Analytical and Bioanalytical Chemistry

, Volume 401, Issue 2, pp 463–481 | Cite as

Recent developments in doping testing for erythropoietin



The constant development of new erythropoiesis-stimulating agents (ESAs), since the first introduction of recombinant erythropoietin (rhEpo) for clinical use, has also necessitated constant development of methods for detecting the abuse of these substances. Doping with ESAs is prohibited according to the World Anti-Doping Code and its prohibited list of substances and methods. Since the first publication of a direct and urine-based detection method in 2000, which uses changes in the Epo isoform profile as detected by isoelectric focusing in polyacrylamide slab gels (IEF-PAGE), the method has been constantly adapted to the appearance of new ESAs (e.g., Dynepo, Mircera). Blood had to be introduced as an additional matrix, because Mircera (a PEGylated Epo) is best confirmed in serum or plasma after immunoaffinity purification. A Mircera ELISA was developed for fast screening of sera. With the appearance of Dynepo and copy epoetins, the additional application of sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE or equivalent) became necessary. The haematological module of the Athlete Biological Passport is the latest development in multivariable indirect testing for ESA doping. The article summarizes the main strategies currently used in Epo anti-doping testing with special focus on new developments made between 2009 and 2010.


Recent developments in doping testing for Epo include mass-based methods. The apparent molecular mass of most rhEpos is higher than for uhEpo, which can be shown by SDS-PAGE


Erythropoietin (Epo) Doping control Electrophoresis Mass spectrometry ELISA Biological passport Direct and indirect detection 


In 1985 Lin et al. [1] described the cloning of the human Epo gene, which was the basis for the expression of the gene in transfected cell lines. Epo is a glycoprotein hormone produced mainly by the kidneys, and is responsible for the development of bone marrow stem cells into red blood cells (RBC, erythrocytes), which transport oxygen to organs and tissues [2]. The entire process which leads to the production of RBC is called erythropoiesis, and substances which stimulate RBC production are called erythropoiesis-stimulating agents (ESAs). An increase in RBC is performance enhancing, especially in endurance sports (e.g., cycling, running, cross-country skiing). The first recombinant human Epo (rhEpo; epoetin alfa; Eprex, Erypo, Procrit) was approved in 1989 for clinical use. Because misuse by athletes was suspected, the International Olympic Committee (IOC) put Epo on its list of prohibited substances in 1990 [3]. However, no IOC-approved test existed at that time. In 2000 a direct test was published by Lasne et al. [4] together with a blood-property-based indirect test [5]. The direct test became the first WADA-approved method for Epo-doping control and was based on the principle that rhEpo application may lead to a temporary change in the endogenous Epo isoform distribution. This observation was made as early as the 1990s by other scientists (see the section “IEF-PAGE”). Further ESAs followed in 1990 (epoetin beta; NeoRecormon), 2001 (darbepoetin alfa; Aranesp or NESP), 2002/2007 (approval/launch of epoetin delta; Dynepo), and 2007 (methoxy polyethylene glycol–epoetin beta; Mircera, continuous erythropoietin receptor activator, or CERA). After the IOC, the World Anti-Doping Agency (WADA; established in 1999) became responsible for the prohibited list. The 2011 version of the list [6] bans ESA doping according to article S2 “Peptide hormones, growth factors and related substances” (prohibited substances) and also M2 “Gene doping” (prohibited methods). The misuse of non-approved substances is covered by article S0 “Non-approved substances”. The detection of Epo-doping has been regulated by WADA in a technical document (TD2009EPO) [7], and comprises mostly gel electrophoretic methods (i.e. isoelectric focusing (IEF), sodium dodecylsulfate polyacrylamide gel electrophoresis (SDS-PAGE) or equivalent). Additional methods have been developed, including enzyme-linked immunoassays (ELISA), membrane-assisted isoform immunoassays (MAIIA), mass spectrometry-based methods, and procedures for the detection of gene doping and sample manipulation with proteases. The purpose of the “Athlete Biological Passport” and its “haematological module” is to enable prolonged detection of Epo doping by statistical application of indirect (blood profile) markers in a multivariable, longitudinal, and athlete-specific. Currently, the following techniques are used by WADA-accredited laboratories for detection of ESA doping: gel electrophoretic methods (IEF-PAGE, SDS-PAGE or equivalent), ELISA-based methods (measurement of Epo and Mircera concentrations), protease tests, and the determination of haematological data (“haematological module” consisting of haematocrit (HCT), haemoglobin (HGB), red blood cells count (RBC), percentage of reticulocytes (RET%), reticulocytes count (RET#), mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC), and index of stimulation (OFF-hr score)).

GEL Electrophoretic methods

Methods based on the electrophoretic separation of Epo molecules and subsequent immunological detection (Western blot) are still dominant in detection of Epo doping. They are direct detection methods similar to immunoassay-based methods, but additionally allow a visualization of the antibody-bound molecules regarding charge heterogeneity (IEF-PAGE) or molecular mass (SDS-PAGE, SAR-PAGE). Based on this information a differentiation between recombinant and endogenous erythropoietins is possible.


A brief description of the method used in doping control

Isoelectric focusing (IEF) of Epo in polyacrylamide slab gels (polyacrylamide gel electrophoresis, PAGE) was the first method which enabled the detection of rhEpo abuse for doping purposes. Because of its charge heterogeneity Epo can be separated into so-called Epo isoforms, which together form an Epo isoform profile. The profiles significantly differ between rhEpos, human urinary Epo (uhEpo), and human serum Epo (shEpo). The first validated doping test using this principle was developed by Lasne et al. (2000) [4, 8]. The test was based on observations by Wide and Bengtsson (1990) [9], Tam et al. (1991) [10], and Wide et al. (1995) [11], who showed by zone electrophoresis (Wide et al.) and IEF in solution (Tam et al.) that:
  1. 1.

    human urinary Epo contains greater amounts of acidic isoforms than rhEpo and shEpo; and

  2. 2.

    this difference can be used to detect the administration of rhEpo in both serum and urine [11].


Key improvements by Lasne et al.—namely the use of PAGE and the development of the so-called Western double-blotting technique—led to a test which, since then, has been widely used for the detection of Epo doping. In its original form the test uses urine (typically 20 mL), which is concentrated after adjustment with 3.75 mol L−1 Tris-HCl buffer to pH 7.4, sedimentation, and microfiltration in two ultrafiltration steps (30 kDa molecular weight cut-off) down to ca 20–40 μL. A mix of protease inhibitors is added during the entire sample-preparation procedure. Adjustment of the pH of the sample is necessary because it keeps (intact) Epo isoforms negatively charged (their isoelectric points (pI) are below pH 7.4); this prevents the action of acidic proteases and glycanases (e.g. neuraminidases) contained in urine, and also helps to dissolve urinary precipitates which may contain Epo. The retentate obtained after ultrafiltration contains Epo but also a large amount of urinary proteins (e.g. 20–40 μg μL−1). To avoid overloading effects (band deformation, smearing) during the subsequent electrophoretic separation, Epo can be affinity purified (e.g. by wheat germ agglutinin (WGA) [8] or anti-Epo antibodies [12, 13, 14]) before IEF-PAGE, the concentration of Epo in the retentate determined by ELISA and adjusted, e.g. to 500 IU L−1, for each sample, or methyl red (a pH indicator with a pI of 3.8) can be added to the IEF catholyte for accurate band identification. Typically, 5% T (total acrylamide) and 3% C (bisacrylamide crosslinker) polyacrylamide gels are used which contain carrier ampholytes in the pH 2–6 range at a final concentration of 2% (w/v), 7 mol L−1 urea, and 5% sucrose for stabilizing the pH gradient. After a prefocusing step to establish the pH gradient, retentates are heated for 3 min at 80 °C, supplemented with 1% Tween-80 or Tween-20, and applied cathodically to the gel by use of sample application pieces, a sample applicator strip, or—preferably—precast wells in the gel (vide infra). The main focusing step is performed at constant wattage (1 W cm−1 gel length and a 1 mm thick gel) and for 3600–4000 V h. Proteins are then semi-dry electroblotted on to a polyvinylidene fluoride (PVDF) membrane and incubated after a reduction (5 mmol L−1 dithiothreitol (DTT) in phosphate-buffered saline, PBS) and a blocking step (5% non-fat milk (NFM) in PBS) with the primary antibody (mouse monoclonal anti-Epo antibody clone AE7A5 in 1% NFM in PBS). The membrane is then washed with 0.5% NFM in PBS and the bound antibody transferred under acidic semi-dry blotting conditions (0.7% acetic acid) to a second PVDF membrane. Lasne [15, 16] introduced this procedure to avoid interfering interactions of the secondary antibody (usually a polyclonal goat anti-mouse IgG antibody) with urinary proteins other than Epo on the first membrane (Note: a similar technique using also a “donor” and “acceptor” membrane for the detection of antibody cross-reacting proteins was published as early as 1992 by Hammerl et al. [17] and is known as “Western cross blot”). The successful introduction of this so-called “double-blotting” technique was a key step in the development of the WADA-approved Epo IEF-PAGE method for doping-control purposes. The second membrane is then treated as a regular Western blot, i.e. blocked (5% NFM in PBS), incubated with a secondary biotinylated antibody, washed (0.5% NFM in PBS), treated with streptavidin horseradish peroxidase (HRP) complex, washed again (PBS), and developed using an enhanced chemiluminescence substrate. A CCD camera is used as signal detector, which enables accurate quantification of the bands obtained within each Epo isoform profile. The IEF profiles are evaluated after quantitative software-based image analysis [18, 19, 20] and in accordance with the criteria of the WADA technical document on Epo analysis (TD2009EPO) [7].

The WADA technical document TD2009EPO

The evolution of criteria for the interpretation of Epo IEF profiles is directly associated with the appearance of new Epo pharmaceuticals on the market. When the IEF method was first published in 2000 only two types of rhEpo, epoetin alfa and beta, were commercially available; these could easily be differentiated from uhEpo by using the percentage of “basic isoforms” as main criterion, which meant that samples exceeding a certain percentage were declared positive (a cut-off of 80% was used for a some time) [21]. In 2001 darbepoetin alfa (Aranesp, NESP) was approved and was rapidly used for doping purposes. Because of its higher content of sialic acids NESP is focussed in the acidic region of the IEF gel [8, 22]. NESP-positive cases were detected during the Winter Olympic Games of Salt Lake in 2002. To harmonize the detection of these three forms of non-endogenous epoetins by IEF-PAGE the World Anti-Doping Agency issued a technical document in 2004 (TD2004EPO). Epo profiles had to pass three types of criteria, namely acceptance, identification, and stability criteria. The “percentage of basic isoforms” rule was replaced by the “most intense bands” criteria (Fig. 1), i.e. rhEpo (epoetin alfa and beta) positive cases had to have the two most intense bands in the basic area, they had to be consecutive, and the most intense band had to be 1, 2, or 3. For NESP-positive cases, the two most intense bands had to be in the acidic area and the most intense band had to be C or D (identification criteria). The “80% basic bands” rule should no longer be used. Occasionally urine samples were unstable, which caused IEF profiles to shift toward the basic region, or even outside the pH 6 region. Instability of urine samples has been ascribed to enzymatic degradation of Epo caused by bacterial contamination (see the section “Detection of proteases”). Hence, TD2004EPO also made the performance of a stability test mandatory for all confirmations and B analyses (stability criteria). Also clarified was the inclusion of positive and negative standards and control samples for the screening and confirmation procedures and the data contained in reports. A new version of the technical document was issued by WADA in 2007 (TE2007EPO). It explicitly mentioned immunoaffinity purification as an acceptable additional step during sample preparation and switched from visual or densitometric band intensity assessment to evaluation by densitometry only. Accordingly, the wording “more intense” was also described by “approximately twice or more”—which was an attempt to describe the visual impression of bands being “more intense”. Hence, the wording “approximately twice or more” was used and not “twice or more”—keeping in mind that the visual impression and band distribution of the entire profile are of importance. In 2007 epoetin delta (Dynepo) entered the European market and also Mircera (CERA) and several biosimilar epoetins were approved in Europe (e.g. Binocrit, Epoetin Alfa Hexal, Abseamed, Silapo, Retacrit). Copy pharmaceuticals containing mostly epoetins alfa and beta were already available before 2007 in non-European countries (e.g. Korea, China, India, South America) [23]. Although all these Epo preparations were detectable in the pH range 2–6 of the IEF-PAGE method [24, 25, 26, 27], their profiles differed partly from the profiles described in TD2007EPO. In 2009 a new technical document was released by WADA (TD2009EPO) [7], which adapted the 2007 document to the new situation. Specifically, identification criteria for CERA (presence of four consecutive bands, which correspond to the CERA standard) and “other epoetins” (e.g. the sum of basic isoforms must be 85% or more) were added and the term “additional evidence” introduced. If IEF band intensity criteria were not fulfilled, “the application of an electrophoretic SDS-PAGE procedure or equivalent where protein separation is based on a different principle (i.e. apparent molecular mass or hydrodynamic volume) can be used complementarily to the IEF method for the purpose of helping to confirm the exogenous or endogenous origin of the finding.” Gaining additional evidence by SDS-PAGE has proved especially useful for the prolonged detection of Dynepo and some biosimilar or copy epoetins (vide infra).
Fig. 1

