analysis of 21-deoxycortisol, a marker of congenital adrenal hyperplasia, in blood by atmospheric...

RAPID COMMUNICATIONS IN MASS SPECTROMETRY

Rapid Commun. Mass Spectrom. 2004; 18: 77–82

Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/rcm.1284

Analysis of 21-deoxycortisol, a marker of congenital

adrenal hyperplasia, in blood by atmospheric pressure

chemical ionization and electrospray ionization using

multiple reaction monitoring

Simone Cristoni1*, Debora Cuccato2, Mariateresa Sciannamblo2, Luigi Rossi Bernardi1,Ida Biunno3, Piermario Gerthoux4, Gianni Russo2, Giovanna Weber2 and Stefano Mora2

1University of Milan, Centre for Biomolecular Interdisciplinary Studies and Industrial Applications CISI, Via Fratelli Cervi 93, 20090 Segrate,

Milan, Italy2Laboratory of Pediatric Endocrinology and Department of Pediatrics, Scientific Institute H. San Raffaele, Via Olgettina 60, 20132 Milan, Italy3CNR-ITB, Via Fratelli Cervi 93, 20090 Segrate, Milan, Italy4University Department of Laboratory Medicine, University of Milano-Bicocca, Hospital of Desio, Via Mazzini 1, 20033 Desio, Milan, Italy

Received 12 August 2003; Revised 27 October 2003; Accepted 27 October 2003

Congenital adrenal hyperplasia (CAH) is an autosomal recessive disorder mainly caused by 21-

hydroxylase deficit (21-OHD). Deletions or mutations of the CYP21 gene induce the impairment

of glucocorticoid and mineralcorticoid synthesis. 17-Hydroxyprogesterone (17-OHP) is the hormo-

nal marker in patients, but not in the heterozygous subjects. Excess 17-OHP is hydroxylated into 21-

deoxycortisol (21-DF), and therefore 21-DF can be used as a specific marker for diagnosis of hetero-

zygous individuals. We report an analytical method for analysis of 21-DF in blood samples using

electrospray (ESI) and atmospheric pressure chemical ionization (APCI), showing that ESI is very

sensitive for the analysis of this marker molecule. The multiple reaction monitoring (MRM)

approach was used to increase the specificity and the sensitivity of the method. Copyright #

2003 John Wiley & Sons, Ltd.

Congenital adrenal hyperplasia (CAH), caused by a 21-

hydroxylase deficit (21-OHD), is an autosomal recessive dis-

order in which deletions or mutations of the CYP21 gene

induce the impairment of glucocorticoid and mineralcorti-

coid synthesis.

The classical form of 21-OHD occurs in about 1:15 000 live

births.1–5 CAH diagnosis is usually performed by measuring

blood levels of adrenal hormones and precursor steroids.6 In

particular, 17-hydroxyprogesterone (17-OHP) is not hydro-

xylated into 11-deoxycortisol (11-DF) and cortisol synthesis is

decreased. Consequently, the secretion of adrenocorticotro-

pic hormone (ACTH), which is the promoter of the cortisol

metabolic pathway, is increased. Overstimulation by ACTH

is responsible for the high concentration of circulating 17-

OHP. For this reason, high 17-OHP plasma levels are used to

diagnose this disorder. The excess 17-OHP is hydroxylated

into 21-deoxycortisol (21-DF) in the adrenal gland, and this

metabolic pathway becomes very important in patients with

21-OHD.7 It is known that ACTH induces much higher 21-DF

plasma levels in heterozygous than in normal subjects.8–12

Several studies8–10,13–15 have demonstrated that 21-DF is a

more sensitive marker than 17-OHP because it allows the

detection of more than 90% of heterozygous carriers. Plasma

levels of 21-DF before and 60 min after ACTH stimulation is

becoming a new approach for detection of heterozygous

carriers of 21-OHD.

