R-Biopharm Rhône Mycotoxin Technical Manual

Transcription

1 R-Biopharm Rhône Mycotoxin Technical Manual GB02/V2

2 R-Biopharm Rhône Mycotoxin Technical Manual 2

3 No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, without prior permission of the publisher (R-Biopharm Rhône Ltd, Block 10, Todd Campus, West of Scotland Science Park, Acre Road, Glasgow, Scotland, G20 0XA). Limit of Liability / Disclaimer of Warranty: While the publisher and author have used their best efforts in preparing this manual, they make no representations or warranties with respect to the accuracy or completeness of the contents and specifically disclaim any implied warranties of merchantability or fitness for a particular purpose. No warranty may be created or extended by sales representatives or written sales materials. The advice and strategies contained herein may not be suitable for your situation. You should consult with a professional where appropriate. Neither the publisher nor author shall be liable for any loss of profit or any other commercial damages, including but not limited to special, incidental, consequential, or other damages. For general information on our products and services or for technical support, please contact: R-Biopharm Rhône Ltd Block 10 Todd Campus West of Scotland Science Park Acre Road, Glasgow Scotland G20 0XA Tel: +44 (0) Fax: +44 (0) info@r-biopharmrhone.com 3

4 Contents 1. Introduction to Mycotoxins 5 2. Testing for Mycotoxins Legislation of Mycotoxins Sampling Detection of Mycotoxins Principle of Immunoaffinity Columns Principle of Derivatisation Derivatisation of Aflatoxins Products Available from R-Biopharm Rhône 9 3. Introduction to Terminology Used in Mycotoxin Detection Abbreviations Used in Mycotoxin Detection Mycotoxin Levels Weights Volumes Dilutions Handling Mycotoxins Hazards Decontamination Calibration Curves An Example Calibration Curve for Aflatoxin Detection An Example Calibration Curve for Deoxynivalenol Detection An Example Calibration Curve for Fumonisin Detection An Example Calibration Curve for Ochratoxin Detection An Example Calibration Curve for T-2 & HT-2 Detection An Example Calibration Curve for Zearalenone Detection Reporting Results Determining the Concentration of Toxin Present AFLAPREP An Example DONPREP An Example EASI-EXTRACT AFLATOXIN An Example EASI-EXTRACT T-2 & HT-2 An Example EASI-EXTRACT ZEARALENONE An Example FUMONIPREP An Example OCHRAPREP An Example Validating a Method Precision Repeatability (r) Reproducibility (R) Limit of Detection Limit of Quantification Determination of LOD and LOQ of an HPLC System Accuracy How to Spike a Sample General Guidelines on Improving Recoveries General Guidelines on Improving Chromatograms High Baseline Noise Mobile Phase Contamination Contamination Downstream from the Pump Poor Reproducibility for Peak Areas Peak Broadening and Peak Splitting Ghost Peaks HPLC Troubleshooting Introduction When to Adjust the Mobile Phase Solvents and Mobile Phase Pump Column and Column Heater Detector Removing Bubbles from Detector Cell Cleaning the Detector Flow Cell Without Disassembly Prevention of Air Bubbles Maintenance of Fluorescence Detectors Glossary 38 R-Biopharm Rhône Mycotoxin Technical Manual 4

5 1 Introduction to Mycotoxins Mycotoxins are toxic chemical compounds that are produced by certain fungi. They are produced under specific conditions of moisture and temperature and are generally associated with diseased or mouldy crops. Not all fungi can produce mycotoxins. Even those with the ability to produce mycotoxins may not produce them all of the time. Growth of the mycotoxin depends on temperature, ph, humidity and the presence of plant substrates. There are many mycotoxins, however only a few of them are regularly found in food and animal feedstuffs. Most mycotoxins are chemically stable so they tend to survive storage and processing even at extreme temperatures, such as freezing. Nevertheless, those that do occur in food have great impact on the health of humans and can cause significant economic losses in terms of plants and livestock. Mycotoxins are responsible for a diverse range of toxic effects because their chemical structures are very different from each other. Acute effects require that high amounts of toxin are present in the food being consumed and such incidents are usually restricted to the less developed parts of the world where resources for control are limited. Chronic effects are caused by the accumulation of a low level of toxin in the body over a long period of time and can affect the long-term health of the population. Some of the most common mycotoxins are carcinogenic, genotoxic or may target the kidney, liver or immune system. 5

6 2 Testing for Mycotoxins 2.1 Legislation of Mycotoxins Legislation for mycotoxins is constantly changing with more emphasis being placed on sampling, method criteria and precision of results. The number of commodities and mycotoxins that are now covered under EU legislation is also increasing. R-Biopharm Rhône has summarised all legislation in an easy to follow booklet and poster to help you understand current legislation, please contact us for a copy or for further information. 2.2 Sampling Mycotoxins are heterogeneously distributed in naturally contaminated commodities. The primary goal of sampling is to obtain a lab sample that accurately represents the lot from which it was taken. In order to obtain a sample that is representative it is recommended to take a higher number of samples of lower weight as the following pictures illustrate. There are several possible sources of variability when sampling a commodity for mycotoxins: Sampling and sub sampling Sample Preparation Analysis For further information on sampling plans see There is also a Sampling Advice booklet for mycotoxins available from the Food Standards Agency in the UK, which provides a simple guide to European Sampling of cereals, dried fruit, nuts, spices, coffee, fruit juices and wines and offers advice on how to record and interpret results. This booklet is available from the Food Standards Agency or from R-Biopharm Rhône on request. Low Number of Samples = High Sample Variance High Number of Samples = Low Sample Variance Sampling photographs supplied by: Kim Esbensen Applied Chemometrics, Analytical Chemistry, Applied Biotechnology, Bioenergy & Sampling Research Group, University of Aalborg Esbjerg, Denmark R-Biopharm Rhône Mycotoxin Technical Manual 6