Identification of Epo isoforms according to TD2009EPO (WADA). Whereas isoforms of uhEpo (National Institute for Biological Standards and Control (NIBSC); Hertfordshire, UK) are mostly found in the “endogenous area” of the IEF gel, isoforms of rhEpo (BRP-Epo, European Directorate for the Quality of Medicines; Strasbourg, France) and Mircera are primarily focussed in the “basic area”. Most of the NESP isoforms migrate into the “acidic area”. Effort and unstable urine samples cause shifts of the uhEpo profile toward the cathode. Source: TD2009EPO [7], World Anti-Doping Agency (WADA) (permission granted)

Recent research results

Zinc alpha-2-glycoprotein (ZAG) was identified in 2008 [28] as a protein which interacts non-specifically with the primary anti-Epo antibody used during the IEF procedure (clone AE7A5). Because ZAG focuses outside the region of the IEF gel which is used to evaluate the IEF profiles, no interference with the final results occur. The binding was known for several years [12, 29, 30] but the protein was not identified until 2008. A shotgun proteomics approach enabled the successful identification. Confirmation experiments using the Western double-blotting procedure and a recombinant version of ZAG showed that the interaction occurs only at high concentrations of ZAG (i.e. μg range on the gel). Earlier observations of non-specific binding of clone AE7A5 antibody made in the context of two-dimensional electrophoresis [31, 32] or SDS-PAGE [32, 33, 34] could either not be confirmed for the one-dimensional carrier ampholyte-based IEF-PAGE method or were shown to occur outside the pI range used for Epo testing [35, 36, 37, 38]. An additional and mandatory immunoaffinity purification step was introduced for SDS-PAGE-based analyses with TD2009EPO [7].

Lamon et al. (2009) [39] performed a study to identify conditions which lead to so-called “effort (atypical) profiles” and “effort urine samples”. Top cyclists had to perform a series of short strenuous exercises (e.g. four 500-m rides at maximum speed), and urine and blood were collected before and after the exercises. After the exercise a ca 300-fold increase in retinol binding protein (RBP) was observed in the urine and 8.5 and 10-fold increases in total protein and Epo, respectively. IEF-PAGE was performed after immunoaffinity purification [12] of urine and serum samples. The authors clearly demonstrated that this type of exercise led to “effort profiles” in urine but to no changes of the shEpo IEF profile. The data were explained by a possible “exercise-induced transient renal dysfunction” [39], which resulted in the excretion of more basic (less acidic) Epo isoforms [40]. As was shown by Lasne et al. in 2007, shEPO is less acidic than uhEpo [12] and this may be used to explain the shift.

Discriminant analysis—a supervised multivariate statistical analysis technique frequently used in chemometrics—was applied by Lasne et al. (2007) [41] and Lamon et al. (2010) [42] for classification of IEF profiles. It could be demonstrated that this technique enabled improved detection of low-dose (microdose) rhEpo application [41, 43] and is also applicable to the detection of Dynepo doping [42]. Lamon et al. proposed use of a so-called “bands intensity (BI) score” as alternative criterion or for additional evidence in the detection procedure of Epo doping by the IEF method.

Because of the low excretion in urine, Mircera should preferably be detected in blood. Usually, serum or plasma samples have to be immunoaffinity purified to remove high-abundance proteins, which would otherwise distort the IEF profile [12, 27]. Lasne et al. (2010) [44] developed a fast precipitation procedure which kept shEpo, rhEpos, NESP, and Mircera in solution and thus enabled direct application of the supernatant to the IEF gel after washing and ultrafiltration.

The effect of heparin on the detection of Epo by IEF-PAGE was investigated by Reichel et al. (2010) [45]. To reduce the risk of life-threatening thromboembolic events (stroke, heart-attack) some athletes use anticoagulants in combination with Epo [46]. Heparin is a highly sulfated glycosaminoglycan and well known anticoagulant. Because of its polyanionic nature, unfractionated heparin (UFH) acts destructively in IEF-PAGE. Concentrations of ca 25 IE UFH or above led to a complete distortion of the Epo profile. No such effect was observable for low-molecular-weight heparins (LMWH), which completely migrated to the anode. UFH on the other hand generated a heparin gradient across the entire pH 2–6 range of the gel and thus disturbed the separation of the Epo isoforms. Immunoaffinity purification of EPO before IEF-PAGE or treatment of the urinary retentate with solid urea removed the effect of UFH. No such interference of UFH was observed in the separation of Epo on SDS-PAGE.

Sample application pieces have frequently been used to apply urinary retentates to IEF gels. In 2007 interference was observed which affected the separation of NESP isoforms and which could be ascribed to faulty application pieces. Reichel (2010) [47] performed a mass spectrometric study on working and defective application pieces and showed that defective application pieces (bought between 2007 and 2010) contained a complex mixture of alcohol ethoxylates, alcohol ethoxysulfates, and alkyl sulfates including SDS. Anionic detergents like these are, in general, incompatible with IEF-PAGE. Casting IEF gels with wells and directly applying the retentates into the wells was demonstrated to be the best solution to this problem.

The applicability of the IEF method for detection of doping of animals (horses, dogs) with rhEpos, NESP, and Mircera was demonstrated in 2005, 2006, and 2009, respectively [48, 49, 50].


Differences between the apparent molecular masses of endogenous erythropoietins (uhEpo, shEpo) and recombinant Epos and analogues have been studied by several authors [51, 52, 53, 54, 55, 56]. Sodium dodecyl sulfate (SDS), an anionic detergent which interacts with proteins and thus charges them negatively, is used with the sieving effect of polyacrylamide gels. Smaller proteins migrate faster through the gel toward the anode than bigger proteins, and so these can be separated from each other by electrophoresis. Because Epo consists of glycoforms with slightly different molecular masses [57] Epo bands on SDS-PAGE are usually broader than bands of non-glycosylated proteins. Typical apparent molecular masses obtained for various epoetins on SDS-PAGE are: uhEpo and shEpo ca 34 kDa, epoetins alfa (Eprex, Erypo) and beta (NeoRecormon) ca 36–38 kDa, epoetin omega (Repotin) ca 34 kDa, darbepoetin alfa (NESP) ca 44–45 kDa, epoetin delta (Dynepo) ca 36 kDa, and Mircera ca 69–78 kDa [58]. Biosimilar compounds and copy epoetins frequently have similar masses to epoetins alfa and beta. However, some formulations contain broader mass distributions because of lower and higher molecular mass glycoforms (e.g. Alfaepoetina, Epocrin; Fig. 2).
Fig. 2

Comparison of uhEpo (NIBSC standard), originator epoetins (Erypo, NeoRecormon, NESP, Dynepo), and some copy epoetins (Beijing 4 Rings, Shanpoietin, Erythrostim, Epocrin, Hemax, Alfaepoetina) after SDS-PAGE and Western double-blot; (A) and (B)

For application of SDS-PAGE to doping control the Epo contained in sample matrices (urine, serum/plasma) must first be immunoaffinity purified (e.g. by ELISA plate [55, 59] or monolithic disk [13]). Otherwise the gel would be overloaded with highly abundant proteins resulting in strong lane and band deformations. The workflows of IEF and SDS-PAGE (and Sarcosyl-PAGE; vide infra) currently used in doping control are compared in Fig. 3. Except for the different principle of separation (charge, molecular mass) they are mostly identical. Usually, mini-gels with 8–12% T and a thickness of 1–1.5 mm are used (the 1.5-mm gels are less prone to band deformation because of heat development during electrophoresis, and hence can be run at room temperature and with run-times below 1 h).
Fig. 3

Different workflows for direct detection of Epo doping by IEF- and SDS- or SAR-PAGE. Because of the high protein capacity of the gel, immunoaffinity purification is optional for urine samples on IEF-PAGE but mandatory for serum and plasma samples on both IEF- and SDS- or SAR-PAGE

Kohler et al. (2008) [60] used two internal standards with masses above and below epoetins alfa, beta, delta, and uhEpo, which enabled the calculation of relative mobility values and subsequent classification of these analytes in samples. Reichel et al. (2009) [59] showed that effort urine samples and unstable (“active”) urine samples can be also easily detected by SDS-PAGE without the need to perform a stability test. Contrary to IEF-PAGE the degradation of Epo in active urine samples causes a shift of the uhEpo band to lower molecular mass (i.e. toward the anode) whereas the IEF profile is shifted toward the cathode. The position of the Epo band of effort urine samples is hardly altered compared with the band of the uhEpo standard and with no shifts to the rhEpo region. These data were confirmed by Voss et al. (2010), who investigated the effect of high-intensity exercise on the relative mobility values of Epo [61]. Since 2009, SDS-PAGE has also been part of the WADA technical document on Epo analysis (TD2009EPO; vide supra).

SDS-PAGE has already been successfully used for detection of Dynepo doping [62]. Because of its narrow glycoform mass distribution [63] Dynepo produces a rather sharp band on SDS-PAGE, which is unique compared with other epoetins [60]. Compared with IEF-PAGE the detection window of Dynepo is longer on SDS-PAGE. On IEF-PAGE the alfa band of Dynepo is more intense than the alfa bands usually observed for epoetins alfa and beta. Nevertheless it is still less intense than the most intense band in the basic area (bands 1, 2, or 3). Because IEF-PAGE evaluates band intensities in the basic (and acidic) area relative to band intensities in the endogenous area (Fig. 1), the higher intensity of the alfa band (endogenous area) leads to a faster drop of this ratio during the washout phase after the application of Dynepo. Figure 4 shows the SDS-PAGE results for three athletes who tested positive for Dynepo.
Fig. 4

Detection of Dynepo in the urine samples of three athletes by SDS-PAGE. Note (1) the significant and narrow band shape of Dynepo in comparison with uhEpo (NIBSC standard; negative quality control (QC) urine), and (2) the almost complete absence of endogenous Epo from the three athletes’ samples. One athlete later publicly admitted having used Dynepo

Sarcosyl (SAR)-PAGE

Because of the prolonged serum half-life of Mircera (ca 130 h) and its limited excretion in urine, serum or plasma should preferably be used for detection of Mircera abuse. Hence, two matrices (urine, blood) must be taken to enable detection of abuse of different types of epoetin. Currently, four methods have to be used for unambiguous proof of their misuse by athletes:
  1. 1.

    IEF-PAGE for detection of epoetins alfa, beta, NESP, and some biosimilar compounds in urine;

  2. 2.

    SDS-PAGE for detection of Dynepo in urine;

  3. 3.

    IEF-PAGE for detection of Mircera in serum and/or plasma; and

  4. 4.

    Mircera-ELISA for screening of serum samples before confirmation with IEF-PAGE.

