibrated and where the biological Measure samples before the start of the experiment using the image analysis system and inhomogeneous samples are sorted out.

advantageous Embodiments are the subject of the dependent claims.

So it is preferable if the device means for automatic Discrimination of sample components such as e.g. Seed coat, sprout, root, Root hair or surface covered by root hair, and for the subsequent quantitative measurement of the components e.g. Length, Width and shadow area based on the collected data by means of image analysis.

Further advantageous are means for the retrieval of individual samples, Sample components or sample groups at periodic measurements the same sample based on specific shape characteristics and information on the spatial change options between the measurements.

If arranged on / in the experimental units in the field of view of the camera barcodes are responsible for the safe allocation of each measurement be analyzed with image analysis and in addition to information on Test also information about the computer-generated placement of the shell at randomized Try included, there is another preferred embodiment in front.

It is also advantageous if a Einrich is provided for automatic calibration with a unit body of known body geometry and color by automatically measuring the image processing. As a result, the sample data can be converted into absolute values in a simple manner.

Especially It is advantageous if the one or more bar codes on the unit body for Calibration are arranged, as this reduces the expenditure on equipment becomes.

If this should be necessary, the image analysis without visible Can perform light a lighting device for evaluating in a dark test environment with a wavelength range, which does not influence the plants in their behavior and in the CCD camera works.

Especially This may be an infrared lighting device.

The in some series of tests required, separate by color and evaluation of the samples is done by a color camera or an optical Realized filter device.

Advantageous is further, means for comparing the data obtained with stored Comparative data are available. This can be done over the course of Time accumulated reference data material are used.

If the recording devices (CCD camera) and the evaluation (Image analysis system) spatially are separated, the evaluation can advantageously central carried out becomes. This does not have to at each test / workstation the full computing power held become.

The CCD camera and the sample cups should be movable relative to each other be arranged so that the CCD camera different sample areas approach, focus and record.

Prefers is further when the evaluation devices (image analysis system) based of color and texture patterns the test plants conclusions which can attract xenobiotics (environmental chemicals). Xenobiotics (environmental chemicals) cause different structural changes in the mentioned plants. The plants can e.g. in color, shape, size, division rate, Change infrared pattern or chlorophyll content. It can be more complete and partial Pigment loss occur. It comes to colony break-up, hunchback, for the loss or extreme formation of root hairs. duckweed can Form antocyans on the underside, or they form Turione. Alles these features can Conclusions on to leave existing xenobiotics. However, most features are According to the current state of the art to determine only by destruction of the organism. Thereby can only individual characteristics are examined, and the specific one Significance of the measurement is not given. Through the integrals Capture all the features, the evaluation can draw conclusions make the present xenobiotics. There are two different ones Strategies pursued. First, the evaluation recognizes predetermined Feature patterns, on the other hand, through a variety of tests new feature links found by the evaluation independently.

One Another advantageous feature of the method is that the evaluation (Image analysis system) the seedlings due to movement patterns, Find color and shape classes even after their growth can. In germination and growth test are a variety of the same plants observed in a test tube. By the enormous similarity it is not possible, individual Organisms to find and clearly identify them for a long time. The receiving unit may e.g. observe up to 100 seedlings simultaneously, and properly assign them based on the described feature patterns. Added to this is the possibility the recognition based on given movement patterns.

Objects that are below the resolution limit of the CCD camera (less than 1 mm) should preferably be reproduced true to the original. The normal resolution limit of CCD cameras with a resolution of 520 × 520 pixels is approximately 3 mm ² for a field of 100 × 100 mm (520 pixels and 3 colors respectively corresponds to 174 pixels per color) at least 3 x 3 pixels). Since the objects are known, the evaluation unit can apply found objects to known patterns, thus increasing the resolution. The objects are then no longer represented as a pixel matrix, but as a mathematical function. This makes it possible to increase the actually limited resolution of the camera and even objects that are smaller than 3 mm ² , to scale.

A further preferred development of the method is provided when the recording device (CCD camera) and the evaluation device can determine the metabolic rate of the plant by means of filter systems on the basis of infrared patterns. An important feature of the fitness of plants is their metabolic rate. Both the photosynthesis performance and the oxidation of starch and fats change the infrared pattern through reflection and the own emission of plants. These changes can be visualized by means of IR filtering through the CCD camera be made. The evaluation unit adjusts the filter systems independently, and can thus distinguish between photosynthetic metabolism and respiration. Until now, these feature patterns could not be detected nondestructively and over a long period of time, since the IR spectra could not be distinguished from the normal background noise. Thanks to special lighting and cooling, even the finest differences can be detected.

Finally is It is advantageous if specific optical filters for lighting and camera for measuring and quantifying the fluorescence of plants are provided. So far, such fluorescence signals are integrated over the entire image area evaluated. It is novel that such Fluorescence signals are processed with morphological filters. Furthermore, the image data with the data and shapes of the image objects superimposed in the visible and IR ranges and evaluated together. This application is particularly suitable for short-term tests in the hourly range.