The methods currently used for the detection of 21-DF are

radioimmunoassay (RIA)8–10,13,14 and solid-phase time-

resolved fluoroimmunoassay (TR-FIA).16 In particular, the

RIA technique leads to high sensitivity (0.6 pg/tube)14 but on

the other hand it is labour intensive and sample loss can occur

during the extraction and preparation procedures.10 For

these reasons, the use of a mass spectrometry approach that is

highly sensitive, fast, is not labour intensive nor dangerous, is

of interest. Both gas and liquid chromatography coupled

with mass spectrometry (GC/MS and LC/MS) have been

used in recent years for the study of steroid compounds.17–19

In particular, several methods, based on coupling GC and LC

to mass spectrometry,20,21 have been developed for the

diagnosis of 21-hydroxylase deficit (21-OHD).

Liquid chromatography with tandem mass spectrometry

(LC/MS/MS) is a technique widely used to characterize and

quantify steroids.21,22 In particular, the ion sources typically

employed to analyze steroids are electrospray (ESI)23–25 and

atmospheric pressure chemical ionization (APCI).26–28 The

use of LC/MS/MS offers the advantage that there is no need

for derivatization of the samples before injection, as is

Copyright # 2003 John Wiley & Sons, Ltd.

*Correspondence to: S. Cristoni, Universita degli Studi di Milano(CISI), Via Fratelli Cervi 93, 20090 Segrate, Milan, Italy.E-mail: [email protected]/grant sponsor: ARFSAG—Lombardia (RegionalAssociation of Families with Cortical Adrenal Hyperplasia).

required in GC/MS. Thus the detection of compounds of

interest can be achieved by directly injecting biological fluids

like urine and saliva.29,30 Kao et al.21 have developed

an approach based on the use of MS/MS for the analysis of

21-DF. They used a TurboIonSpray ionization source work-

ing at a high flow rate (1 mL/min) using a 4.6 mm i.d. column.

However, under these conditions, they can detect the 21-DF

only in serum from patients with 21-hydroxylase deficiency

(21-OHD). Furthermore, they did not use the MRM approach

that can improve the instrumental performance in terms of

specificity and sensitivity.

The aim of this study is to compare APCI and ESI

performance in the analysis of 21-DF. The 21-DF content

was extracted from the blood of four subjects, analyzed

by MRM methods, and the results thus obtained are

reported here.

EXPERIMENTAL

ChemicalsStandard 11-deoxycortisol (11-DF) and 21-deoxycortisol

(21-DF) were purchased from Sigma Aldrich (Milan, Italy).

Methanol (CH3OH) was purchased from J. T. Baker

(Deventer, Holland). Trifluoroacetic acid (TFA, CF3COOH)

was purchased from Lancaster (Eastgate, White Lund, Mor-

ecambe, UK). Adrenocorticotropic hormone (ACTH,

Synacthen) was purchased from Ciba-Geigy (Basel, Switzer-

land). Isooctane (C8H18) and ethyl acetate (CH3COOC2H5)

were purchased from E. Merck (Darmstadt, Germany).

Figure 1. Structures of (a) 21-DF and (b) 11-DF.

Figure 2. (a) LC/APCI-MS/MS analysis of a mixture containing both 21- and 11-DF, each at

a concentration of 50 ng/mL. A volume of 20 mL was injected. (b) MS/MS spectrum of

protonated 21-DF. (c) MS/MS spectrum of protonated 11-DF.

Copyright # 2003 John Wiley & Sons, Ltd. Rapid Commun. Mass Spectrom. 2004; 18: 77–82

78 S. Cristoni et al.

Sample preparationThree heterozygous females at random stages of their men-

strual cycles and one healthy man were studied. The subjects

were relatives of CAH patients. The heterozygous state was

determined by molecular and biochemical screening.

Informed consent was obtained from each subject prior to

the study.

The ACTH stimulation test was performed by injecting

250mg of ACTH intravenously. Two blood samples were

taken both before and 60 min after ACTH stimulation.