Unfortunately, SDS-PAGE has been shown to be less sensitive for Mircera than for the other epoetins, because of interference of the primary antibody used for the Western blot (clone AE7A5) with the large (30 kDa) PEG group of Mircera [64]. Reichel et al. (2009) solved the problem of the reduced sensitivity by exchanging the SDS in sample and running buffers of SDS-PAGE for sarcosyl [64]. Sarcosyl (SAR, sodium N-lauroylsarcosinate) is, similar to SDS, an anionic detergent but, unlike SDS, hardly interacts with the PEG group of Mircera and only with the protein chain. As a result Mircera migrates as a sharp band in SAR-PAGE (as the method was named by the author) and is detected with the same sensitivity as the other epoetins by the primary antibody. Because shEpo has a molecular mass apparently similar to that of uhEpo (in both SDS-PAGE and SAR-PAGE), SAR-PAGE can be used not only for detection of Mircera abuse in serum and/or plasma but also for detection of all other rhEpos and analogues. Another benefit of using serum or plasma is that strenuous exercise does not affect shEPO, as shown by Lamon et al. [39] and that no stability test for urine has to be performed. As shown in Fig. 5, 200 μL serum is sufficient for direct detection of recombinant epoetins (e.g. NeoRecormon, Dynepo, NESP), Mircera, and shEpo in a single experiment and with a single matrix (“one matrix – one method approach”).
Fig. 5

Subcutaneous single dose-administration studies of (A) Mircera (0.58 μg kg−1), (B) Dynepo (35 IU/kg), and (C) NeoRecormon (66 IU kg−1) after immunoaffinity purification of 200 μL serum and Sarcosyl (SAR)-PAGE. Both shEpo and the Epo pharmaceuticals were detectable on all three gels. Note the reappearance of the shEpo band in the Dynepo and NeoRecormon studies after several days and also the significant faint smear above the shEpo band on day 7 of the NeoRecormon study, which is characteristic of the washout phase. Source: Reichel et al. (2009) [64], Wiley (permission granted)

Two-dimensional electrophoresis

By combining charge-based separation (IEF-PAGE) with the molecular mass-based separation (SDS-PAGE) two-dimensional (2D) separation of proteins can be achieved on a single gel (2D-PAGE). Instead of performing IEF in carrier ampholytes, so-called immobilized pH-gradient (IPG) gels in strip form are normally used, which simplifies transfer of the IEF gel to the second dimension and also ensures highly reproducible results. Two-dimensional PAGE is a well established technique in proteome research, in which high-capacity and high-resolution methods are required. During recent years several authors have used 2D-PAGE for separation of recombinant erythropoietins, and were able to demonstrate that Epo isoforms have different apparent molecular masses [63, 65, 66, 67]. A 2D-PAGE method for detection of Epo doping was published by Khan et al. in 2005 [68], but the article caused discussion among experts [29]. However, 2D PAGE as a method has some technical disadvantages, which make its application for Epo-doping control unlikely, for example:
  1. 1.

    it is not a high-throughput method (the capacity is only one sample per gel);

  2. 2.

    very acidic epoetins, for example NESP, are only poorly resolved on commercial IPG strips, which end at pH 3 in the acidic region (although Görg et al. (2009) [69] demonstrated that good resolution of NESP is possible on non-commercial IPG strips which started at pH 2.5); and

  3. 3.

    it requires that all isoforms of the different epoetins migrate reproducibly and equally well from the first dimension IPG strip into the second dimension SDS-PAGE gel (otherwise artificial changes in the isoform distribution might occur).


ELISA-based methods

Commercial human Epo ELISAs (e.g. Stemcell Technologies Human Epo ELISA, R&D Systems Human Quantikine IVD ELISA Kit, Immulite Epo Assay) were designed to measure the concentration of human Epo, only, in serum or plasma. Because of the low concentration of Epo in human urine and interferences in the urinary matrix these ELISAs cannot/should not be used for direct measurement of the Epo concentration in urine. One way to measure the concentration of Epo in urine is to first ultrafilter the urine to concentrate the Epo and, in parallel, perform a buffer exchange (e.g. by washing with PBS or 50 mmol L−1 Tris-HCl buffer, pH 7.4) and then use the obtained retentate and mix it with the ELISA’s sample-diluent buffer before performing the assay. However, currently there is no commercial ELISA available which is capable of directly differentiating between plain recombinant and endogenous human Epos (the commercial ELISA for Mircera uses an artificial modification (PEGylation) of rhEpo for the detection; vide infra). To make commercial Epo ELISAs useful for screening purposes in human doping control, at least three principles have been applied during recent years. The samples were either:
  1. 1.

    enzymatically pretreated (NESP), or

  2. 2.

    pretreated with a precipitant (Mircera) before the application of the ELISA, OR

  3. 3.

    the elution characteristics of the ELISA’s capturing antibodies were used after sample incubation.


A fourth principle, use of species-specific differences in the primary structure of Epo, has been used for detection of human recombinant Epo in non-human samples (vide infra).

Wide et al. (2003) [70] showed that NESP could be differentiated from rhEpo and shEpo by treating serum samples with α(2→3,6,8,9) neuraminidase from Arthrobacter ureafaciens. Sera were first buffer adjusted to pH 5.6 by dilution with acetate buffer and then incubated for 1 or 24 h with the neuraminidase. Before and after this treatment the Epo concentration was measured with two commercial ELISA kits (R&D Systems, Medac). Because of its hyperglycosylation NESP has up to 22 terminal sialic acids whereas standard rhEpos (e.g. epoetin alfa and beta) have only up to 14 sialic acids [71]. Hyperglycosylation reduces the affinity of NESP for the Epo receptor and thus increases the serum half-life and biological activity of the molecule in vivo. On removing the sialic acids Wide et al. observed that the interaction between Epo and the ELISA antibodies (“immunoactivity”) was increased and that the relative increase in immunoactivity was much higher for NESP than for rhEpo and shEpo (Fig. 6). This difference was observable for both commercial ELISA kits and was already significant after incubation for 1 h with neuraminidase (for the Medac kit the mean relative increase after 1 h was 42% for shEpo and 282% for NESP; for the R&D Systems kit 118% and 231%, respectively). On the basis of this principle NESP could be detected in serum samples of patients up to 14 days after the last injection. However, so far the method has not been evaluated for doping-control purposes.
Fig. 6

Differentiation of NESP and shEpo after neuraminidase treatment by ELISA. The increase in immunoactivity is much higher for NESP than for shEpo after removal of sialic acids, which can be used for differentiation. Serum samples were collected between 2 and 14 days after NESP administration. Source: Wide et al. (2003) [70], Upsala Medical Society (permission granted)

In an attempt to generate anti-NESP antibodies Giménez et al. (2007) [72] raised polyclonal anti-sera against two NESP peptides, LVNSSQVNETLQLHVC and QVNETLQLHVDKAVSGLRSC (amino acids 81–95 and 86–104), and one Epo-peptide, LVNSSQPWEPLQLHVC (amino acids 81–95). The main idea was to exploit differences between the primary structures of rhEpo and NESP to generate NESP-specific antibodies (a total of five amino acids were exchanged in the amino acid sequence of Epo for NESP, namely the amino acids in positions 30 (A→N), 32 (H→T), 87 (P→V), 88 (W→N), and 90 (P→T)). A specific interaction for NESP was observed for the anti-NESP (84–106) antibody but only after immunoaffinity purification using a NESP (86–91) peptide column. No cross-reaction with NESP was observed for the Epo (81–95) antibody. Nevertheless, despite their potential the specific application of these antibodies for anti-doping-control purposes has not been reported.

The different amino acid sequences of human Epo, NESP, and horse Epo were exploited by Lasne et al. (2005) [48] for detection of horse doping with rhEpo and NESP. An ELISA from R&D Systems was used for the screening procedure. Because antibodies of the ELISA had only weak cross-reactivity with equine Epo, a single dose injection of 200 μg NESP led to a significant increase in measurable Epo in plasma during the first 24 h after the injection (156 IU L−1). After ten days the basal value (<2.5 IU L−1) was reached again. A similar but weaker reaction was also observed after treatment with epoetin alfa. However, after the last injection the concentration in plasma rapidly decreased and reached the starting value within 48–72 h. The results were confirmed by use of the IEF method with horse urine as sample matrix. A similar approach was described by Bartlett et al. (2006) [49] for the application of ELISA kits in canine doping control. Dogs received 2500 IU of epoetin alfa (subcutaneously and intravenously) and urine samples were taken before and after administration of the drug. Three human Epo ELISA kits were tested (R&D Systems, Neogen, Immulite). The R&D systems kit gave the highest Epo values. Typically, the maximum concentration was observed 3.75 h after administration (e.g. 179 mIU mL−1), which dropped below 0.2 mIU mL−1 after 32.3–49.3 h. Again, the ELISA kits were used for screening purposes only, and confirmation was done with the IEF method. The method was tested with ca 6000 greyhound sport samples and resulted in four positive cases—three for rhEpo and one for uhEpo (because of submission of human urine instead of greyhound urine to the control stewards). However, it must be pointed out that administration of rhEpo to horses and dogs may lead to the production of Epo antibodies, because of the different amino acid sequences of human, equine, and canine Epo, and, consequently, harmful immunological reactions after repeated use of rhEpo [73].

The applicability of a differential elution strategy in combination with a commercial ELISA (R&D Systems) was investigated by Mallorquí et al. in 2010 [74]. The authors used the discriminating power of the monoclonal capture antibody of the ELISA (clone 9C21D11) for basic Epo isoforms [12, 75]. Under acidic elution conditions (0.7% acetic acid, pH 2.8, 2 min) basic isoforms were eluted preferentially from the well, and no isoform discrimination occurred under basic elution conditions (0.4 mol L−1 glycine per 6 mol L−1 urea buffer, pH 11.3). Sample matrix was human urine, which was concentrated by ultrafiltration as described for the IEF method. After the first ultrafiltration step, the retentate obtained (100 μL) was diluted 1:1 with ELISA assay diluent (R&D Systems) and incubated for 1.5 h in the ELISA well. The unbound fraction was removed and the well washed. The bound Epo was first eluted under acidic conditions; the well was then washed again and the remaining bound Epo was subsequently eluted under basic conditions. Both eluates were concentrated by ultrafiltration and volume adjusted with Immulite assay diluent (350 μL final volume). The Immulite Epo immunoassay was then used for quantifying the Epo contained in both fractions. Recovery was determined for both conditions (%A, %B) and a so-called “ratio QA” (quotient between %A of the sample and %A of NIBSC uhEpo as reference compound) calculated for discrimination between endogenous and recombinant erythropoietins. The reference range for blank urine samples (ratio QA 0.57–1.15, mean value 0.86; Fig. 7) was established by measuring 30 samples of volunteers and calculating a 95% confidence interval. The method was tested with spiked urine samples (Mircera, Dynepo, BRP-Epo, NESP) and samples obtained from excretion studies with rhEpo (epoetins alfa and beta). Performance characteristics were similar to those of the IEF method for rhEpo but discriminating power for NESP was lower. The latter was explained by the different elution behaviour of uhEpo obtained from urine and the NIBSC standard (as was apparent from the <1 ratio QA mean value for blank urine samples). The authors concluded that the QA value will be affected similarly to the IEF method by “effort” and unstable (“active”) urine samples. Problematic might be the presence of rhEpo and NESP in the same sample, which might lead to false negative results. Such cases can best be detected by SDS-PAGE or the IEF method.
Fig. 7

Different elution of erythropoietins bound to an ELISA plate. Acidic and basic buffers were used and the ratio of the Epo concentration in both fractions calculated (box plots of ratio QA). Results for (a) CERA, (b) Dynepo, (c) rhEpo standard, (d) uhEpo (NIBSC standard), (e) NESP, (f) blank urine samples, (g) blank urine used for spiking experiments, (hj) blank urine spiked with 71%, 87%, and 90% rhEpo standard, (k, l) urine from Eprex and NeoRecormon excretion studies. Dotted lines represent the limits of the confidence interval (95%) for blank samples. Source: Mallorquì et al. (2010) [74], Elsevier (permission granted)

A screening method for detection of Mircera in serum using a commercial Epo ELISA assay (Access Epo assay; Beckman Coulter) in combination with fractionated protein precipitation was published by Van Maerken et al. (2010) [76]. The method is based on the observation that PEGylation of rhEpo (i.e. Mircera) leads to an increase in water solubility and that this increased solubility can be used for specific enrichment of Mircera contained in serum by precipitation of serum proteins with PEG-6000 [53]. A final PEG-6000 concentration of 25% (w/v) was obtained by mixing 150 μL serum with 150 μL PEG-6000 (50% (w/v); dissolved in 0.15 mol L−1 saline solution). After vortex mixing and incubation at 37 °C (15 min) the samples were vortex mixed again and centrifuged (9300 g for 10 min). The supernatant was collected, diluted with 0.15 mol L−1 saline solution (1:4), and the Epo concentration was measured by use of the Access Epo assay—a magnetic bead-based immunoassay with chemiluminescent detection. Data were normalized by calculating the ratio between the Epo concentration obtained with and without PEG-6000 pretreatment (in the latter case 150 μL 0.15 mol L−1 saline solution was used instead of the 50% PEG-6000 solution). The authors observed a significant increase in the PEG:control ratio for haemodialysis patients (average value 2.15) treated intravenously with Mircera, compared with untreated patients (average value 0.92 for untreated non-renal patients and 0.82 for untreated haemodialysis patients). Additional longitudinal studies showed that the test was able to detect Mircera for two weeks in all the Mircera-treated patients (receiving 50–350 μg Mircera every four weeks) and in most cases during the third and fourth weeks also. It was concluded that the assay might also be useful for screening purposes in doping control, despite the fact it was developed using samples from clinically ill persons.