7 2 Testing for Mycotoxins 2.3 Detection of Mycotoxins Mycotoxins occur in small quantities in foodstuffs therefore their identification and quantitative assessment requires sophisticated sampling, preparation, extraction and analytical techniques such as - High pressure liquid chromatography (HPLC) Immunoassays (ELISAs) Lateral flow (LF) or dipsticks Membrane cards Mass spectroscopy (MS) Thin layer chromatography (TLC) Gas chromatography (GC) Liquid chromatography with tandem mass spectrometry (LC-MS/MS) 2.4 Principle of Immunoaffinity Columns RBR mycotoxin immunoaffinity columns contain a gel suspension of monoclonal antibodies specific to the toxin of interest. The use of a monoclonal antibody makes the test highly specific for a target mycotoxin and offers improved sensitivity. Immunoaffinity columns have the added advantage in that they can be automated for largescale analysis of samples. Following extraction of the toxin the sample extract is filtered, diluted and passed through the immunoaffinity column. Any toxin that is present in the sample is retained by the antibody within the gel suspension. The column is washed to remove unbound material and the toxin is then released by the antibody following elution with solvent. The eluate can then be analysed by HPLC or LC-MS/MS. The total extraction and clean-up procedure takes approximately 20 minutes. The result is improved clean-up and concentration of the toxin from food and feed samples giving a much cleaner chromatogram, therefore providing more accurate and sensitive detection. R-Biopharm Rhône products are developed, manufactured and dispatched under ISO 9001 and ISO registered quality systems, guaranteeing products of consistent quality that always meet our performance criteria. Sample Washing Elution The sample extract (containing the toxin) is passed through the column. Mycotoxins The antibodies isolate and concentrate the toxin and retains it in the column. Other Material Passage of solvent through the column denatures the antibody and releases the toxin. 7

8 2 Testing for Mycotoxins 2.5 Principle of Derivatisation Some mycotoxins do not fluoresce naturally therefore they require to be derivatised prior to injection onto the HPLC system. Derivatisation is the technique that transforms a chemical compound into a product of similar chemical structure for enhanced detection. Generally, a specific functional group of the compound is targeted during derivatisation. The resulting new chemical compound is generally easier to detect and can be used for quantification. membrane. These layers are sandwiched between rigid plastic housings. The KOBRA CELL is fitted between the HPLC column and the detector and generates the derivatisation agent, bromine, online from potassium bromide and nitric acid which are present in the mobile phase. The derivatisation results in a significant increase in the fluorescent signals of the modified forms of aflatoxin B1 and G1 allowing improved detection using a fluorescence detector. 2.6 Derivatisation of Aflatoxins When testing for aflatoxins, only B2 and G2 fluoresce brightly under UV light and therefore can easily be detected by HPLC with a fluorescence detector. Aflatoxins B1 and G1 do not fluoresce to such a high degree naturally and therefore must be derivatised. The KOBRA CELL is a unique system, which offers a popular derivatisation method when testing for aflatoxins. The KOBRA CELL is an electrochemical cell consisting of a platinum working electrode and a stainless steel auxiliary electrode separated from each other by a Derivatisation without KOBRA CELL Derivatisation with KOBRA CELL R-Biopharm Rhône Mycotoxin Technical Manual 8

9 2 Testing for Mycotoxins 2.7 Products Available from R-Biopharm Rhône Immunoaffinity, solid phase and molecularly imprinted polymer columns are available for the quantitative analysis of the following mycotoxins both individually or in combination when used in conjunction with HPLC or LC-MS/MS - Aflatoxin B1, B2, G1, G2 and M1 Deoxynivalenol Fumonisin B1 and B2 Ochratoxin A Patulin T-2 & HT-2 Zearalenone Additional products for analysis of mycotoxins: RBR also provide a number of screening tests such as cards and ELISA systems for the detection of a range of mycotoxins. Standards and reference materials: RBR manufacture and supply a range of mycotoxin standards and reference materials for use with HPLC. Please contact your local R-Biopharm distributor for further information. KOBRA CELL for derivatisation of aflatoxins B1 and G1: The only electrochemical cell recommended in EC standard methods and used by key institutions, food companies and government laboratories worldwide. 9

10 3 Introduction to Terminology Used in Mycotoxin Detection 3.1 Abbreviations Used in Mycotoxin Detection µ micro IAC immunoaffinity column n nano dh 2 O distilled / deionised water m milli MeOH methanol L litre ACN acetonitrile g gram PBS phosphate buffered saline ppb parts per billion AA acetic acid ppm parts per million r repeatability ppt parts per trillion R reproducibility 3.2 Mycotoxin Levels When talking about mycotoxins, parts-per notation is routinely used, which helps to describe low value proportions in measured quantities. Common parts-per notation that are used in science are parts per million (ppm), parts per billion (ppb) and parts per trillion (ppt). Parts per million: 1 part per 1,000,000 parts or one part in 10 6 g / ml = mg / µl µg / ml = mg / L µg / g = µg / ml = ng / mg 1 g in 1,000 g or 1 kg = 1,000 ppm For example, if we have 1 mg / g, this is 1000 x greater than µg / g; therefore it is equal to 1000 ppm. Parts per billion: To convert parts per million and parts per billion: 1 ppm = 1,000 ppb = 1,000,000 ppt ppm = 1 ppb = 1000 ppt ppm = ppb = 1 ppt 1 ppm = 1,000,000 ppt Summary: ppm --> ppb --> ppt (each differing by a factor of 1,000) 1 part per 1,000,000,000 parts or one part in 10 9 µg / L = ng / ml ng / g = µg / kg 1 µg in 1,000 g or 1 kg = 1 ppb R-Biopharm Rhône Mycotoxin Technical Manual 10