Plasma was obtained by blood centrifugation at 1550 g for

10 min at room temperature. Plasma samples (2 mL) were

extracted twice with isooctane/ethyl acetate (1:1, v/v). The

first extraction was made with 40 mL of the solvent mixture

after vortexing for 1 min; then the pellet was frozen with

liquid nitrogen. The extract containing steroids was sepa-

rated from the frozen pellet, and the same pellet was

extracted a second time using 20 mL of the solvent mixture

following the procedure described above. Both steroid

extracts were collected and dried with nitrogen. The dried

extract was resuspended in 40 mL water/methanol (50:50),

and 20mL were analyzed. 20 mL of standard solutions in the

concentration range 0.250–600 ng/mL were injected to

obtain the APCI and ESI calibration curves.

ChromatographyA Surveyor mHPLC system (ThermoFinnigan, Palo Alto, CA,

USA) was used. The chromatographic column was a reverse-

phase C18 (250� 2.1 mm, 5mm, 300 A). A HPLC gradient was

performed using two eluents: (A) H2Oþ 0.025% TFA; (B)

CH3OHþ 0.025% TFA. A linear gradient was used, from

52% to 77% of B in 10 min. The eluent flow was 200 mL/min.

A divert valve was used in order to send the eluent to waste

for the first 4 min of HPLC analysis.

Mass spectrometryThe APCI mass spectra were obtained using a LCQXP ion trap

(ThermoFinnigan). The source temperature was 3508C and

the entrance capillary temperature was 1508C. The corona

discharge voltage was 5 kV. The flow rate of nebulizing gas

(nitrogen) was 2.00 L/min. The He pressure inside the trap

was kept constant; the pressure directly read by ion gauge

(in the absence of N2 stream) was 2.8� 10�5 Torr. The maxi-

mum injection scan time was 200 ms, 5 microscans were used,

and the automatic gain control was turned on.

The same instrument was used to obtain the ESI mass

spectra. The needle voltage was 5 kV. The entrance capillary

temperature was 2408C. The flow of nebulizing gas (nitrogen)

was 1.5 L/min. The He pressure inside the trap was kept

constant; the pressure directly read by ion gauge (in the

absence of N2 stream) was 2.8� 10�5 Torr. The maximum

injection scan time was 200 ms, 5 microscans were used, and

the automatic gain control was turned on.

HPLC/ESI and HPLC/APCI chromatograms were

acquired using tandem mass spectrometry (MS/MS) and

MRM in positive acquisition mode. The isolation width

(baseline) of the precursor ion was 3 Th. The fragment ion

mass width (baseline) was also 3 Th. The collision energy was

30% of its maximum value (5 V peak to peak). Five

microscans were used and the microscan time in MRM mode

was 20 ms. Under these conditions the instrumental resolu-

tion was effectively unit mass.

Data analysisThe signal/noise (S/N) ratio was calculated using the RMS

algorithm. The chromatographic data were processed using

Xcalibur qualbrowser and Exel software.

RESULTS AND DISCUSSION

Two isomers of deoxycortisol (21-DF and 11-DF) are present

in blood samples (Figs. 1(a) and 1(b)), but only 21-DF is used

to detect heterozygous individuals. Preliminary experiments

were performed using the APCI source. As reported in the lit-

erature,26–28 this ionization source is very efficient in the

analysis of steroids. In order to separate and quantify

the two isomers, an LC/MS/MS method was developed.

The MS/MS chromatogram, and the MS/MS spectra of the

[MþH]þ ions at m/z 347 of 21- and 11-DF isomers, are shown

in Figs. 2(a), 2(b) and 2(c), respectively (positive acquisition

mode was used). 20mL of a mixture of the 11- and 21-DF iso-

mers, each with a concentration of 50 ng/mL (1 ng injected

on-column), were injected in order to obtain the MS/MS

chromatogram. Under these conditions the signal response

of 21-DF is about 5 times higher than that of 11-DF. The

S/N ratio of the chromatographic peak of 21-DF at retention

time 7.02 min is about 100 (calculated using the RMS algo-

rithm). Under these chromatographic conditions, chromato-

graphic peaks for 21- and 11-DF are well resolved.

Scheme 1. Fragmentation pathway proposed for 11-DF.