Lamon et al. (2009) [77] described the development and validation of an ELISA for detection of Mircera doping in human serum. The assay used a streptavidin-precoated microtitre plate for immobilization of a biotinylated polyclonal anti-Epo capture antibody. The coating procedure was preferably performed overnight. After a washing step 100 μL sample was added to the ELISA well and incubated for 2 h at room temperature. The well was washed again and then the bound Mircera was detected by a monoclonal anti-PEG antibody, which was labelled with digoxigenin (DIG; incubation time 1 h). To detect the bound anti-PEG antibody another incubation step was necessary—this time with an anti-DIG-Fab-HRP conjugate and a subsequent colorimetric reaction (2,2′-azinobis[3-ethylbenzothiazoline-6-sulfonic acid] diammonium salt (ABTS) substrate solution; measurement of absorption at 405 nm, with 490 nm as reference wavelength). The lower limit of quantification (LLOQ) of the assay was 30 pg mL−1 and the lower limit of quality control (LLQC) 50 pg mL−1. For determining the cut-off limit of the ELISA (100 pg mL−1) 140 blank serum samples and serum samples from six Mircera administration studies were analysed and an ROC curve calculated. On the basis of this cut-off value the specificity and sensitivity of the assay were 100% and 80%, respectively, for a period of four weeks and a single injection of 200 μg Mircera. Depending on the route of application (intravenous, subcutaneous) and interindividual variability, detection windows between eight days and more than four weeks were obtained. Currently, the test is used by several doping-control laboratories for screening purposes and in combination with the IEF method for confirmation. The ELISA is commercially available from MicroCoat Biotechnologie (Germany). However, alternative strategies include the combination of IEF+IEF-PAGE, Sarcosyl-PAGE [64] +IEF-PAGE, and SDS-PAGE+IEF-PAGE.

A competitive ELISA for the detection of Hematide in plasma samples from test animals (rat, monkey) was published by Fan et al. (Affymax) in 2006 [78]. Sample dilutions were first pre-incubated with a fixed amount of a polyclonal anti-Hematide rabbit antibody and then transferred to microtitre wells, which were coated with Hematide. Only antibodies which were not blocked by the Hematide in the sample were still able to bind to the Hematide on the wells. After a washing step the assay was developed by using an anti-rabbit IgG conjugate. The quantification limit of the ELISA was 31.5 ng mL−1 and it detected both the free and the PEGylated peptides. However, it was not reported whether this assay might also be useful for detection of Hematide in human samples.

Membrane-assisted isoform immunoassay (MAIIA)

An on-line combination of lectin affinity chromatography with lateral flow immunoassay (strip) for differentiation of recombinant and endogenous erythropoietins was developed by Lönnberg et al. [79, 80, 81, 82]. The technology was named “membrane-assisted isoform immunoassay” (MAIIA) [83]. The so-called “EPO WGA MAIIA” test uses wheat germ agglutinin (WGA) in the “separation zone” of the strip—a lectin with specificity for N-acetylglucosamine and N-acetylneuraminic acid-containing glycans and which was already known for its non-selectivity regarding the capture of Epo isoforms [8]. However, the interaction was observed to be stronger for the isoforms of rhEpo than for those of uhEpo, probably because of the greater content of lactosamine units in rhEpo. Desorption of isoforms by use of buffers containing a competing monosaccharide in low concentration (N-acetylglucosamine (GlcNAc); e.g. 2–10 mmol L−1) enabled differentiation of epoetins, including NESP and Mircera [84, 85]. Compared with endogenous Epo (serum, urine) Mircera had a weak interaction with WGA; that of rhEpos was stronger, and that of NESP was strongest [81, 85]. At a GlcNAc concentration of 100 mmol L−1 or above all the isoforms were eluted from the lectin [83]. In addition to the separation zone the EPO WGA MAIIA strip contains a “capture (detection) zone” (Fig. 8) in which the eluted Epo isoforms reacted with an immobilized anti-Epo antibody. The bound Epo was then detected with a second anti-Epo antibody, which was labelled with carbon black nano-strings [86], and which coloured the capture zone from grey to black. The degree of “blackness” was directly proportional to the concentration of the bound Epo. A flatbed scanner was used for quantification [87, 88] and the exact Epo concentration was calculated after background correction (“delta blackness” value) and use of an Epo calibration curve [80, 89]. To distinguish between different Epo pharmaceuticals and endogenous Epos two Epo concentrations had to be measured for each sample—one at a low GlcNAc concentration in the desorption buffer, the other at a high concentration. The so-called “percentage of migrated isoforms” (PMI) is the amount of Epo desorbed at the low GlcNAc concentration and expressed as a percentage of the amount obtained at the high GlcNAc concentration [85]. Typical PMI values for shEpo and uhEpo were, e.g., ca 75%; for rhEpos (Eprex, NeoRecormon, Dynepo, Epomax, NESP, some biosimilar compounds) they were between, e.g., ca 7% and 52%, and for Mircera they were ca 90% [85]. For improved detection of Mircera the authors recommended additional calculation of the so-called “relative analyte migration” value (RAM), which evaluated the interaction profile of Epo with the antibody in the capture zone (“EPO AbQ MAIIA” method and algorithm) [81].
Fig. 8

(A) Main steps in the EPO WGA MAIIA test procedure. (B) Design of the MAIIA strip with WGA lectin zone, for Epo separation, and anti-Epo detection zone. The small inset shows a schematic representation of carbon black nano-strings. Source: (2010) [82], MAIIA Diagnostics (permission granted)

However, the Epo contained in biological samples (e.g. serum, urine) must first be immunoaffinity purified before it can be applied to the EPO WGA MAIIA test strip. For this purpose Lönnberg et al. developed disposable monolith columns (6 μL, 0.15 mm length), which enabled rapid Epo purification [13]. Approximately 20–40 μg of a mouse monoclonal anti-Epo antibody (clone 3 F6) was immobilized on each column. Bound Epo was desorbed under acidic conditions and in a small volume (ca 50 μL) of buffer. Average recoveries of endogenous Epo in urine were ca 65%. Recovery depended on the type of Epo and was lowest for Mircera (ca 30% in urine). The columns were also tested for a possible change of the Epo IEF profile after immunoaffinity purification, but no alteration was observed [14]. The applicability of membrane-assisted isoform immunoassay as a screening procedure in Epo-doping control is currently in the validation phase.

Detection of proteases

The observation that Epo profiles of some urine samples were unstable led to the speculation that enzymes contained in these urine samples might be the reason. Hence, a stability test has to be routinely performed on all Epo IEF-PAGE confirmation analyses (vide supra). A small aliquot of the urine (500 μL) is supplemented with a mixture of protease inhibitors including pepstatin-A (an inhibitor of acid proteases), buffer-exchanged to pH 5.5 with acetate buffer, and spiked with rhEpo and NESP [7]. Any shifts in the Epo profile are indicative of an unstable (“active”) urine (vide supra). The nature of this instability was ascribed to the presence of, e.g., proteases, glycosidases, sulfatases, or neuraminidases (the last are very active under the pH conditions of the test) because of bacterial proliferation during transport of the samples [40, 90]. To inhibit the growth of microorganisms, freezing of urine samples immediately after collection [90] or addition of 1% sodium azide [91] was therefore recommended. High enzymatic activity also led to completely empty IEF profiles in the pH range 2–6, because Epo degradation products had higher pI values. The effect of successive neuraminidase degradation on the IEF profile of rhEpo, which finally leads to asialoerythropoietin, was shown by Imai et al. as early as 1990 [92]. Degradation by neuraminidases also leads to a decrease in molecular mass of Epo, which was demonstrated by Yanagawa et al., by SDS-PAGE, in 1984 [93]. Belalcazar et al. [94] performed controlled digestions of uhEpo, rhEpo, and NESP with neuraminidase and arylsulfatase, and obtained profiles similar to those observed for unstable urine samples. Anielski et al. (2009) [95] showed, by urine spiking experiments, that bacteria causing urinary tract infections (E. coli, K. pneumoniae) were able to alter Epo IEF profiles in a manner similar to that observed for unstable urine samples.

However, undetectable Epo IEF profiles might also be caused by the intentional addition of enzymes to the urine during sample collection by the athlete. Thevis et al. (2007) developed a procedure which enabled detection of such manipulation [96]. A threshold of 15 μg mL−1 protease concentration (measured with a protease activity test) was established, above which further investigations should be made. The threshold was based on the observation that in most tested urine samples endogenous protease activity was below 6 μg mL−1 and that at 20 μg mL−1 (the minimum protease concentration expected after manipulation) urinary proteins were completely degraded. Samples exceeding the threshold were analyzed by SDS-PAGE and Coomassie staining and showed either completely empty lanes or lanes with few bands, which enabled identification of the added proteases by mass spectrometry (either by in-gel digestion of the excised band or by determination of proteolysis and autolysis products of the enzymes [97]). By using the latter approach Kohler et al. [98] were able to report the presence of exogenous bacillolysin in two doping-control samples from elite athletes. Another approach was published by Lamon et al. (2007) [99], who suggested the use of protease-induced digestion of endogenous urinary albumin as a test for protease activity (revealed by Western blotting with an anti-albumin antibody after SDS-PAGE).

To protect the Epo contained in urine samples from enzymatic degradation, Tsivou et al. (2010) proposed addition of a stabilization mixture containing protease inhibitors (including, e.g., PMSF, pepstatin-A, trypsin inhibitor) and antimicrobial substances (sodium azide, pen-strep-fungizone) [100]. As an alternative, Sanchis-Gomar et al. (2010) [101] considered cleaning the urethra by first collecting a small volume of urine and then collecting the urine for the A and B samples—a less expensive alternative for preventing athletes from using proteases.

Mass spectrometric methods

During the past two decades mass spectrometry (MS) of erythropoietin has been mostly focussed on the characterization of recombinant Epos. This has been largely because the amount of commercially available uhEpo has been very limited and its purity low. Consequently, most attempts, which had initially started with the idea of structurally characterizing uhEpo, ended with the characterization of rhEpos—a necessary basis for discovering possible differences from uhEpo. Hence, only a very limited amount of articles addressing these differences has so far appeared in print [102, 103, 104, 105].

Structural characterization of Epo by mass spectrometry can be performed either at the level of the intact glycoprotein or after controlled enzymatic digestion at the level of its glycopepides, peptides, or glycans. Two methods of ionization, matrix-assisted laser desorption ionization (MALDI) and electrospray ionization (ESI), in combination with various types of mass analyzer (e.g. linear or 3D ion traps, orbitrap, triple quadrupoles, time of flight (TOF or TOF/TOF), and hybrid instruments, which combine two types of analyzers) have been most frequently used in these studies. These investigations revealed that human Epo is a glycoprotein consisting of 165 amino acids [103], with disulfide bridges at positions C7–C161 and C29–C33, containing three N-glycans of the bi, tri, and tetraantennary complex type (with or without lactosamine repeating units) at positions N24, N38, and N83, and one O-glycan at position S126 of the core 1 or core 2 type [105, 106]. A varying number of terminal N-acetylneuraminic acids (NeuAc) was found on these glycans, with N-acetylation at position 5 (NeuAc) [107] and—less frequently—additional acetylation, e.g. at position 9 (Neu5,9Ac2) [108]. N-Glycolylneuraminic acid (NeuGc) was found to exist only in recombinant erythropoietins and only in minor amounts (e.g. 2% NeuGc, 98% NeuAc) [107, 108]. Neuraminic acids of rhEpos were exclusively α(2→3) linked to galactose and uhEpo also contained α(2→6) linkages [104]. Fucosylations were mostly found in α(1→6) linkage to the first core N-acetylglucosamine (GlcNAc) [104, 105, 109]. Also shown was the presence of mannose-6-phosphate [110] and sulfated N-acetylglucosamines (GlcNAc) [111, 112, 113]. However, no mass spectrometric method which enables direct detection of doping of human athletes with rhEpos (e.g. epoetins alfa, beta, delta, zeta, theta, omega, biosimilar compounds, or copy epoetins) has yet been developed. Only if natural (i.e. species specific) or artificial changes in the amino acid sequence (as in the human Epo analogue NESP) were present was mass spectrometric detection of doping successful. Guan et al. (2007) [114] was the first to use species-specific differences between human and horse Epo for confirming the presence of rhEpo and NESP in equine plasma by electrospray mass spectrometry (linear ion trap, LTQ) in combination with upstream liquid chromatographic (LC) separation. Epos were first immunoaffinity purified with polyclonal anti-Epo antibodies immobilized on magnetic beads, then digested with trypsin, separated by LC, and then identified after ionization and fragmentation in the mass spectrometer (MS–MS) by collision-induced dissociation (CID) according to their retention time and product ion characteristics. The two tryptic peptides of rhEpo and NESP used (46VNFYAWK52 (T6), 144VYSNFLR150 (T17)) were specific for human Epo. The limits of detection and confirmation were 0.1 ng mL−1 and 0.2 ng mL−1, respectively. Because this method did not enable differentiation between rhEpo and NESP, Guan et al. (2008) [115] developed a slightly modified version of the method using the deglycosylated tryptic peptides T5 (rhEpo; 21EAENITTGCAEHCSLNENITVPDTK45) and T9 (NESP; 77GQALLVNSSQVNETLQLHVDK97) for identification. The method enabled detection of these human Epos in horse plasma with limits of detection and identification of 0.05 ng mL−1 (NESP) and 0.1 ng mL−1 (rhEpo), and 0.1 ng mL−1 (NESP) and 0.2 ng mL−1 (rhEpo), respectively. In 2009 Guan et al. [116] adapted this method for detection of NESP in human plasma using the deglycosylated tryptic T9 peptide of NESP (Fig. 9). The limits of detection and identification achieved were 0.1 ng mL−1 and 0.2 ng mL−1, respectively.
Fig. 9