11 3 Introduction to Terminology Used in Mycotoxin Detection 3.3 Weights The gram is a unit of mass and is the most commonly used unit of measurement for solid material. To convert grams and milligrams: 1 g = 1,000 mg (g x 1,000 = mg) 1 mg = g (mg / 1,000 = g) To convert grams and micrograms: 1 g = 1,000,000 µg (g x 1,000,000 = µg) 1 g = g (g / 1,000,000 = g) To convert milligrams to nanograms: 1 mg = 1,000,000 ng (mg x 1,000,000 = ng) 1 ng = mg (ng / 1,000,000 = mg) To convert micrograms to nanograms: 1 g = 1,000 ng (g x 1,000 = ng) 1 ng = µg (ng / 1,000 = µg) To convert grams and nanograms: 1 g = 1,000,000,000 ng (g x 1,000,000,000 = ng) 1 ng = g (ng / 1,000,000,000 =g) Summary: g --> mg --> µg --> ng (each differing by a factor of 1,000) To convert milligrams to micrograms: 1 mg = 1,000 µg (mg x 1,000 = µg) 1 g = mg (g / 1,000 = mg) 3.4 Volumes The litre is a unit of volume and is the most commonly used unit of measurement for liquid material. To convert litres and millilitres: 1 L = 1,000 ml (L x 1,000 = ml) 1 ml = L (ml / 1,000 = L) To convert litres and micro litres: 1 L = 1,000,000 µl (L x 1,000,000 = µl) 1 µl = L (µl / 1,000,000 = L) To convert millilitres to micro litres: 1 ml = 1,000 µl (ml x 1,000 = µl) 1 µl = ml (µl / 1,000 = ml) Summary: L --> ml --> µl (each differing by a factor of 1,000) 11

12 3 Introduction to Terminology Used in Mycotoxin Detection 3.5 Dilutions The following are examples of percentage solutions used in mycotoxin testing: 1 % Sodium Bicarbonate Soution 1 g of sodium bicarbonate in 100 ml water. 2 % Sodium Bicarbonate Solution 2 g of sodium bicarbonate in 100 ml water. 60 % Methanol 60 ml of methanol plus 40 ml of water. 80 % Methanol 80 ml of methanol plus 20 ml of water. Methanol : Acetic Acid (98 : 2 (v/v)) 98 ml of methanol plus 2 ml of acetic acid. 1 In 2 Dilution Or Doubling Dilution Or Serial Dilutions A serial dilution can be used to accurately create diluted solutions and is routinely used in producing calibration curves. For example: Vial 1: Measure 2 ml of 100 % methanol. Vial 2: Take 1 ml from vial 1 and add 1 ml of water (equivalent to 50 % methanol). Vial 3: Take 1 ml from vial 2 and add 1 ml of water (equivalent to 25 % methanol). Vial 4: Take 1 ml from vial 3 and add 1 ml of water (equivalent to 12.5 % methanol). R-Biopharm Rhône Mycotoxin Technical Manual 12

13 4 Handling Mycotoxins 4.1 Hazards Mycotoxins are very hazardous substances. Only laboratories equipped to handle toxic materials and solvents should perform analyses. Suitable protective clothing, including gloves, safety glasses and lab coats should be worn throughout the analysis. Flammable solvents should be stored in an explosion-proof cabinet. Use a chemical hood and protective equipment as applicable. 4.2 Decontamination Prior to disposal, excess standard solutions should be treated with at least one-tenth their volume of 5 % sodium hypochlorite. Labware and contaminated waste should be immersed in 5 % sodium hypochlorite solution for 30 minutes followed by the addition of 5 % acetone for 30 minutes. Flush with copious amounts of water before disposal. After decontamination labware should be thoroughly washed. Incinerate waste if regulations permit. The columns contain 0.01 % (w/v) thimerosal. Skin or eye splashes should be washed immediately with quantities of water. Contact your local R-Biopharm distributor for a Material Safety Data Sheet for further information if required. 13

14 5 Calibration Curves The sample components are eluted as Guassian shaped peaks in the chromatogram. The retention times provide the qualitative aspect of the chromatogram. The retention time of a compound should be the same under identical chromatographic conditions. The peak height, or peak area, is related to the quantity of analyte. For determination of the actual amount of analyte, the area or height is compared against standards of a known concentration. 100 % will produce a series of readings. For most analyses a plot of response versus concentration will create a linear relationship. The response of the unknown can then be measured using the calibration curve. The calibration curve when plotted should be linear, and deviations from this straight line give an indication about the precision of the result. For example, a correlation of >98 % will indicate that all standards produced create a good straight line and with good certainty in the result. With a correlation of <98 %, there will be less certainty about the results. Peak Height W5 % 5 % LW RW A calibration curve is used to determine the concentration of a substance in an unknown sample by comparing the unknown to a set of standard samples of known concentration. The calibration curve is a plot of how the instrument responds, the so-called analytical signal, changes with changing concentration of analyte. A series of standards across a range of concentrations will be prepared, preferably with serial dilutions, with the middle standard usually matching the concentration of interest. Analysing each of the standards using the chosen technique Certain criteria must be met in developing a calibration curve: It is recommended to run at least a 3-5 point calibration curve. In constructing a calibration curve the levels of the calibration standards should bracket or include the range of expected results / measurements. For example, if the sample is 20 ppb the curve should be 5, 10, 20, 40 and 80 ppb. R-Biopharm Rhône Mycotoxin Technical Manual 14