Characterization and quantification of steroids 79


In the MS/MS spectrum of 11-DF (Fig. 2(c)) two peaks at

m/z 317 and 299 are observed, while in the case of 21-DF

(Fig. 2(b)) these fragments are absent. Scheme 1 shows the

fragmentation pathway proposed for 11-DF that could lead to

these fragments. The peak at m/z 317 is due to the loss of a

CH2O neutral molecule with a hydrogen rearrangement. The

peak at m/z 299 is due to a further water loss, again with

hydrogen rearrangement. The two most abundant peaks

obtained in the 21-DF fragmentation (Fig. 2(b)) are those at

m/z 311 and 293 (Scheme 2) that correspond to the loss of two

and three neutral water molecules, respectively, and are also

present in the MS/MS spectrum of 11-DF (Fig. 2(c)).

This difference in fragmentation behaviour of the two

molecules was used to perform multiple reaction monitoring

(MRM), monitoring the fragment ions atm/z 317, 311, 299 and

293. The peaks at m/z 311 and 293 are present in the

fragmentation spectra of the two isomers and thus they are

not useful for selective discrimination of the two steroids, but

they were chosen for their high abundance in the spectrum of

21-DF. As mentioned, the peaks atm/z 317 and 299 are present

only in the 11-DF fragmentation spectrum; these peaks were

monitored to increase the selectivity of the method since they

are not present in the MS/MS spectrum of 21-DF at

the retention time of 7.02 min but are clearly detected in the

MS/MS spectrum of 11-DF at the retention time of 8.31 min.

Thus, two parameters (retention time and the lack of the

peaks at m/z 317 and 299 in the MS/MS spectrum of 21-DF)

allow the reliable accurate identification of 21-DF.

The same analysis as that reported above was performed

under ESI conditions in order to determine whether the same

experimental data were reproduced also by using this

ionization source and in order to evaluate the efficiency of

this ionization approach. The same fragmentation behaviour

of the two isomers as was obtained using APCI was observed

with ESI. Moreover, increased sensitivity and linear dynamic

range were observed by using this technique. The calibration

curves for 21-DF obtained using ESI and APCI are shown in

Figs. 3(a) and 3(b), respectively. In both cases good linearity

was achieved (R2¼ 0.9995 for the ESI approach and

R2¼ 0.9976 for APCI). However, the limit of detection

obtained using the ESI source (0.20 ng/mL injecting 20mL,

corresponding to 4 pg injected on-column) was lower than

that obtained with the APCI source (1 ng/mL injecting 20mL,

corresponding to 20 pg injected on-column). The linear

dynamic range was also enhanced, from 30–600 ng/mL

(obtained under APCI conditions) to 0.25–600 ng/mL (ESI

conditions). The limit of quantitation of the developed

LC/ESI-MS/MS-MRM approach is lower than that obtained

Scheme 2. Fragmentation pathway proposed for 21-DF.



previously using LC/MS/MS but without using MRM

(50 ng/mL injecting 17mL, corresponding to 850 pg injected

on-column).21 The absolute limits of quantitation of RIA

(0.75 pg/tube)14,16 and GC/MS (0.5 ng/mL, injecting 2mL,

corresponding to 1 pg injected on-column) are about 4–7

times lower than that obtained using our approach based on

the use of ESI and MRM (5 pg injected on column). However,

both RIA and GC/MS approaches require more time

consumption with respect to the LC/ESI-MS/MS-MRM

method. Furthermore, the linear range obtained (0.25–

600 ng/ml, injecting 20 mL, corresponding to 5–12 000 pg

injected on-column) is good enough if compared with

that used in the case of LC/MS/MS without using MRM

(50–250 ng/mL, injecting 17 mL, corresponding to 850–

4250 pg injected on-column)21, RIA (0.75–100 pg/tube)14,16

and GC/MS (0.5–20 ng/mL, injecting 2mL, corresponding to

1–40 pg injected on-column)20,31 approaches.

Thus, ESI was chosen in order to detect and quantify 21-DF

in eight blood samples obtained from four selected subjects.

Figures 4(a) and 4(b) show the MRM chromatograms

obtained by injecting a standard solution of 21-DF (50 ng/

mL injecting 20mL, corresponding to 1 ng injected on-

column) and a solution obtained by extracting 21-DF from a

blood sample of a selected subject. The peak corresponding to

21-DF was clearly observed in both cases.