Detection of NESP in human plasma samples by mass spectrometry. Fragment ion spectra of deglycosylated tryptic T9 peptides of NESP (A) and rhEpo (B) are shown. Whereas shEpo and rhEpo cannot be distinguished from each other, differences in the amino acid sequence of NESP enable discrimination. Source: Guan et al. (2009) [116], Thieme Medical Publishers (permission granted)

In 2010 Yu et al. [117] reported a slightly modified version of the 2007 procedure of Guan et al. in which the T6 peptide (vide supra) was used to confirm the presence of rhEpo, NESP, and Mircera in equine plasma. A triple-quadrupole ion-trap mass spectrometer (QTrap 4000) run in selected reaction monitoring (SRM) mode was used in combination with nano-LC. Four SRM transitions of the doubly positively charged m/z 465.5 precursor ion were used for identification (m/z 465.5→214.4, 714.8, 811.9, and 828.7). The limits of confirmation obtained were 0.1, 0.2, and 1.0 ng mL−1 for rhEpo, NESP, and Mircera. Chang et al. (2010) [118] combined an ELISA method (based on the method published by Van Maerken et al.; vide supra) with the SRM assay of Yu et al. for screening and confirmation of Mircera in horse plasma. Guan et al. (2010) [119] was able to further increase the sensitivity of his method for detection of Epo doping in horses by:
  1. 1.

    using the tryptic T8 Epo peptide (54MEVGQQAVEVWQGLALLSEAVLR76) instead of the T6, T9, and T17 peptides used in the earlier publications; and

  2. 2.

    introducing a PEG-6000 precipitation step before immunoaffinity purification of the plasma.


The T8 peptide is common to rhEpo, NESP, and Mircera, and hence does not enable differentiation between these three Epo pharmaceuticals. The limits of detection and identification obtained were 0.1/0.05/0.3 ng mL−1 and 0.2/0.1/0.5 ng mL−1 for rhEpo/NESP/Mircera, respectively.

In 1996 Wrighton et al. [120] discovered, by phage display methodology, that small peptide molecules without sequence relation to Epo can also interact with the Epo receptor (Epo-R) causing the same receptor dimerization and signal transduction as observed for Epo. The peptides, called Epo mimetic peptides (EMP), were cyclic, because of an internal disulfide bond, and contained a fourteen amino acid minimum consensus sequence. One of these peptides (EMP1; 1GGTYSCHFGPLTWVCKPQGG20; the disulfide bridge is between C6 and C15, conserved amino acids are underlined) was further investigated. It was shown that covalent dimerization (either by a two-amino-acid linker via the C-termini [121] or by a bifunctional PEG-linker via the N-termini [122]) led to increased affinity for Epo-R. Hematide (Peginesatide, AF-37702; Affymax and Takeda Pharmaceutical) was developed on the basis of these discoveries and is currently in clinical phase 3. The chemical structure has been published by several authors [123, 124, 125]. According to these data Hematide is an EMP dimer containing three unusual amino acids (N-acetylglycine, 1-naphthylalanine, and N-methylglycine (sarcosine) at positions 1, 13, and 20, respectively) and two 20-kDa PEG groups. Doses used in clinical trials were in the range of ca 0.025 to 0.1 mg kg−1 every four weeks [126, 127]. Because of the large amount of PEGylation, doping with Hematide is probably best detected in serum or plasma. On the basis of the unusual amino acids and the rather high doses, mass spectrometric detection of Hematide for doping control purposes should be feasible. Mass spectrometric data for the monomeric peptide were published by Thevis et al. in 2009 [128].

For the immunological detection of Hematide refer to the section on ELISA-based methods.

Athlete biological passport

The concept of the biological passport is based on the observation that Epo doping:
  1. 1.

    changes haematological data; and

  2. 2.

    that these changes are best monitored individually and longitudinally for each athlete.

In contrast with the methods described above, the passport uses markers for altered erythropoiesis instead of directly detecting the recombinant hormone—hence is a multivariable indirect detection method. Apart from changes in the haematological profile (“haematological module” of the passport) the concept is open for inclusion of profiles of other drugs (e.g. steroid markers, “endocrine module”). The idea behind the passport is that the haematological profile should enable longer and more efficient (by targeted testing of athletes) detection of Epo doping than non-targeted direct methods, which suffer from the relatively short detection window, especially when microdoses of rhEpo are used [42]. A guideline regarding the implementation of the haematological module was published by WADA in 2010 [129]. It describes the scientific basis, test implementation, administration, and documentation of the biological passport. It also includes four technical documents, which are mandatory for anti-doping organizations using the passport. Nine markers of erythropoiesis have to be measured and longitudinally recorded: haematocrit (HCT), haemoglobin (HGB), red blood cells count (RBC), percentage of reticulocytes (RET%), reticulocytes count (RET#), mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH), mean corpuscular haemoglobin concentration (MCHC), and index of stimulation (OFF-hr score) [129]. The last marker was published in 2003 (OFF-hr=HGB−60 (RET%)1/2) [130] and was part of the so-called ON and OFF models which were developed starting in 2000 (Sydney Olympic Games) for the indirect detection of Epo doping [5, 130, 131]. Mandatory requirements regarding blood collection, transport, analysis, and results management are detailed in the technical documents. The profiles are individually and longitudinally evaluated for each athlete (thus reducing the higher inter-individual variability of blood data and generating individual limits) by a so-called adaptive model, which uses a Bayesian statistical approach (Fig. 10) [132, 133, 134]. Abnormal profile data have to be reviewed by an expert panel and may lead to discovery of an anti-doping rule violation.
Fig. 10

Longitudinal profiles of (A) haemoglobin (HGB) and (B) OFF score as used for the Athlete Biological Passport. Upper and lower lines represent the threshold limit values calculated by the Bayesian statistical model. The centre line shows the measurement results for the athlete’s blood samples. Source: Sottas et al. (2010) [134], Springer (permission granted)

Detection of EPO gene doping

Gene doping is a prohibited method according to article M3 of the WADA “World Anti-Doping Code—The 2011 Prohibited List” [6]. It would be also covered by article S0 (“non-approved substances”), in case the substances used have not been approved for human therapeutic use (e.g. substances under clinical development). By definition, gene doping comprises not only the “transfer of nucleic acids or nucleic sequences” but also the “use of normal or genetically modified cells” and the use of agents which alter gene expression (article M3). Current methods and strategies for detection of gene doping have recently been reviewed [135, 136, 137] and the Epo gene identified as one of the main target genes for gene doping. However, so far no clinically approved gene therapy method which involves the Epo gene is available. Repoxygen, an Epo gene therapy method developed by Oxford BioMedica several years ago, has not yet entered clinical trial phases. A viral vector system was used for delivery of the Epo gene to muscle tissue. To assess the detectability of this type of gene therapy, Lasne et al. [138] compared the serum Epo IEF profiles of macaques before and after Epo cDNA treatment using a recombinant adeno-associated virus for in vivo gene transfer. A significant shift of the IEF profiles to more basic isoforms was found after the treatment—similar to that observed in humans after treatment with rhEpo. The results were explained by a tissue specific expression of Epo isoforms and were later confirmed for retina tissue [139] and also for ex vivo gene therapy which used hepatocytes and a liver-specific lentiviral vector system (Fig. 11) [140].
Fig. 11

IEF-PAGE of macaque serum Epo (A) before (NT, non-transplanted; monkey not specified) and (B–E) after transplantation of lentivirally transduced hepatocytes. (B) monkey #2 on day 2, (CE) monkey #1 (E), #2 (C), and #3 (D) on day 8 after the transplantation. Source: Menzel et al. (2009) [140], Nature Publishing Group (permission granted)

Aside from attempting to directly detect the transgenic protein and possible differences by comparison with the non-transgenic protein, an alternative strategy is the direct detection of the transgenes, preferably not in the treated tissue but in blood. Beitler et al. [141] used the observation that transgenes usually lack the intron sequences normally present in human genes. A so-called single-copy primer-internal intron-spanning PCR method (spiPCR) was developed. The primers were designed in a way that their 3′ and 5′ ends only hybridize with two neighbouring exons, which are not separated by introns. Consequently, PCR amplification occurs only if the transgene is present. The method was developed to screen for six prime candidate genes (including Epo and vascular endothelial growth factors) in blood and was tested in a mouse model for vascular endothelial growth factor A (VEGF-A) [142]. The transgene was delivered intramuscularly by an adeno-associated virus system. Only 20 μL blood was required and this enabled detection of the gene therapy for up to 56 days. A similar approach was reported by Baoutina et al. (2010) [143].

Apart from gene transfer, Epo gene expression also can be regulated by small orally administered molecules (e.g. transcription factor stabilizers). Prolylhydroxylase inhibitors (e.g. FG-2216, FG-4592; FibroGen; currently clinical phase 2 studies have been halted) stabilize hypoxia-inducible factor 1 (HIF-1α subunit) by preventing its enzymatic hydroxylation and subsequent degradation and thus increase Epo gene expression. Direct detection of the inhibitors might be possible by mass spectrometry [128].


Erythropoiesis-stimulating agents have been continually developed further since the first approval of recombinant Epo alfa for clinical use in 1989. This evolution fostered the development of new methods for detection of ESA doping and adaptation of existing methods to an ever changing situation. Recent advances have been made in the direct detection of Epo by gel electrophoretic methods, with the introduction of molecular mass-based methods (SDS-PAGE, SAR-PAGE) in screening and confirmation procedures. This strategy was especially useful for obtaining evidence of the abuse of Dynepo, NESP, and Mircera from urine and blood samples. Because of its increased sensitivity for Mircera, SAR-PAGE should preferably be used for serum and plasma samples instead of SDS-PAGE (one matrix one method approach). However, both methods are ideally suitable for confirming the presence of unstable (“active”) and “effort” urine samples as detected by IEF-PAGE. Targeted screening for the abuse of Mircera has been made possible with the development of an ELISA method for CERA. Another innovative method is the so-called “EPO WGA MAIIA” test, which is currently under evaluation for use as a screening method for the abuse of many different Epos. Adulteration of urine samples during collection by addition of proteases by the athlete has resulted in the development of procedures for protease detection and identification. The applicability of mass spectrometry for direct detection of NESP in human plasma and for rhEPO, NESP, and Mircera in horse plasma has been demonstrated in a series of publications between 2007 and 2010. However, these methods are not useful for obtaining evidence of the presence of rhEpo in human plasma. Because of the short serum half-life of rhEpos, indirect (marker) methods using blood data have been developed. The “Athlete Biological Passport” (haematological module) is the latest development in multivariable and longitudinal indirect testing. Its Bayesian statistical approach enables calculation of athlete-specific blood data limits and hence more efficient and targeted testing. Because of its open concept the passport is ready for inclusion of future advances in the detection of ESA doping by alternative methods, for example transcriptomics, proteomics, and metabolomics.