15 5 Calibration Curves 5.1 An Example Calibration Curve for Aflatoxin Detection Take 5 ml methanol and place in a 5 ml amber vial. Remove 400 µl. Add 400 µl of 1,000 ng/ml total aflatoxin standard to give an 80 ng/ml solution. HPLC Standard 4 Take 2.5 ml of the 80 ng/ml solution and put in a 5 ml amber vial. Add 2.5 ml water to give a 40 ng/ml standard. This is equivalent to 1 ng per aflatoxin (4 ng total) per 100 µl injection. HPLC Standard 3 Take 2.5 ml of standard 4 and put in a 5 ml amber vial. Add 2.5 ml 50 % methanol to give a 20 ng/ml standard. This is equivalent to 0.5 ng per aflatoxin (2 ng total) per 100 µl injection. HPLC Standard 2 Take 2.5 ml of standard 3 and put in a 5 ml amber vial. Add 2.5 ml 50 % methanol to give a 10 ng/ml standard. This is equivalent to 0.25 ng per aflatoxin (1 ng total) per 100 µl injection. HPLC Standard 1 Take 2.5 ml of standard 2 and put in a 5 ml amber vial. Add 2.5 ml 50 % methanol to give a 5 ng/ml standard. This is equivalent to ng per aflatoxin (0.5 ng total) per 100 µl injection. 15

16 5 Calibration Curves 5.2 An Example Calibration Curve for Deoxynivalenol Detection Diluent solution 15 % methanol. HPLC Standard 5 Take 5 ml of diluent solution and place in a 5 ml amber vial. Remove 400 µl. Add 400 µl of 50 µg/ml deoxynivalenol solution to give a 4 µg/ml (4 ppm = 4,000 ng/ml) standard. This is equivalent to 400 ng per 100 µl injection. HPLC Standard 4 Take 2 ml of standard 5 solution and put in a 5 ml amber vial. Add 2 ml of diluent solution to give a 2 µg/ml (2 ppm = 2,000 ng/ml) standard. This is equivalent to 200 ng per 100 µl injection. HPLC Standard 2 Take 2 ml of standard 3 and put in a 5 ml amber vial. Add 2 ml of diluent solution to give a 0.5 µg/ml (0.5 ppm = 500 ng/ml) standard. This is equivalent to 50 ng per 100 µl injection. HPLC Standard 1 Take 2 ml of standard 2 and put in a 5 ml amber vial. Add 2 ml of diluent solution to give a 0.25 µg/ml (0.25 ppm = 250 ng/ml) standard. This is equivalent to 25 ng per 100 µl injection. HPLC Standard 3 Take 2 ml of standard 4 and put in a 5 ml amber vial. Add 2 ml of diluent solution to give a 1 µg/ml (1 ppm = 1,000 ng/ml) standard. This is equivalent to 100 ng per 100 µl injection. R-Biopharm Rhône Mycotoxin Technical Manual 16

17 5 Calibration Curves 5.3 An Example Calibration Curve for Fumonisin Detection Take 7.5 ml of acetonitrile : methanol : water (25 : 25 : 50 (v/v/v)) and place in a 10 ml amber vial. Remove 200 µl. Add 200 µl of 150,000 ng/ml fumonisin standard to give a 4,000 ng/ml solution. HPLC Standard 3 Take 500 µl of the 4000 ng/ml solution and put in a 5 ml amber vial. Add 1.5 ml of 50 % methanol to give a 1,000 ng/ml standard. This is equivalent to 100 ng per 100 µl injection. HPLC Standard 2 Take 1 ml of standard 3 and put in a 5 ml amber vial. Add 1 ml of 50 % methanol to give a 500 ng/ml standard. This is equivalent to 50 ng per 100 µl injection. HPLC Standard 1 Take 1 ml of standard 2 and put in a 5 ml amber vial. Add 1 ml of 50 % methanol to give a 250 ng/ml standard. 5.4 An Example Calibration Curve for Ochratoxin Detection Take 5 ml of methanol and place in a 5 ml amber vial. Remove 500 µl. Add 500 µl of 1,000 ng/ml ochratoxin standard to give a 100 ng/ml solution. HPLC Standard 4 Take 100 µl of the 100 ng/ml solution and put in a 5 ml amber vial. Add 1.4 ml of acetic acid : methanol (2 : 98 (v/v)). Add 1.5 ml of water to give a 3.33 ng/ml standard. This is equivalent to 0.33 ng per 100 µl injection. HPLC Standard 3 Take 1.5 ml of standard 4 and put in a 5 ml amber vial. Add 750 µl of acetic acid : methanol (2 : 98 (v/v)). Add 750 µl of water to give a 1.67 ng/ml standard. This is equivalent to ng per 100 µl injection. HPLC Standard 2 Take 1.5 ml of standard 3 and put in a 5 ml amber vial. Add 750 µl of acetic acid : methanol (2 : 98 (v/v)). Add 750 µl of water to give a ng/ml standard. This is equivalent to ng per 100 µl injection. HPLC Standard 1 Take 1.5 ml of standard 2 and put in a 5 ml amber vial. Add 750 µl of acetic acid : methanol (2 : 98 (v/v)). Add 750 µl of water to give a ng/ml standard. This is equivalent to ng per 100 µl injection. 17

18 5 Calibration Curves 5.5 An Example Calibration Curve for T-2 & HT-2 Detection HPLC Standard 3 30 µl of 10 µg/ml T-2 & HT-2 standard should be added to the glass tube. Blow down to dryness, follow derivatisation as per IFU. Reconstitute in 2 ml of mobile phase in order to obtain 150 ng/ml solution. This is equivalent to 15 ng/ml per 100 µl injection. HPLC Standard 1 Take 1ml of standard 2 and put in a 5 ml amber vial. Add 1ml of mobile phase to give a 37.5 ng/m l standard. This is equivalent to 3.75 ng per 100 µl injection. HPLC Standard 2 Take 1 ml of standard 3 and put in a 5 ml amber vial. Add 1 ml of 70 % acetonitrile to give a 75 ng/ml standard. This is equivalent to 7.5 ng per 100 µl injection. 5.6 An Example Calibration Curve for Zearalenone Detection Take 3 ml of acetonitrile and place in a 5 ml amber vial. Remove 1.8 ml. Add 1.8 ml of 1,000 ng/ml zearalenone standard to give a 600 ng/ml solution. HPLC Standard 4 Take 2 ml of the 600 ng/ml solution and put in a 5 ml amber vial. Add 2 ml of water to give a 300 ng/ml standard. This is equivalent to 30 ng per 100 µl injection. HPLC Standard 3 Take 2 ml of standard 4 and put in a 5 ml amber vial. Add 2 ml of 50 % acetonitrile to give a 150 ng/ml standard. This is equivalent to 15 ng per 100 µl injection. HPLC Standard 2 Take 2 ml of standard 3 and put in a 5 ml amber vial. Add 2 ml of 50 % acetonitrile to give a 75 ng/ml standard. This is equivalent to 7.5 ng per 100 µl injection. HPLC Standard 1 Take 2 ml of standard 2 and put in a 5 ml amber vial. Add 2 ml of 50 % acetonitrile to give a 37.5 ng/ml standard. This is equivalent to 3.75 ng per 100 µl injection. R-Biopharm Rhône Mycotoxin Technical Manual 18