The ESI approach was therefore used to quantitate the 21-

DF in blood samples. Table 1 summarizes the results obtained

by analyzing blood samples obtained from four selected

subjects. The samples were collected before and after ACTH

stimulation in order to distinguish between normal and

heterozygous carriers. For the healthy subject (number 2,

Table 1) no increase in the level of 21-DF was observed after

stimulation. In contrast, clearly different behaviour was

observed for heterozygous carriers (numbers 1, 3 and 4,

Table 1) in which the level of 21-DF strongly increased (about

10–15 times) after ACTH stimulation. These results fully

confirm the literature data obtained by using the radio-

immunoassay technique.8–10,13,14

Figure 3. (a) ESI calibration curve obtained using a 0.250–

600ng/mL solution concentration range of 21-DF. MRMmode

was used. (b) APCI calibration curve obtained using a 30–

600ng/mL solution concentration range of 21-DF. MRMmode

was used.

Figure 4. ESI MRM chromatograms obtained by injecting (a) 20 mL of a 21-DF

standard solution at a concentration of 50 ng/mL and (b) a plasma extract of a healthy

subject.

Table 1. Quantitative analysis of 21-DFextracted from blood

samples of four selected subjects obtained before and after

60min stimulation with ACTH

Subjects*

21-DF concentrationbefore ACTH stimulation

(ng/mL)

21-DF concentration afterACTH stimulation

(ng/mL)

1 3.1� 0.5 27.8� 3.02 2.1� 0.5 2.2� 0.53 3.5� 0.5 39.1� 2.54 1.9� 0.6 32.7� 2.8

* Heterozygous carriers are subjects # 1, 3 and 4.

Characterization and quantification of steroids 81


CONCLUSIONS

A method based on the use of LC/MS/MS with MRM, that

permits discrimination between 21-DF (a marker of congeni-

tal adrenal hyperplasia disease) and 11-DF, has been per-

formed using an APCI source. The data obtained using

APCI were compared with those obtained under ESI condi-

tions. The latter technique leads to a higher sensitivity and

linear dynamic range in the analysis of 21-DF, and for this rea-

son it was chosen for detecting and quantifying this com-

pound in blood samples of four selected subjects. In all

samples the 21-DF marker was readily monitored with a

10 min chromatographic analysis. Furthermore, the data

obtained clearly show that the limit of quantitation and linear

range achieved makes it possible to analyze and quantify

21-DF molecules in both healthy and heterozygous carrier

subjects; thus this method could be developed for success-

ful application in the clinic as an alternative to the labour-

intensive RIA8–10,13,14 and GC/MS21,30 approaches.

Future investigations will be focused on increasing the

sensitivity of the techniques by reducing the diameter of the

chromatographic column from 2.1 mm to 0.5 mm. The

developed methods will also be applied to the study of a

large number of blood samples in order to fully validate this

approach for the diagnosis of CAH disease.

AcknowledgementsThis work was supported by ARFSAG—Lombardia (Regio-

nal Association of Families with Cortical Adrenal Hyperpla-

sia). The authors also thank Drs Maria Carla Proverbio and

Ilaria Zamproni for their technical support.

REFERENCES

1. Appan S, Hindmarsh PC, Brook CGD. Arch. Dis. Childhood1989; 64: 1235.

2. Bergstrand CG. Acta Pediatr. Scand. 1966; 55: 463.3. Brook CGD. Clin. Endocrinol. 1990; 33: 559.4. Pang S, Wallace MA, Hofman L, Thuline HC, Dorche C,

Lyon ICT, Dobbins RH, Kling S, Fufjieda K, Suwa S. Pedia-trics 1988; 81: 866.

5. Speiser PW, Dupont B, Rubinstein P, Piazza A, Kastelan A,New MI. Am. J. Hum. Genet. 1985; 37: 650.

6. New MI, Rapaport R. Pediatric Endocrinology, Sperling MA(ed). W.B. Saunders: Philadelphia, 1996; 281–314.

7. Finkelstein M, Shaefer JM. Physiol. Rev. 1979; 59: 353.8. Milewicz A, Vecsei P, Korth-Schutz S, Haack D, Roseler A,

Lichtwald K, Lewiscka S, Mittelstaedt GV. J. Steroid Biochem.1984; 21: 185.