The device used to carry out the method (see. 1 ) is characterized by full compatibility with the existing test guidelines by avoiding the above problems while creating new, more sensitive measures from. The automatic image analysis excludes subjective factors and drastically reduces the necessary personnel expenses. This makes multiple measurements easily possible and creates a valid, continuous documentability.

• – Erkennung von Nekrosen, Chlorosen und Blattmißbildungen;
• – Unterscheidung von toten und lebenden Blättern;
• – Bestimmung der Wachstumsgeschwindigkeit und Keimungsrate und andere kontinuierliche oder engmaschige Bestimmungen von Parametern (z.B. Teilungsrate, Flächenzunahme, Chlorophyllentwicklung, Wachstumsdynamik, Absterbedynamik), die von Hand wegen des hohen Zeitaufwands kaum (wirtschaftlich) möglich sind;
• – Bestimmung der Blattanzahl;
• – Bestimmung der Frondzahl;
• – Bestimmung von Gesamt- und Einzelfrondfläche;
• – Bestimmung von Gesamt- und Einzelgrünheit;
• – Bestimmung des Chlorophyllgehaltes über die Grünheit der Blätter
• – Bestimmung des Trockengewichts über Volumenanalyse und Flächenanalyse;
• – Bestimmung von Effektkonzentrationen (EC_x);
• – Bestimmung von Blattgrößen, Blattgrößenverteilungen und Blattformen;
• – Gesamterfassung einer höheren Pflanze;
• – automatischer Vergleich zu unbelasteten sogenannten Nullkontrollen, sowohl aus parallelen Versuchsansätzen als auch aus gespeicherten Datensätzen;
• – Erfassung von IR-Mustern und Überlagerung mit sichtbaren morphologischen Strukturen;
• – Archivierung der bildanalytischen Daten z.B. für eventuelle spätere Zweitgutachten.

The actual evaluation is limited only by the software used. For example, it is possible to carry out the following investigations with the data obtained:

• - detection of necrosis, chlorosis and leaf malformations;
• - distinction of dead and living leaves;
• - Determination of the growth rate and germination rate and other continuous or close-meshed determinations of parameters (eg, division rate, area increase, chlorophyll development, growth dynamics, Absterbedynamik), which are hardly possible (economically) by hand because of the high expenditure of time;
• - Determination of the number of sheets;
• - determination of frond number;
• - determination of total and single frontal area;
• - determination of total and single greenness;
• - Determination of chlorophyll content on the greenness of the leaves
• - determination of dry weight via volume analysis and area analysis;
• - Determination of effect concentrations (EC _x );
• - determination of sheet sizes, sheet size distributions and leaf shapes;
• - total survey of a higher plant;
• - automatic comparison to unloaded so-called zero checks, both from parallel test approaches and from stored data sets;
• - acquisition of IR patterns and overlay with visible morphological structures;
• - Archiving of the image-analytical data eg for possible later second appraisals.

there is it possible in particular the data obtained by remote data transmission to a central To transmit the analysis device there, e.g. evaluate by comparison with large stocks of standard data and to retransmit the evaluation.

The in the 1 and 2 shown devices may be used in the practice of the method.

In 1 a device is shown in which the sample located in a sample cup is placed on a stage. Depending on the question, there is either no illumination or illumination from the illumination devices arranged next to the camera as incident light or from the illumination device arranged below the object table as transmitted light. The camera, for example a CCD camera, records the images, which are then processed by image analysis by means of an evaluation unit, preferably in the form of a PC. The data is displayed on a monitor.

In 2 is shown as an alternative, as the camera is arranged above a climate chamber with several samples movable. The camera can thus approach the individual samples at predetermined time intervals and make appropriate recordings as described above.

below Two application examples for the method will be explained.

Application example root length test

The Computer-assisted image analysis according to the invention detected each seedling individually. With image analytical methods are then Seed coat, sprout and root, If necessary, also root hair separated and then quantified separately. Possible measured variables are e.g. Seed swelling, germination (root after a certain incubation period longer 1-4 mm), root length, Root shadow area, mean root diameter, root length to root ratio Diameter, area covered by the root hairs, shoot length, shoot area and leaf area.

This allows it, without additional Work effort also previously poorly detectable effects such as a change the root thickness or leaf area under the influence of pollutants too detect.

The only through the automation practicable periodic survey all samples in Petri dish tests during the germination and growth phase allows the determination of temporal growth parameters such as germination time and individual growth curves. This also allows automatic Separation of non-germinated seeds and those in the germination or first growth phase have died. For individual retrieval The seedlings that move on the ground will be at each Measuring the spatial Location of specific seedling characteristics of the individual seedlings image-analytically determined and stored with. These features allow at the next Measuring the individual assignment of seedlings.