  1. 1.
    Lin FK, Suggs S, Lin CH, Browne JK, Smalling R, Egrie JC, Chen KK, Fox GM, Martin F, Stabinsky Z, Badrawi SM, Lai PH, Goldwasser E (1985) Cloning and expression of the human erythropoietin gene. Proc Natl Acad Sci USA 82:7580–7584CrossRefGoogle Scholar
  2. 2.
    Israels LG, Israels ED (2003) Erythropoiesis: an overview. In: Molineux G, Foote MA, Elliott SG (eds) Erythropoietins and erythropoiesis, 1st edn. Birkhäuser, BaselGoogle Scholar
  3. 3.
    Catlin DH, Hatton CK, Lasne F (2003) Abuse of recombinant erythropoietins by athletes. In: Molineux G, Foote MA, Elliott SG (eds) Erythropoietins and erythropoiesis, 1st edn. Birkhäuser, BaselGoogle Scholar
  4. 4.
    Lasne F, de Ceaurriz J (2000) Recombinant erythropoietin in urine. Nature 405:635CrossRefGoogle Scholar
  5. 5.
    Parisotto R, Gore CJ, Emslie KR, Ashenden MJ, Brugnara C, Howe C, Martin DT, Trout GJ, Hahn AG (2000) A novel method utilising markers of altered erythropoiesis for the detection of recombinant human erythropoietin abuse in athletes. Haematologica 85:564–572Google Scholar
  6. 6.
    World Anti-Doping Agency (2010) The World Anti-Doping Code. The 2011 Prohibited List. International Standard, Montreal. Accessed 10 January 2011
  7. 7.
    World Anti-Doping Agency (2009) Harmonization of the method for the identification of recombinant erythropoietins (i.e. epoetins) and analogues (e.g. darbepoetin and methoxypolyethylene glycol-epoetin beta), Montreal. Accessed 6 January 2011
  8. 8.
    Lasne F, Martin L, Crepin N, de Ceaurriz J (2002) Detection of isoelectric profiles of erythropoietin in urine: differentiation of natural and administered recombinant hormones. Anal Biochem 311:119–126CrossRefGoogle Scholar
  9. 9.
    Wide L, Bengtsson C (1990) Molecular charge heterogeneity of human serum erythropoietin. Br J Haematol 76:121–127CrossRefGoogle Scholar
  10. 10.
    Tam RC, Coleman SL, Tiplady RJ, Storring PL, Cotes PM (1991) Comparisons of human, rat and mouse erythropoietins by isoelectric focusing: differences between serum and urinary erythropoietins. Br J Haematol 79:504–511CrossRefGoogle Scholar
  11. 11.
    Wide L, Bengtsson C, Berglund B, Ekblom B (1995) Detection in blood and urine of recombinant erythropoietin administered to healthy men. Med Sci Sports Exerc 27:1569–1576Google Scholar
  12. 12.
    Lasne F, Martin L, Martin JA, de Ceaurriz J (2007) Isoelectric profiles of human erythropoietin are different in serum and urine. Int J Biol Macromol 41:354–357CrossRefGoogle Scholar
  13. 13.
    Lönnberg M, Dehnes Y, Drevin M, Garle M, Lamon S, Leuenberger N, Quach T, Carlsson J (2010) Rapid affinity purification of erythropoietin from biological samples using disposable monoliths. J Chromatogr A 1217:7031–7037CrossRefGoogle Scholar
  14. 14.
    Dehnes Y, Lamon S, Lönnberg M (2010) Erythropoietin (EPO) immunoaffinity columns - a powerful tool for purifying EPO and its recombinant analogues. J Pharm Biomed Anal 53:1028–1032CrossRefGoogle Scholar
  15. 15.
    Lasne F (2001) Double-blotting: a solution to the problem of non-specific binding of secondary antibodies in immunoblotting procedures. J Immunol Meth 253:125–131CrossRefGoogle Scholar
  16. 16.
    Lasne F (2003) Double-blotting: a solution to the problem of nonspecific binding of secondary antibodies in immunoblotting procedures. J Immunol Meth 276:223–226CrossRefGoogle Scholar
  17. 17.
    Hammerl P, Hartl A, Freund J, Thalhamer J (1992) A method for the detection of serologically crossreacting antigens both within and between protein mixtures: the western cross blot. J Immunol Meth 151:299–306CrossRefGoogle Scholar
  18. 18.
    Bajla I, Holländer I, Gmeiner G, Reichel Ch (2005) Quantitative analysis of images in erythropoietin doping control. Med Biol Eng Comput 43:403–409CrossRefGoogle Scholar
  19. 19.
    Bajla I, Holländer I, Minichmayr M, Gmeiner G, Reichel Ch (2005) GASepo - a software solution for quantitative analysis of digital images in Epo doping control. Comput Meth Programs Biomed 80:246–270CrossRefGoogle Scholar
  20. 20.
    Stolc S, Bajla I, 4 (2006) Improvement of band segmentation in Epo images via column shift transformation with cost functions. Med Biol Eng Comput 44:257–274, Erratum in: Med Biol Eng Comput 44:605–7CrossRefGoogle Scholar
  21. 21.
    Segura J, Pascual JA, Gutiérrez-Gallego R (2007) Procedures for monitoring recombinant erythropoietin and analogues in doping control. Anal Bioanal Chem 388:1521–1529CrossRefGoogle Scholar
  22. 22.
    Catlin DH, Breidbach A, Elliott S, Glaspy J (2002) Comparison of the isoelectric focusing patterns of darbepoetin alfa, recombinant human erythropoietin, and endogenous erythropoietin from human urine. Clin Chem 48:2057–2059Google Scholar
  23. 23.
    Schellekens H (2004) Biosimilar epoetins: how similar are they? EJHP 2004(3):43–47Google Scholar
  24. 24.
    Belalcazar V, Ventura R, Segura J, Pascual JA (2008) Clarification on the detection of epoetin delta and epoetin omega using isoelectric focusing. Am J Hematol 83:754CrossRefGoogle Scholar
  25. 25.
    Reichel C, Kulovics R, Jordan V, Watzinger M, Geisendorfer T (2009) SDS-PAGE of recombinant and endogenous erythropoietins: benefits and limitations of the method for application in doping control. Drug Test Anal 1:43–50CrossRefGoogle Scholar
  26. 26.
    Macdougall IC, Ashenden M (2009) Current and upcoming erythropoiesis-stimulating agents, iron products, and other novel anemia medications. Adv Chron Kidney Dis 16:117–130CrossRefGoogle Scholar
  27. 27.
    Lasne F, Martin L, Martin JA, de Ceaurriz J (2009) Detection of continuous erythropoietin receptor activator in blood and urine in anti-doping control. Haematologica 94:888–890CrossRefGoogle Scholar
  28. 28.
    Reichel C (2008) Identification of zinc-alpha-2-glycoprotein binding to clone AE7A5 antihuman EPO antibody by means of nano-HPLC and high-resolution high-mass accuracy ESI-MS/MS. J Mass Spectrom 43:916–923CrossRefGoogle Scholar
  29. 29.
    Rabin OP, Lasne F, Pascual JA, Saugy M, Delbeke FJ, Van Eenoo P (2006) New urinary EPO drug testing method using two-dimensional gel electrophoresis. Clin Chim Acta 373:186–187CrossRefGoogle Scholar
  30. 30.
    Lasne F (2006) No doubt about the validity of the urine test for detection of recombinant human erythropoietin. Blood 108:1778–1779CrossRefGoogle Scholar
  31. 31.
    Khan A, Grinyer J, Truong ST, Breen EJ, Packer NH (2005) New urinary EPO drug testing method using two-dimensional gel electrophoresis. Clin Chim Acta 358:119–130CrossRefGoogle Scholar
  32. 32.
    Khan A, Baker MS (2007) Non-specific binding of monoclonal human erythropoietin antibody AE7A5 to Escherichia coli and Saccharomyces cerevisiae proteins. Clin Chim Acta 379:173–175CrossRefGoogle Scholar
  33. 33.
    Franke WW, Heid H (2006) Pitfalls, errors and risks of false-positive results in urinary EPO drug tests. Clin Chim Acta 373:189–190CrossRefGoogle Scholar
  34. 34.
    Beullens M, Delanghe JR, Bollen M (2006) False-positive detection of recombinant human erythropoietin in urine following strenuous physical exercise. Blood 107:4711–4713CrossRefGoogle Scholar
  35. 35.
    Rabin OP, Lasne F, Pascual JA, Saugy M, Delbeke FJ, Van Eenoo P (2006) New urinary EPO drug testing method using two-dimensional gel electrophoresis. Clin Chim Acta 373:186–187CrossRefGoogle Scholar
  36. 36.
    Catlin D, Green G, Sekera M, Scott P, Starcevic B (2006) False-positive Epo test concerns unfounded. Blood 108:1778CrossRefGoogle Scholar
  37. 37.
    Lasne F (2006) No doubt about the validity of the urine test for detection of recombinant human erythropoietin. Blood 108:1778–1779CrossRefGoogle Scholar
  38. 38.
    Segura J, Pascual JA, Gutiérrez-Gallego R (2007) Procedures for monitoring recombinant erythropoietin and analogues in doping control. Anal Bioanal Chem 388:1521–1529CrossRefGoogle Scholar
  39. 39.
    Lamon S, Martin L, Robinson N, Saugy M, Ceaurriz J, Lasne F (2009) Effects of exercise on the isoelectric patterns of erythropoietin. Clin J Sport Med 19:311–315CrossRefGoogle Scholar
  40. 40.
    Lasne F (2009) Erythropoietin doping – detection in urine. In: Fourcroy JL (ed) Pharmacology, doping and sports. Routledge, LondonGoogle Scholar
  41. 41.
    Lasne F, Thioulouse J, Martin L, de Ceaurriz J (2007) Detection of recombinant human erythropoietin in urine for doping analysis: interpretation of isoelectric profiles by discriminant analysis. Electrophoresis 28:1875–1881CrossRefGoogle Scholar
  42. 42.
    Lamon S, Boccard J, Sottas PE, Glatz N, Wuerzner G, Robinson N, Saugy M (2010) IEF pattern classification-derived criteria for the identification of epoetin-delta in urine. Electrophoresis 31:1918–1924CrossRefGoogle Scholar
  43. 43.
    Ashenden M, Varlet-Marie E, Lasne F, Audran M (2006) The effects of microdose recombinant human erythropoietin regimens in athletes. Haematologica 91:1143–1144Google Scholar
  44. 44.
    Lasne F, Martin L, Martin JA (2010) A fast preparative method for detection of recombinant erythropoietin in blood samples. Drug Test Anal 2:494–495CrossRefGoogle Scholar
  45. 45.
    Reichel C, Benetka E, Geisendorfer T, Scheiblhofer V, Abzieher F (2010) The interference of heparin on IEF-PAGE of erythropoietins. Drug Test Anal 2:568–575CrossRefGoogle Scholar
  46. 46.
    Catlin DH, Hatton CK (2009) Abuse of recombinant erythropoietins and blood products by athletes. In: Elliott SG, Foote MA, Molineux G (eds) Erythropoietins, erythropoietic factors and erythropoiesis, 2nd edn. Birkhäuser, BaselGoogle Scholar
  47. 47.
    Reichel C (2010) Practicing IEF-PAGE of EPO: the impact of detergents and sample application methods on analytical performance in doping control. Drug Test Anal 2:603–619CrossRefGoogle Scholar
  48. 48.
    Lasne F, Popot MA, Varlet-Marie E, Martin L, Martin JA, Bonnaire Y, Audran M, de Ceaurriz J (2005) Detection of recombinant epoetin and darbepoetin alpha after subcutaneous administration in the horse. J Anal Toxicol 29:835–837Google Scholar
  49. 49.
    Bartlett C, Clancy GJ, Cowan DA, Healy JF (2006) Detection of the administration of human erythropoietin (HuEPO) to canines. J Anal Toxicol 30:663–669Google Scholar
  50. 50.
    Catlin DH, Maylin GA, Benchaar S, Neades SM, Timmons MS, McKeever KH (2009) Detection of pegylated epoetin β in horse plasma after intravenous administration. Comp Exerc Physiol 6:105–108CrossRefGoogle Scholar