19 6 Reporting Results All results should be expressed in parts-per notation and in order to accurately report results the analyst should account for any losses during the method. It is recommended that a reference sample of the same commodity type as the material being tested is run through the complete immunoaffinity column procedure. The recoveries obtained with the spiked sample can then be used to correct the results obtained with the test sample. For example, known reference sample is at 10 ppb, results obtained from HPLC were 8 ppb. This is equivalent to 80 % recovery. For all future samples, similar to the reference standard you can add 20 % onto the result in order to account for losses, i.e. if you obtain a value of 2 ppb and you know this is only 80 % recovery, adding 20 % gives a value of 2.5 ppb. Current legislation states that for results to be accepted recoveries should fall within the following range: Commodity Total Contamination Level Acceptable Range Of Recoveries All Foods <1.0 ppb % 1-10 ppb % >10 ppb % RBR products are developed, manufactured and dispatched under an ISO 9001 and ISO registered quality management systems, guaranteeing products of consistent quality that meet CEN specifications of % recovery. 19

20 7 Determining the Concentration of Toxin Present When reporting results, they should be expressed in parts per notation. However, with most HPLC systems the value given will be expressed in nanograms per injection. Therefore, you must convert nanograms to the required parts-per notation. In order to do this you must know the gram equivalent of sample that was injected onto the HPLC system. From there you can calculate a multiplication factor to convert nanograms per injection to parts-per notation. In the example below 0.05 g of sample was injected onto the HPLC system, which produced a reading of ng for aflatoxin G2. In order to present this value as ppb (i.e. nanograms per gram of sample) it is necessary to multiply the value obtained for G2 (i.e. nanograms per 0.05 g of sample injected) by a factor of 20 in order to calculate how many nanograms would be present in 1 g of sample. In this case, you would report the contamination of the unknown sample as x 20 = ppb aflatoxin G2. Example printout from HPLC system: Sample No. Sample Name Ret. Time Min G2 Fluorescence 1 STD 1 5 ng / ml Area ml / *min G2 Fluorescence Height mv Fluorescence G2 Amount ng G2 Fluorescence STD 2 10 ng / ml 3 STD 3 20 ng / ml 4 STD 4 40 ng / ml unknown unknown Average: Rel. Std. Dev: % % % % R-Biopharm Rhône Mycotoxin Technical Manual 20

21 7 Determining the Concentration of Toxin Present The following section works through some examples for each toxin on how to determine the multiplication factor for converting nanograms per injection to parts per billion. 7.1 AFLAPREP An Example 50 g of sample --> 500 ml of extraction buffer. 1 g of sample <-- 10 ml of filtrate taken and passed through IAC. Elute in total volume of 2 ml. 1 g of sample --> 2ml g < µl sample injected onto HPLC. An estimated total of 0.05 g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of ng/g = 1 ppb. 1 g per 0.05 g = multiplication factor of 20. Therefore, ng obtained x 20 = ppb. 7.2 DONPREP An Example 25 g of sample --> 200 ml of extraction buffer g <-- 2 ml of filtrate taken and passed through IAC. Elute in total volume of 1.5 ml. Evaporate to dryness g of sample --> 1 ml reconstitution volume g < µl sample injected onto HPLC. An estimated total of g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of ng/g = 1 ppb. 1 g per g = 40. Therefore, ng obtained x 40 = ppb. 7.3 EASI-EXTRACT AFLATOXIN An Example 50 g of sample < ml of extraction buffer. 1 g of sample <-- 2 ml of filtrate taken and passed through IAC. Elute in total volume of 3 ml. 1 g of sample --> 3 ml g < µl sample injected onto HPLC. An estimated total of g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of ng/g = 1 ppb. 1 g per g = 30. Therefore, ng obtained x 30 = ppb. 7.4 EASI-EXTRACT T-2 & HT-2 An Example 25 g of sample --> 200 ml of extraction buffer g <-- 2 ml of filtrate taken and passed through IAC. Elute in total volume of 1.5 ml. Evaporate to dryness g of sample --> 1 ml reconstitution volume g < µl sample injected onto HPLC. An estimated total of g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of ng/g = 1 ppb. 1 g per g = 40. Therefore, ng obtained x 40 = ppb. 21