The Evaluation of the experiment takes place when tested in Petri dishes either by removing the shells from the incubation room (if necessary Phytotron) into the measuring device or by installing the camera in a horizontally movable carriage in the incubation room. This allows fully automatic multiple measurements and the measurement and test evaluation on non-working days.

at Other testing requirements include plants from the growth matrix manually freed and then measured by image analysis.

to Planarization can be a pane of glass on the seedlings / plants be placed.

All Raw data can directly evaluated statistically.

Around To exclude confusion can be at / in each vessel / trial batch In the field of view of the camera a barcode can be attached, which in the image analysis each is read in. It contains information about Test also information about the computer-generated placement of the shell at randomized To attempt.

The Lighting is dependent executed by the plant to be examined as transmitted light or reflected light. One special case represents the infra-red illumination. With infrared illumination can they Control seedlings without affecting their growth, because infrared light has no effect on the photosynthesis of plants Has. On the other hand, the CCD camera is sensitive to infrared light.

For an automatic Calibration of the measured variable can a defined body be introduced into the shell. The pixel values and their color values, the can be recorded by the image processing, so converted into absolute measures become. this makes possible a comparison with manual evaluations.

example Duckweed test

The computerized Image analysis detected each leaf of duckweed in the test tube individually determines the area and allows over it the number out size distributions and total area display. This also allows so far only bad detectable To recognize effects, such as a reduction of the floating leaves under pollutant influence.

The Evaluation of the experiment is carried out either by removing the Shells from the phytotron and placed in the measuring device or by installing the camera on a mobile carriage in the phytotron itself or the sample. This allows fully automatic Multiple measurements and the measurement and the test evaluation days off.

The Lighting for the measurement is done from the top with additional lighting around the camera, where appropriate, the light of the phytotrons is shielded. For a better distinction of green, yellow and "white" leaflets a color camera or a grayscale camera with moving filters used and possibly also worked with transmitted light. The Determination of the sum of all floating leaves takes place after switching off the additional lighting without filter. For direct measurements in the phytotron the auxiliary lighting is switched off, so that a quasi-indirect lighting / transmitted light is present.

to Documentation of the respective starting conditions in each test vessel before the start of the experiment all duodenal lens groups to be used with the Measure image analysis and groups that have tight quality requirements not to homogeneity comply with, sorted out. This reduces possible standard deviations significantly and creates a test homogeneity that matches the human Eye alone would not be achievable.

During the If necessary, you can try it at defined intervals without big ones Effort measurements performed and temporal shifts in growth dynamics documented become.

For the final evaluation, the statistical evaluation of the raw data and the marking by means of barcodes, the above statements apply analogous.

Claims (16)

Method for determining the phytotoxicity of a Test substance, wherein in a bioassay at least one biological Sample of the test substance exposed and the phytotoxicity by Comparison of data of the sample and reference data of at least one neutral Reference sample is determined, taking the data and the reference data recorded by a digital camera and by means of an image analysis system be compared, the image analysis system by measuring a Unit body known body geometry and color is automatically calibrated and where the biological Measure samples before the start of the experiment using the image analysis system and inhomogeneous samples are sorted out. The method of claim 1, wherein the inclusion of the Data and reference data, their forwarding to the image analysis system and the comparison of the data is done automatically. Method according to claim 1 or 2, wherein the data and reference data in the image analysis system can be reduced. Method according to one of claims 1 to 3, wherein means of the image analysis system with repeated recording of data of the same Sample an individual component of this sample again recognized and its spatial change is described. Method according to one of claims 1 to 4, wherein the samples and reference samples are automatically placed. Method according to one of claims 1 to 5, wherein the bioassay performed on foliage plants becomes. The method of claim 6, wherein by means of the image analysis system Components of leafy plants are distinguished and quantitatively measured become. Method according to one of claims 1 to 5, wherein the bioassay performed on duckweed becomes. Method according to one of claims 1 to 8, wherein as illumination Infrared light is used. Method according to one of claims 1 to 9, wherein a planarizing element, by means of which the samples and the reference samples can be planarized are used. Method according to one of claims 1 to 10, wherein means of the image analysis system on the samples and reference samples and these identifying barcodes are evaluated. Method according to one of claims 1 to 11, wherein a lighting device for evaluation in a dark test environment with a wavelength range, which does not affect the behavior of the samples and reference samples the digital camera works, so the image analysis without synonymous visible light performed can be used. Method according to one of claims 1 to 12, wherein a color camera or an optical filter device by means of which the samples and Reference samples can be recorded and evaluated separately according to color is used. Method according to one of claims 1 to 13, wherein a memory element for storing comparison data, wherein the data of the samples and comparable to the comparative data is used. Method according to one of claims 1 to 14, wherein the digital Camera and the samples and reference samples movable relative to each other be arranged so that the digital camera different Approach, focus and record sample areas. Method according to one of claims 1 to 15, wherein specific optical filters for Lighting and camera for measuring and quantifying fluorescence used by plants.