  51. 51.
    Kung CK, Goldwasser E (1997) A probable conformational difference between recombinant and urinary erythropoietins. Proteins 28:94–98CrossRefGoogle Scholar
  52. 52.
    Masuda S, Okano M, Yamagishi K, Nagao M, Ueda M, Sasaki R (1994) A novel site of erythropoietin production. Oxygen-dependent production in cultured rat astrocytes. J Biol Chem 269:19488–19493Google Scholar
  53. 53.
    Skibeli V, Nissen-Lie G, Torjesen P (2001) Sugar profiling proves that human serum erythropoietin differs from recombinant human erythropoietin. Blood 98:3626–3634CrossRefGoogle Scholar
  54. 54.
    Skibeli V, Nissen-Lie G (2003) Immunomagnetic extraction as a tool for the analysis of human serum erythropoietin. DYNALogue 1:13–15Google Scholar
  55. 55.
    Desharnais P, Ayotte C (2007) Towards the screening of urinary Epo by SDS-PAGE analysis. In: Schänzer et al (eds) Recent advances in doping analysis (15). Sportverlag Strauss, CologneGoogle Scholar
  56. 56.
    Egrie JC, Dwyer E, Browne JK, Hitz A, Lykos MA (2003) Darbepoetin alfa has a longer circulating half-life and greater in vivo potency than recombinant human erythropoietin. Exp Hematol 31:290–299CrossRefGoogle Scholar
  57. 57.
    Balaguer E, Neusüss C (2006) Intact glycoform characterization of erythropoietin-alfa and erythropoietin-beta by CZE-ESI-TOF-MS. Chromatographia 64:351–357CrossRefGoogle Scholar
  58. 58.
    Reichel C, Gmeiner G (2010) Erythropoietins and analogues. In: Thieme D, Hemmersbach P (eds) Doping in sports (Handbook of Experimental Pharmacology 195). Springer, BerlinGoogle Scholar
  59. 59.
    Reichel C, Kulovics R, Jordan V, Watzinger M, Geisendorfer T (2009) SDS-PAGE of recombinant and endogenous erythropoietins: benefits and limitations of the method for application in doping control. Drug Test Anal 1:43–50CrossRefGoogle Scholar
  60. 60.
    Kohler M, Ayotte C, Desharnais P, Flenker U, Lüdke S, Thevis M, Völker-Schänzer E, Schänzer W (2007) Discrimination of recombinant and endogenous urinary erythropoietin by calculating relative mobility values from SDS gels. Int J Sports Med 29:1–6CrossRefGoogle Scholar
  61. 61.
    Voss S, Lüdke A, Romberg S, Schänzer E, Flenker U, deMarees M, Achtzehn S, Mester J, Schänzer W (2010) Effects of high intensity exercise on isoelectric profiles and SDS-PAGE mobility of erythropoietin. Int J Sports Med 31:367–371CrossRefGoogle Scholar
  62. 62.
    Gmeiner G, Reichel C, Kulovics R, Scheiblhofer V (2010) Defending Dynepo detection. Drug Test Anal 2:526–532CrossRefGoogle Scholar
  63. 63.
    Llop E, Gutiérrez-Gallego R, Segura J, Mallorquí J, Pascual JA (2008) Structural analysis of the glycosylation of gene-activated erythropoietin (epoetin delta, Dynepo). Anal Biochem 383:243–254CrossRefGoogle Scholar
  64. 64.
    Reichel C, Abzieher F, Geisendorfer T (2009) SARCOSYL-PAGE: a new method for the detection of Mircera- and EPO-doping in blood. Drug Test Anal 1:494–504CrossRefGoogle Scholar
  65. 65.
    Schlags W, Lachmann B, Walther M, Kratzel M, Noe CR (2002) Two-dimensional electrophoresis of recombinant human erythropoietin: a future method for the European Pharmacopoeia? Proteomics 2:679–682CrossRefGoogle Scholar
  66. 66.
    Caldini A, Moneti G, Fanelli A, Bruschettini A, Mercurio S, Pieraccini G, Cini E, Ognibene A, Luceri F, Messeri G (2003) Epoetin alpha, epoetin beta and darbepoetin alfa: two-dimensional gel electrophoresis isoforms characterization and mass spectrometry analysis. Proteomics 3:937–941CrossRefGoogle Scholar
  67. 67.
    Llop E, Gallego RG, Belalcazar V, Gerwig GJ, Kamerling JP, Segura J, Pascual JA (2007) Evaluation of protein N-glycosylation in 2-DE: erythropoietin as a study case. Proteomics 7:4278–4291CrossRefGoogle Scholar
  68. 68.
    Khan A, Grinyer J, Truong ST, Breen EJ, Packer NH (2005) New urinary EPO drug testing method using two-dimensional gel electrophoresis. Clin Chim Acta 358:119–130CrossRefGoogle Scholar
  69. 69.
    Görg A, Drews O, Lück C, Weiland F, Weiss W (2009) 2-DE with IPGs. Electrophoresis 30(Suppl 1):S122–S132CrossRefGoogle Scholar
  70. 70.
    Wide L, Wikström B, Eriksson K (2003) A new principle suggested for detection of darbepoetin-alpha (NESP) doping. Ups J Med Sci 108:229–238CrossRefGoogle Scholar
  71. 71.
    Egrie JC, Browne JK (2001) Development and characterization of novel erythropoiesis stimulating protein (NESP). Nephrol Dial Transplant 16(Suppl 3):3–13Google Scholar
  72. 72.
    Giménez E, de Bolós C, Belalcazar V, Andreu D, Borrás E, De la Torre BG, Barbosa J, Segura J, Pascual JA (2007) Anti-EPO and anti-NESP antibodies raised against synthetic peptides that reproduce the minimal amino acid sequence differences between EPO and NESP. Anal Bioanal Chem 388:1531–1538CrossRefGoogle Scholar
  73. 73.
    Giampaoli S, Facciabene A, Mennuni C (2003) A protocol to guide development of a sensitive ELISA for canine erythropoietin. Vet Clin Pathol 32:199–201CrossRefGoogle Scholar
  74. 74.
    Mallorquí J, Gutiérrez-Gallego R, Segura J, de Bolòs C, Pascual JA (2010) New screening protocol for recombinant human erythropoietins based on differential elution after immunoaffinity purification. J Pharm Biomed Anal 51:255–259CrossRefGoogle Scholar
  75. 75.
    Mallorquí J, Llop E, de Bolòs C, Gutiérrez-Gallego R, Segura J, Pascual JA (2010) Purification of erythropoietin from human plasma samples using an immunoaffinity well plate. J Chromatogr B Anal Technol Biomed Life Sci 878:2117–2122CrossRefGoogle Scholar
  76. 76.
    Van Maerken T, Dhondt A, Delanghe JR (2010) A rapid and simple assay to determine pegylated erythropoietin in human serum. J Appl Physiol 108:800–803CrossRefGoogle Scholar
  77. 77.
    Lamon S, Giraud S, Egli L, Smolander J, Jarsch M, Stubenrauch KG, Hellwig A, Saugy M, Robinson N (2009) A high-throughput test to detect C.E.R.A. doping in blood. J Pharm Biomed Anal 50:954–958CrossRefGoogle Scholar
  78. 78.
    Fan Q, Leuther KK, Holmes CP, Fong KL, Zhang J, Velkovska S, Chen MJ, Mortensen RB, Leu K, Green JM, Schatz PJ, Woodburn KW (2006) Preclinical evaluation of Hematide, a novel erythropoiesis stimulating agent, for the treatment of anemia. Exp Hematol 34:1303–1311CrossRefGoogle Scholar
  79. 79.
    Lönnberg M, Carlsson J (2006) Lab-on-a-chip technology for determination of protein isoform profiles. J Chromatogr A 1127:175–182CrossRefGoogle Scholar
  80. 80.
    Lönnberg M, Drevin M, Carlsson J (2008) Ultra-sensitive immunochromatographic assay for quantitative determination of erythropoietin. J Immunol Meth 339:236–244CrossRefGoogle Scholar
  81. 81.
    Carlsson J, Lönnberg M, Garle M (unknown) Membrane Assisted Isoform Immunoassay (MAIIA I) a unique method for rapid detection of rhEPO in doping, Montreal. Accessed 9 January 2011
  82. 82.
    MAIIA Diagnostics (2010) Rapid EPO MAIIA doping test, Upsala. Accessed 9 January 2011
  83. 83.
    Lönnberg M, Carlsson J (2000) Membrane assisted isoform immunoassay. A rapid method for the separation and determination of protein isoforms in an integrated immunoassay. J Immunol Methods 246:25–36CrossRefGoogle Scholar
  84. 84.
    Franco Fraguas L, Carlsson J, Lönnberg M (2008) Lectin affinity chromatography as a tool to differentiate endogenous and recombinant erythropoietins. J Chromatogr A 1212:82–88CrossRefGoogle Scholar
  85. 85.
    Carlsson J, Garle M, Lönnberg M (2008) MAIIA method for EPO detection, Montreal. Accessed 9 January 2011
  86. 86.
    MAIIA Diagnostics (2010) Carbon Black Nano-Strings, Upsala.,%20100%20nm.pdf. Accessed 9 January 2011
  87. 87.
    Lönnberg M, Carlsson J (2001) Quantitative detection in the attomole range for immunochromatographic tests by means of a flatbed scanner. Anal Biochem 293:224–231CrossRefGoogle Scholar
  88. 88.
    MAIIA Diagnostics (2010) Scanner Detection Instruction, Upsala. Accessed 9 January 2011
  89. 89.
    MAIIA Diagnostics (2010) EPO Quantification Kit, Upsala. Accessed 9 January 2011
  90. 90.
    Lasne F, Martin L, de Ceaurriz J (2005) Active urine and detection of recombinant erythropoietin. In: Schänzer et al (eds) Recent advances in doping analysis (13). Sportverlag Strauss, CologneGoogle Scholar
  91. 91.
    Belalcazar V, Pascual JA, Abanades S, Farré M, de la Torre R (2005) Stability of EPO in urine stabilized by sodium azide. In: Schänzer et al (eds) Recent advances in doping analysis (13). Sportverlag Strauss, CologneGoogle Scholar
  92. 92.
    Imai N, Higuchi M, Kawamura A, Tomonoh K, Oh-Eda M, Fujiwara M, Shimonaka Y, Ochi N (1990) Physicochemical and biological characterization of asialoerythropoietin. Suppressive effects of sialic acid in the expression of biological activity of human erythropoietin in vitro. Eur J Biochem 194:457–462CrossRefGoogle Scholar
  93. 93.
    Yanagawa S, Hirade K, Ohnota H, Sasaki R, Chiba H, Ueda M, Goto M (1984) Isolation of human erythropoietin with monoclonal antibodies. J Biol Chem 259:2707–2710Google Scholar
  94. 94.
    Belalcazar V, Gutiérrez Gallego R, Llop E, Segura J, Pascual JA (2006) Assessing the instability of the isoelectric focusing patterns of erythropoietin in urine. Electrophoresis 27:4387–4395CrossRefGoogle Scholar
  95. 95.
    Anielski P, Schwenke D, Monecke S, Jacobs E, Thieme D (2009) Bacterial proteases in urine samples – effects on EPO analysis? In: Schänzer et al (eds) Recent advances in doping analysis (17). Sportverlag Strauss, CologneGoogle Scholar
  96. 96.
    Thevis M, Maurer J, Kohler M, Geyer H, Schänzer W (2007) Proteases in doping control analysis. Int J Sports Med 28:545–549CrossRefGoogle Scholar
  97. 97.
    Thomas A, Kohler M, Walpurgis K, Schänzer W, Thevis M (2009) Proteolysis and autolysis of proteases and the detection of degradation products in doping control. Drug Test Anal 1:81–86CrossRefGoogle Scholar
  98. 98.
    Kohler M, Thomas A, Geyer H, Horta L, Schänzer W, Thevis M (2009) Detection of the protease bacillolysin in doping-control urine samples. Drug Test Anal 1:143–145CrossRefGoogle Scholar
  99. 99.
    Lamon S, Robinson N, Sottas PE, Henry H, Kamber M, Mangin P, Saugy M (2007) Possible origins of undetectable EPO in urine samples. Clin Chim Acta 385:61–66CrossRefGoogle Scholar
  100. 100.
    Tsivou M, Dimopoulou HA, Leontiou IP, Georgakopoulos DG, Koupparis MA, Atta-Politou J, Georgakopoulos CG (2010) Stabilization of human urine doping control samples: III. Recombinant human erythropoietin. Clin Chim Acta 411:448–452CrossRefGoogle Scholar