22 7 Determining The Concentration Of Toxin Present 7.5 EASI-EXTRACT ZEARALENONE An Example 25 g of sample --> 125 ml of extraction buffer. 4 g <-- 20 ml of filtrate taken and diluted with PBS. 1 g <-- 25 ml of diluted filtrate taken and passed through IAC. Elute in total volume of 3 ml. 1 g of sample --> 3 ml g < µl injected onto HPLC. An estimated total of g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of ng/g = 1 ppb. 1 g per g = 30. Therefore, ng obtained x 30 = ppb. 7.6 FUMONIPREP An Example 25 g of sample --> 125 ml of extraction buffer. 2 g <-- 10 ml of filtrate taken and made up to 50 ml. 2 g --> 50 ml. 0.4 g --> 10 ml passed through IAC. Elute in total volume of 3 ml. 0.4 g of sample --> 3 ml g <-- 70 µl taken to derivatise g < µl total volume after derivatisation g < µl injected onto HPLC. An estimated total of g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of OCHRAPREP An Example Example 1: 50 g of sample --> 200 ml of extraction buffer. 1 g <-- 4 ml of filtrate taken and passed through IAC. Elute in total volume of 3 ml. 1 g of sample --> 3 ml g < µl injected onto HPLC. An estimated total of g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of ng/g = 1 ppb. 1 g per g = 30. Therefore, ng obtained x 30 = ppb. Example 2: 10 g of sample --> 200 ml of extraction buffer g <-- 5 ml of filtrate taken and passed through IAC. Elute in total volume of 3 ml g of sample --> 3 ml g < µl injected onto HPLC. An estimated total of g of sample would be injected onto the system for analysis. However, to report your results in ppb, i.e. ng/g of sample, you should multiply the estimated ng of sample detected by a factor of ng/g = 1 ppb. 1 g per g = 40. Therefore, ng obtained x 120 = ppb. 1 ng/g = 1 ppb. 1 g per g = Therefore, ng obtained x = ppb. R-Biopharm Rhône Mycotoxin Technical Manual 22

23 8 Validating a Method During the validation process it is important to determine if the method is fit for purpose, i.e. do the characteristics of the method meet with the minimum characteristics required. Validation is about establishing the analyte and matrices to which the method can be applied, establishing the LOD, LOQ, the concentration range over which the method can be employed, and the accuracy and precision of the method. Once the characteristics of a method have been established, they should be compared with the method requirements for a particular purpose to establish whether they are a good match. For a method to be fit for purpose there should be a good match between characteristics and requirements. In order to validate a method, reference material or spiked samples should be used to determine recoveries and the method should show good precision, repeatability and reproducibility. R-Biopharm Rhône would recommend that a minimum of 10 columns should be used to validate a method. 8.1 Precision The precision of an analytical procedure expresses the scatter between a series of measurements obtained from multiple sampling of the same sample under certain conditions. The precision is expressed as % relative standard deviation (% RSD) for a statistically significant number of samples, e.g. > Repeatability (r) Expresses the precision under the same operating conditions over a short interval of time. It is therefore, a measure of precision using: the same analyst. the same apparatus. close time interval between replicate analytes. the same reagents. the same laboratory, etc. For example, validation of a method within a lab would involve testing 10 columns on the same day. During method validation, method repeatibility and column repeatibility can be determined. Column repeatibility is determined by running the same extract through 10 columns. Method repeatibility is determined by extracting the same sample 6 times and passing each extract through one column. RBR would recommend the % RSD is <10 %. 8.3 Reproducibility (R) Express the precision between different laboratories. It is therefore, a measure of precision using: different analysts. different laboratories. different equipment. different sources of supply of reagents. For example, large-scale validation as in a ring trial would involve several labs testing the same method. % RSD = (STDEV / MEAN % RECOVERY) x 100 Precision is normally determined as repeatability (r) and reproducibility (R) 23

24 8 Validating a Method 8.4 Limit of detection The limit of detection (LOD) for a given analyte is the minimal amount needed to be able to distinguish the analyte signal above the background detector noise. A peak is at the LOD when the signal value measured at its maximum is 3 times the noise signal. 8.5 Limit of quantification The limit of quantification (LOQ) of an analyte is the lowest amount of that analyte in a sample which can be accurately measured with reliability. The LOQ is generally taken as being 3 times the LOD. 8.6 Determination of LOD and LOQ of an HPLC system The LOD / LOQ varies depending on method, HPLC conditions, derivatisation technique and the HPLC equipment used. LOD = 3 x signal / noise ratio LOQ = 3 x LOD 8.7 Accuracy The accuracy of an analytical method expresses the closeness of a measured level of an analyte to the known true level of that analyte. Accuracy is frequently reported as % recovery. The recovery provides an insight into the extraction efficiency of a method. R-Biopharm Rhône Mycotoxin Technical Manual 24

25 9 How to Spike a Sample Reference material can be purchased, however it can be difficult to source the matrix of interest therefore samples can be spiked in order to validate a method. To spike a sample a known blank sample should be taken and the required volume of sample should be weighed out into a suitable container. The calculated volume of standard should then be added to the sample and left overnight in the dark. If the sample is naturally contaminated it is advised not to spike the sample as discrepancies in results can occur. However, if you are required to spike a contaminated sample, i.e. in FAPAS rounds, you should spike the sample at approximately 10 times the level of contamination. C1 = V1 = C1 x V1 = C2 x V2 Concentration of standard to be used for spiking Volume of standard to be added to sample C2 = Required concentration of sample V2 = Required volume of sample 40 ml of milk was taken and spiked at 25 ppt using a 40 ng/ml Standard: C1 x V1 = C2 x V2 40 ng/ml x V1 = 25 ppt (0.025 ppb) x 40 ml 40 ng/ml x V1 = ng/ml x 40 ml 40 ng/ml x V1 = 1 ng V1 = 1 ng per 40 ng/ml V1 = ml 25 g of turmeric was taken and spiked at 5 ppb using a 1,000 ng/ml Standard: C1 x V1 = C2 x V2 1,000 ng/ml x V1 = 5 ppb x 25 g 1,000 ng/ml x V1 = 5 ng/ml x 25 g 1,000 ng/ml x V1 = 125 ng V1 = 125 ng per 1,000 ng/ml V1 = ml 5 g of maize was taken and spiked at 2 ppm using a 1,000 ng/ml Standard: C1 x V1 = C2 x V2 1,000 ng/ml x V1 = 2 ppm (2,000 ppb) x 5 g 1,000 ng/ml x V1 = 2,000 ng/ml x 5 g 1,000 ng/ml x V1 = 10,000 ng V1 = 10,000 ng per 1,000 ng/ml V1 = 10 ml 25