9. Gueux B, Fiet J, Pham-Huu-Trung MT, Villette JM,Gourmelen M, Galons H, Brerault JL, Vexiau P, Julien R.Acta Endocrinol. (Copenhagen) 1985; 108: 537.

10. Fiet J, Gueux B, Gourmelen M, Kutten F, Vexiau P,Coullin Ph, Pham-Huu-Trung MT, Villette JM,Raux-Demay MC, Galons H, Julien R. J. Clin. Endocrinol.Metab. 1988; 66: 659.

11. Gourmelen M, Gueux B, Pham-Huu-Trung MT, Fiet J,Raux-Demay MC, Girard F. Acta Endocrinol. (Copenhagen)1987; 116: 507.

12. Milewicz A, Afelska A, Wasikowa R, Romer T. Endokrynol.Pol. 1987; 44: 187.

13. Nahoul K, Adeline J, Bercovici JP. J. Steroid Biochem. 1989;33: 1167.

14. Fiet J, Villette JM, Galons H, Boudou Ph, Burthier JM, HardyN, Soliman H, Julien R, Vexiau P, Gourmelen M, Kutten F.Ann. Clin. Biochem. 1994; 31: 56.

15. Hill M, Lapcik O, Hampl R, Starka L, Putz Z. Steroids 1995;60: 615.

16. Fiet J, Boundi A, Giton F, Villette JM, Boudou Ph, SolimanH, Morieau G, Galons H. J. Steroid Biochem. Mol. Biol. 2000;72: 55.

17. Shimada K, Mitamura K, Higashi T. J. Chromatogr. A 2001;935: 141.

18. Appelblad P, Irgum K. J. Chromatogr. A 2002; 955: 151.19. Dehennin L, Nahoul K, Scholler R. J. Steroid Biochem. 1987;

26: 337.20. Caulfield MP, Lynn T, Gottschalk ME, Jones KL, Taylor NF,

Malunowicz EM, Shackleton CH, Reitz RE, Fisher DA.J. Clin. Endocrinol. Metab. 2002; 87: 3682.

21. Kao PC, Machacek DA, Magera MJ, Lacey JM, Rinaldo P.Ann. Clin. Lab. Sci. 2001; 31: 199.

22. Antignac JP, Le Bizec B, Monteau F, Poulain F, Andre F.Rapid Commun. Mass Spectrom. 2000; 14: 33.

23. Mulholland M, Whelan TJ, Rose H, Keegan J. J. Chromatogr.A 2000; 870: 135.

24. Yang L, Amad M, Winnik WM, Schoen AE, SchweingruberH, Mylchreest I, Rudewicz PJ. Rapid Commun. Mass Spec-trom. 2002; 16: 2060.

25. Van Den Hauwe O, Perez JC, Claereboudt J, Van PeteghemC. Rapid Commun. Mass Spectrom. 2001; 15: 857.

26. Wu Z, Zhang C, Yang C, Zhang X, Wu E. Analyst 2000; 125:2201.

27. Joos PE, Van Ryckeghem M. Anal. Chem. 1999; 71: 4701.28. Lagana A, Bacaloni A, Fago G, Marino A. Rapid Commun.

Mass Spectrom. 2000; 14: 401.29. Leinonen A, Kuuranne T, Kostiainen R. J. Mass Spectrom.

2002; 37: 693.30. Shindo N, Yamauchi N, Murayama K, Fairbrother A, Korlik

S. Biomed. Chromatogr. 1990; 4: 171.31. Ichimura K, Yamanaka H, Chiba K, Shinozuka T, Shiki Y,

Saito K, Kusano S, Ameniya S, Oyama K, Nozaki Y, Kato K.J. Chromatogr. 1986; 374: 5.



analysis of 21-deoxycortisol, a marker of congenital adrenal hyperplasia, in blood by atmospheric...

Documents