  101. 101.
    Sanchis-Gomar F, Martinez-Bello VE, Nascimento AL, Gomez-Cabrera MC, Viña J (2010) Alternate methods to prevent protease use as a masking agent in sport. J Sci Med Sport 13:473–474CrossRefGoogle Scholar
  102. 102.
    Sasaki H, Bothner B, Dell A, Fukuda M (1987) Carbohydrate structure of erythropoietin expressed in Chinese hamster ovary cells by a human erythropoietin cDNA. J Biol Chem 262:12059–12076Google Scholar
  103. 103.
    Recny MA, Scoble HA, Kim Y (1987) Structural characterization of natural human urinary and recombinant DNA-derived erythropoietin. Identification of des-arginine 166 erythropoietin. J Biol Chem 262:17156–17163Google Scholar
  104. 104.
    Takeuchi M, Takasaki S, Miyazaki H, Kato T, Hoshi S, Kochibe N, Kobata A (1988) Comparative study of the asparagine-linked sugar chains of human erythropoietins purified from urine and the culture medium of recombinant Chinese hamster ovary cells. J Biol Chem 263:3657–3663Google Scholar
  105. 105.
    Rahbek-Nielsen H, Roepstorff P, Reischl H, Wozny M, Koll H, Haselbeck A (1997) Glycopeptide profiling of human urinary erythropoietin by matrix-assisted laser desorption/ionization mass spectrometry. Mass Spectrom 32:948–958CrossRefGoogle Scholar
  106. 106.
    Rush RS, Derby PL, Smith DM, Merry C, Rogers G, Rohde MF, Katta V (1995) Microheterogeneity of erythropoietin carbohydrate structure. Anal Chem 67:1442–1452CrossRefGoogle Scholar
  107. 107.
    Hokke CH, Bergwerff AA, van Dedem GW, van Oostrum J, Kamerling JP, Vliegenthart JF (1990) Sialylated carbohydrate chains of recombinant human glycoproteins expressed in Chinese hamster ovary cells contain traces of N-glycolylneuraminic acid. FEBS Lett 275:9–14CrossRefGoogle Scholar
  108. 108.
    Hokke CH, Bergwerff AA, Van Dedem GW, Kamerling JP, Vliegenthart JF (1995) Structural analysis of the sialylated N- and O-linked carbohydrate chains of recombinant human erythropoietin expressed in Chinese hamster ovary cells. Sialylation patterns and branch location of dimeric N-acetyllactosamine units. Eur J Biochem 228:981–1008CrossRefGoogle Scholar
  109. 109.
    Nimtz M, Martin W, Wray V, Klöppel KD, Augustin J, Conradt HS (1993) Structures of sialylated oligosaccharides of human erythropoietin expressed in recombinant BHK-21 cells. Eur J Biochem 213:39–56CrossRefGoogle Scholar
  110. 110.
    Nimtz M, Wray V, Rüdiger A, Conradt HS (1995) Identification and structural characterization of a mannose-6-phosphate containing oligomannosidic N-glycan from human erythropoietin secreted by recombinant BHK-21 cells. FEBS Lett 365:203–208CrossRefGoogle Scholar
  111. 111.
    Kawasaki N, Haishima Y, Ohta M, Itoh S, Hyuga M, Hyuga S, Hayakawa T (2001) Structural analysis of sulfated N-linked oligosaccharides in erythropoietin. Glycobiology 11:1043–1049CrossRefGoogle Scholar
  112. 112.
    Kawasaki N, Ohta M, Itoh S, Hyuga M, Hyuga S, Hayakawa T (2002) Usefulness of sugar mapping by liquid chromatography/mass spectrometry in comparability assessments of glycoprotein products. Biologicals 30:113–123CrossRefGoogle Scholar
  113. 113.
    Ohta M, Kawasaki N, Itoh S, Hayakawa T (2002) Usefulness of glycopeptide mapping by liquid chromatography/mass spectrometry in comparability assessment of glycoprotein products. Biologicals 30:235–244CrossRefGoogle Scholar
  114. 114.
    Guan F, Uboh CE, Soma LR, Birks E, Chen J, Mitchell J, You Y, Rudy J, Xu F, Li X, Mbuy G (2007) LC-MS/MS method for confirmation of recombinant human erythropoietin and darbepoetin alpha in equine plasma. Anal Chem 79:4627–4635CrossRefGoogle Scholar
  115. 115.
    Guan F, Uboh CE, Soma LR, Birks E, Chen J, You Y, Rudy J, Li X (2008) Differentiation and identification of recombinant human erythropoietin and darbepoetin Alfa in equine plasma by LC-MS/MS for doping control. Anal Chem 80:3811–3817CrossRefGoogle Scholar
  116. 116.
    Guan F, Uboh CE, Soma LR, Birksz E, Chen J (2009) Identification of darbepoetin alfa in human plasma by liquid chromatography coupled to mass spectrometry for doping control. Int J Sports Med 30:80–86CrossRefGoogle Scholar
  117. 117.
    Yu NH, Ho EN, Wan TS, Wong AS (2010) Doping control analysis of recombinant human erythropoietin, darbepoetin alfa and methoxy polyethylene glycol-epoetin beta in equine plasma by nano-liquid chromatography-tandem mass spectrometry. Anal Bioanal Chem 396:2513–2521CrossRefGoogle Scholar
  118. 118.
    Chang Y, Maylin GM, Matsumoto G, Neades SM, Catlin DH (2010) Screen and confirmation of PEG-epoetin β in equine plasma. Drug Test Anal. 2010 Dec 29. [Epub]Google Scholar
  119. 119.
    Guan F, Uboh CE, Soma LR, Maylin G, Jiang Z, Chen J (2010) Confirmatory analysis of continuous erythropoietin receptor activator and erythropoietin analogues in equine plasma by LC-MS for doping control. Anal Chem 82:9074–9081CrossRefGoogle Scholar
  120. 120.
    Wrighton NC, Farrell FX, Chang R, Kashyap AK, Barbone FP, Mulcahy LS, Johnson DL, Barrett RW, Jolliffe LK, Dower WJ (1996) Small peptides as potent mimetics of the protein hormone erythropoietin. Science 273:458–464CrossRefGoogle Scholar
  121. 121.
    Wrighton NC, Balasubramanian P, Barbone FP, Kashyap AK, Farrell FX, Jolliffe LK, Barrett RW, Dower WJ (1997) Increased potency of an erythropoietin peptide mimetic through covalent dimerization. Nat Biotechnol 15:1261–1265CrossRefGoogle Scholar
  122. 122.
    Johnson DL, Farrell FX, Barbone FP, McMahon FJ, Tullai J, Kroon D, Freedy J, Zivin RA, Mulcahy LS, Jolliffe LK (1997) Amino-terminal dimerization of an erythropoietin mimetic peptide results in increased erythropoietic activity. Chem Biol 4:939–950CrossRefGoogle Scholar
  123. 123.
    Wiecek A, Nieszporek T, Chudek J (2007) Perspectives in the treatment of renal anaemia new concepts and new drugs. Prilozi 28:225–237Google Scholar
  124. 124.
    Macdougall IC (2008) Hematide, a novel peptide-based erythropoiesis-stimulating agent for the treatment of anemia. Curr Opin Investig Drugs 9:1034–1047Google Scholar
  125. 125.
    Perugini M, Varelias A, Sadlon T, D'Andrea RJ (2009) Hematopoietic growth factor mimetics: from concept to clinic. Cytokine Growth Factor Rev 20:87–94CrossRefGoogle Scholar
  126. 126.
    Stead RB, Lambert J, Wessels D, Iwashita JS, Leuther KK, Woodburn KW, Schatz PJ, Okamoto DM, Naso R, Duliege AM (2006) Evaluation of the safety and pharmacodynamics of Hematide, a novel erythropoietic agent, in a phase 1, double-blind, placebo-controlled, dose-escalation study in healthy volunteers. Blood 108:1830–1834CrossRefGoogle Scholar
  127. 127.
    Macdougall IC, Rossert J, Casadevall N, Stead RB, Duliege AM, Froissart M, Eckardt KU (2009) A peptide-based erythropoietin-receptor agonist for pure red-cell aplasia. N Engl J Med 361:1848–1855CrossRefGoogle Scholar
  128. 128.
    Thevis M, Thomas A, Kohler M, Beuck S, Schänzer W (2009) Emerging drugs: mechanism of action, mass spectrometry and doping control analysis. J Mass Spectrom 44:442–460CrossRefGoogle Scholar
  129. 129.
    World Anti-Doping Agency (2010) Athlete biological passport. Operating guidelines and compilation of required elements, Montreal. Accessed 9 January 2011
  130. 130.
    Gore CJ, Parisotto R, Ashenden MJ, Stray-Gundersen J, Sharpe K, Hopkins W, Emslie KR, Howe C, Trout GJ, Kazlauskas R, Hahn AG (2003) Second-generation blood tests to detect erythropoietin abuse by athletes. Haematologica 88:333–344Google Scholar
  131. 131.
    Sharpe K, Ashenden MJ, Schumacher YO (2006) A third generation approach to detect erythropoietin abuse in athletes. Haematologica 91:356–363Google Scholar
  132. 132.
    Sottas PE, Robinson N, Giraud S, Taroni F, Kamber M, Mangin P, Saugy M (2006) Statistical classification of abnormal blood profiles in athletes. Int J Biostatistics 2(1):article 3Google Scholar
  133. 133.
    Sottas PE, Robinson N, Saugy M, Niggli O (2008) A forensic approach to the interpretation of blood doping markers. Law Probab Risk 7:191–210CrossRefGoogle Scholar
  134. 134.
    Sottas PE, Robinson N, Saugy M (2010) The Athlete’s biological passport and indirect markers of blood doping. In: Thieme D, Hemmersbach P (eds) Doping in sports (Handbook of Experimental Pharmacology 195). Springer, BerlinGoogle Scholar
  135. 135.
    Mansour MM, Azzazy HM (2009) The hunt for gene dopers. Drug Test Anal 1:311–322CrossRefGoogle Scholar
  136. 136.
    Azzazy HM, Mansour MM, Christenson RH (2009) Gene doping: of mice and men. Clin Biochem 42:435–441CrossRefGoogle Scholar
  137. 137.
    Azzazy HM. Gene doping. In: Thieme D, Hemmersbach P (eds) Doping in Sports (Handbook of Experimental Pharmacology 195). Springer, BerlinGoogle Scholar
  138. 138.
    Lasne F, Martin L, de Ceaurriz J, Larcher T, Moullier P, Chenuaud P (2004) “Genetic Doping” with erythropoietin cDNA in primate muscle is detectable. Mol Ther 10:409–410CrossRefGoogle Scholar
  139. 139.
    Stieger K, Le Meur G, Lasne F, Weber M, Deschamps JY, Nivard D, Mendes-Madeira A, Provost N, Martin L, Moullier P, Rolling F (2006) Long-term doxycycline-regulated transgene expression in the retina of nonhuman primates following subretinal injection of recombinant AAV vectors. Mol Ther 13:967–975CrossRefGoogle Scholar
  140. 140.
    Menzel O, Birraux J, Wildhaber BE, Jond C, Lasne F, Habre W, Trono D, Nguyen TH, Chardot C (2009) Biosafety in ex vivo gene therapy and conditional ablation of lentivirally transduced hepatocytes in nonhuman primates. Mol Ther 17:1754–1760CrossRefGoogle Scholar
  141. 141.
    Beiter T, Zimmermann M, Fragasso A, Armeanu S, Lauer UM, Bitzer M, Su H, Young WL, Niess AM, Simon P (2008) Establishing a novel single-copy primer-internal intron-spanning PCR (spiPCR) procedure for the direct detection of gene doping. Exerc Immunol Rev 14:73–85Google Scholar
  142. 142.
    Beiter T, Zimmermann M, Fragasso A, Hudemann J, Niess AM, Bitzer M, Lauer UM, Simon P (2010) Direct and long-term detection of gene doping in conventional blood samples. Gene Ther [Epub]Google Scholar
  143. 143.
    Baoutina A, Coldham T, Bains GS, Emslie KR (2010) Gene doping detection: evaluation of approach for direct detection of gene transfer using erythropoietin as a model system. Gene Ther 17:1022–1032CrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  1. 1.Doping Control Laboratory, AIT Seibersdorf LaboratoriesSeibersdorfAustria

Personalised recommendations