26 10 General Guidelines on Improving Recoveries CEN specifications state that recoveries should be between %. If however, recoveries obtained are below the acceptable 70 % check the following Check the ph of the sample lies within the range A ph out-with this range can affect the performance of the antibody. Use the recommended IFU or application notes. If there is no specific application note available contact R-Biopharm Rhône for assistance. Check the volume applied to the column. The sample should not take longer than 30 minutes to pass through the column. If necessary filter or centrifuge the sample after dilution with PBS to remove precipitates. Exposing the antibody to low levels of solvent over a prolonged period of time can have an adverse effect. Check the concentration of solvent being passed through the immunoaffinity column. Higher solvent concentrations can decrease the performance of the antibody. The following solvent concentrations should not be exceeded Use KOBRA CELL for derivatisation of aflatoxins. Use the standards provided by RBR. Ensure that backflushing has been carried out during the elution step. Backflushing increases the amount of time the solvent is in contact with the antibody gel, ensuring that all of the toxin is eluted. Aflatoxin Ochratoxin Zearalenone Fumonisin T-2 & HT-2 30 % methanol 2.5 % acetonitrile 5 % acetonitrile 15 % acetonitrile 5 % methanol 5 % acetonitrile 18 % methanol R-Biopharm Rhône Mycotoxin Technical Manual 26

27 11 General Guidelines on Improving Chromatograms Each of the following items needs to be optimised in order to generate a satisfactory separation and a chromatogram that is suitable for quantitative purposes: Mobile phase composition. Columns and packing dimensions. Injection volume. Sample pre-treatment. Mobile phase flow rate. Column temperature. Detector parameters High baseline noise Stop the pump. If noise disappears, the cause lies in the pump (refer to section 12 or contact your HPLC provider). If the noise remains the cause lies in the detector (refer to section 12 or contact your HPLC provider) Mobile phase contamination 11.3 Contamination downstream from the pump Remove the device that appears to be contaminated. If baseline settles, clean the device Poor reproducibility for peak areas When the retention time reproducibility is normal the pump is most likely not the cause. The source of the problem is likely to be found in the auto sampler OR Due to deterioration of the sample itself Peak broadening and peak splitting Problems with peak resolution indicate: Improper mobile phase (adjust mobile phase). Deteriorated column. A failed fitting. Excessive injection of sample. Increase the detector wavelength. If the drift decreases there is a problem with the mobile phase. Prepare fresh mobile phase. Use higher grade solvents. 27

28 11 General Guidelines on Improving Chromatograms 11.6 Ghost peaks A ghost peak is a peak that has come from a previous injection, or from contaminated equipment or mobile phase. Cause Confirmation Method What do to Contamination of pump or equipment upstream from pump. Mobile phase contamination. Contamination in auto sampler. Watch for the ghost peaks while injecting mobile phase. Clean using an acid or alkaline solution: Acid : Nitric acid 4-6 N. Alkaline : Methanol 5-10 % W/V. Use higher grade solvents / water. Clean equipment. R-Biopharm Rhône Mycotoxin Technical Manual 28

29 12 HPLC Troubleshooting 12.1 Introduction Critical criteria for successful HPLC analysis are: Correct mobile phase. Analytical column and guard column in good condition (change every 2-3 months). Lamp for the detector in good condition (<1,000 hours use). Once the conditions for HPLC analysis have been optimised the next critical step is sample preparation. Hence, to optimise the clean-up effect of the immunoaffinity column all samples must be analysed using protocols recommended in the application notes. This is especially critical for samples with high pigmentation e.g. spices and dried fruits. Note: R-Biopharm Rhône only recommend the following information as a general guide to troubleshooting your HPLC system. For further information we would advise you to contact your HPLC provider. Solvent Rack & Reservoires Pump Auto sampler Column heater and column Detector 29

30 12 HPLC Troubleshooting 12.2 When to Adjust the Mobile Phase More water should be added: If unable to get baseline separation between peaks. This will allow complete elution of the first toxin before elution of the second toxin begins. The run time will increase but good peak separation is essential for accurate chromatography. If toxins are eluted too quickly and merge with the solvent front. The toxins will be retained on the column for longer and allow the baseline to settle before elution of the toxins. More solvent should be added when: Peaks are too broad (starting to merge). Retention time is too long. There is a contaminating peak from the matrix. Adjusting the mobile phase is not always successful. Sample preparation and clean up must be optimised to improve HPLC chromatography. If there is a contaminating peak from the matrix. R-Biopharm Rhône Mycotoxin Technical Manual 30

31 12 HPLC Troubleshooting 12.3 Solvents and Mobile Phase Solvents that are not miscible will partition from each other, resulting in incomplete removal of the original solvent from the HPLC system. Therefore, when changing the mobile phase the following sequence of solvents should be used: Aqueous Solvents Buffer Solutions or Salt Solutions Water Ethanol, Isopropanol or Acetone Organic Solvents Acetonitrile or Methanol Chloroform or Ethyl Acetate n-hexane or ISO-Octane 31

32 12 HPLC Troubleshooting 12.4 Pump Abnormal operating pressure of the pump may be observed and can either be: Too high Too low Variable Pump Pressure Too High Cause Confirmation Method What To Do Downstream side of pump is clogged, e.g. clog in auto sampler, column, filter, tubing, fittings. Line filter is clogged (behind pump outlet). Damper or tubing is clogged inside the pump. Remove and examine devices and / or parts that may be clogged in the flow route. Operate the pump without anything connected to the outlet port. If the pressure is 20 kg / cm 2 or higher when pumping at 10 ml / minute, the line filter is clogged. Remove the outlet port filter and pump the solvent. If the pressure is 20 kg / cm 2 or higher when pumping at 10 ml / minute, the internal damper or internal tubing is clogged. Refer to the appropriate manual. Replace the line filter. Requires engineer or pump to be returned for service. R-Biopharm Rhône Mycotoxin Technical Manual 32

33 12 HPLC Troubleshooting Pump Pressure Too Low Cause Confirmation Method What To Do Pump head contains air bubbles. Inlet filter is clogged (use of buffer solutions often leads to clogging). Plunger seal is leaking. Inlet tube fitting is leaking. Leak at downstream side of pump (auto sampler, column, fittings). Pump head deterioration. When disconnected solvent drips rather than flows from inlet line. Perform pressure test (put screw into pump outlet, at 0.1 ml / min pressure should reach 500 kg / cm² smoothly). Open the purge valve and remove the air bubbles. Replace the inlet filter. Tighten the pump head securing screws. Replace plunger seal. Tighten the fitting at the pump inlet. Refer to appropriate manual. Clean pump head thoroughly with acid or alkaline solution. Variations in Pump Pressure Cause Confirmation Method What To Do Pump head contains bubbles. Inlet filter is clogged. Insufficient degassing of the solvent. Inlet tube fitting is leaking. Plunger seal is leaking. Open the purge valve and remove the bubbles. Replace the inlet filter. Degas the solvent. Tighten the fitting. Tighten the pump head securing screws. Replace the plunger seal. Pump head deterioration. Perform pressure test. Clean the pump head thoroughly using an acid or alkaline solution. Abnormal plunger movement. Sound of metal contacting metal can be heard. Loosen the pump head securing screws slightly. 33

34 12 HPLC Troubleshooting 12.5 Column and Column Heater Always use a column heater to prevent: Periodic baseline drift. Lengthening of retention time. Shortening of retention time. Poor reproducibility of retention time Detector Dealing with noise: Cause Confirmation Method What to do Bubble in cell. Monitor the baseline with pump on and off. Bubbles cause abnormally large baseline variation when the flow rate changes. 1. Remove air from the solvent using a degasser. 2. Install a back pressure coil. Suspension in cell. Use a magnifying glass to carefully examine the cell. Again monitor baseline as for 'Bubble in cell' method. Clean the cell: wash through or disassemble. Fluid is leaking from cell. Remove the cell from the instrument. Examine the cell and the inside of the instrument. 1. Clean and tighten connections. 2. Replace damaged cell. Energy reduction. If the output is 0.4 V at 250 nm, the lamp or mirror has deteriorated. Replace the lamp (1000 hours) or mirror. (5000 hours). Solvent absorption or cell window contamination. 1. Use high grade solvents. 2. Check the wavelength. 3. Wash through cell. R-Biopharm Rhône Mycotoxin Technical Manual 34

35 12 HPLC Troubleshooting Electronic noise. Ground the instrument. 1. Solvent absorption. 2. Cell window contamination. 1. Use high grade solvents. 2. Clean the cell. 3. Column equilibrium. 1. Increase in bubble size. 2. Column contamination. 3. Reduce flow of solvent. Use 'Bubble in cell' method. 1. Remove air from the solvent using a degasser. 2. Install a back pressure coil. 3. Wash / change column. 4. Check pump and / or blocked tubing. Decrease in bubble size. Use 'Bubble in cell' method. 1. Remove air from the solvent using a degasser. 2. Install a back pressure coil. 35

36 12 HPLC Troubleshooting 12.7 Removing Bubbles from Detector Cell Increase the pump flow. If bubbles remain: Connect the outlet port of the pump directly to the inlet of the cell and pump at 2 ml per minute. If bubbles remain: Attach a stainless steel tube to the outlet of the flow cell and cover with a teflon tube. Bubbles may be compressed and pass through the cell if pressure is applied within the cell. Apply pressure on syringe and continually bend and release the teflon tube. If the baseline moves in the negative direction the bubbles in the cell are decreasing in size. Continue until all bubbles have passed through the cell Cleaning the Detector Flow Cell Without Disassembly Contamination of the cell wall can sometimes be removed by passing the following organic solvents or strong nitric acid through the cell. When sufficient cleaning cannot be obtained using this method, disassemble and clean the flow cell. Replace the solvent in the flow cell with distilled water. Pass concentrated nitric acid through the cell. (4 N, 1 ml / minute for 30 minutes) Pass distilled water through the cell. (1 ml / minute for 30 minutes) Pass acetone through the cell. (1 ml / minute for 30 minutes) Pass distilled water through the cell. (1 ml / minute for 30 minutes) Caution: Disconnect the column before running this decontamination protocol. R-Biopharm Rhône Mycotoxin Technical Manual 36

37 12 HPLC Troubleshooting 12.9 Prevention of Air Bubbles Bubbles can be prevented by always using a degasser. Solvent Absorption can be prevented by always using HPLC grade solvent. Cell suspensions and tube blockages can be prevented by flushing the system with water following use of a mobile phase which contains salt. When using a methanol wash before or after a mobile phase containing salt take care to ensure the salt is soluble in methanol. If uncertain use water before and after running the mobile phase. Advisable washing regime would be 30 minutes water at 1 ml per minute followed by 30 minutes methanol at 1 ml per minute. When dealing with organic mobile phase replace the solvent inside the cell with a solvent which is miscible with the organic mobile phase and water, do not mix liquids which are not miscible (e.g. chloroform and water). Use methanol as a solvent which is miscible with both Maintenance of Fluorescence Detectors The lamp should be changed every 1000 hours. This can be done in-house and the maintenance manual usually has a very good description of how to change the lamp. If having difficulty changing the lamp watch the engineer change it or have the engineer supervise changing the lamp. Before changing the lamp set excitation wave length to 400 nm and emission wavelength to 500 nm. Record the excitation and emission out put (V) values. These values should increase when the new lamp is installed. The dust filters should be changed each time the lamp is changed to prevent dust building up inside the detector. When the air filter is clogged the lamp cooling efficiency is reduced, contributing to faster deterioration of the lamp. The best way to clean the dust from the inside of the detector is to use an air pump and blow the dust away. This avoids touching any parts and the risk of causing damage. 37