Mechanisms of Ageing and Development
121 (2000) 59 – 67
www.elsevier.com/locate/mechagedev

A simple method for the measurement of
sjTREC levels in blood
Richard Aspinall *, Jeff Pido, Deborah Andrew
Department of Immunology, Imperial College School of Medicine, Chelsea and Westminister Hospital,
369 Fulham Road, London SW 10 9NH, UK
Received 2 July 2000; received in revised form 21 July 2000; accepted 30 July 2000

Abstract
We have developed a relatively rapid, safe and simple method for the quantification of
sjTREC levels in samples of peripheral blood. The assay uses an image analysis package to
measure the brightness of PCR product bands on an image of the standard agarose gel.
Comparison of the brightness of the band with that obtained from a standard curve provides
a read-out of the amount of sjTRECs in the sample. We have compared the sjTREC levels
we obtained with this method with those obtained from real time analysis PCR using a
Lightcycler and found that they are comparable. © 2000 Elsevier Science Ireland Ltd. All
rights reserved.
Keywords: sjTREC; Thymic output; PCR

1. Introduction
Extrapolation of the rate of loss of thymic tissue in humans suggests that it
would be completely absent by the time the individual reaches the age of 120 years
(Steinmann, 1986). Loss of thymic tissue over this period is not uniform but more
likely biphasic with a period of rapid loss preceding middle age followed by a
period of relatively slow tissue loss calculated by some to be less than 1% per year
(Steinmann, 1986; Kendall et al., 1980; Bertho et al., 1997). In addition to the loss
of thymopoetically active tissue there is the gradual deposition of fat within the
* Corresponding author. Tel.: +44-20-87465993; fax: + 44-20-87465997.
E-mail address: r.aspinall@ic.ac.uk (R. Aspinall).
0047-6374/00/$ - see front matter © 2000 Elsevier Science Ireland Ltd. All rights reserved.
PII: S 0 0 4 7 - 6 3 7 4 ( 0 0 ) 0 0 1 9 7 - 4

60

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

organ. At 20 years of age approximately 20% of the wet weight of the thymus is
contributed by fat, but by 60 years of age this figure is closer to 70% (Kendall et
al., 1980). The major function of the thymus is to produce T cells for the peripheral
T cell pool, and a result of this loss of active thymic tissue is that the contribution
of naıve T lymphocytes by the thymus to the peripheral T cell pool declines with
¨
increasing age.
Measurement of thymic output in humans is indirect and dependant upon
following the changes in the number of naıve T cell with age, a naıve T cell being
¨
¨
defined as a T cell which has not yet encountered its cognate antigen presented
correctly to the T cell receptor. Such naıve T cells have in the past been identified
¨
on the basis of phenotypic markers displayed on the cell’s surface. Using this
technique several groups have shown that there is a decline in the number naıve T
¨
cells in the blood and an increase in the number of memory T cells with increasing
age (Cossarizza et al., 1996; Hulstaert et al., 1994).
More recently measurement of change in thymic output has been achieved by the
analysis of the amount of specific DNA excision products known as T cell receptor
rearrangement excision circles (TRECs) within the T cell population. These excision
circles are a by-product of the process of TCR gene rearrangement and recombination and are present within the T cell but do not have the capacity to replicate
during cellular proliferation. Thus only one of the daughter cells produced during
division carried the TREC which consequently becomes diluted within the population during subsequent divisions. TREC levels are therefore highest in populations
of T cell recently produced by the thymus and lower in T cells populations which
have undergone several rounds of division (Douek et al., 1998; Kong et al., 1998;
Al Harthi et al., 2000; McFarland et al., 2000; Poulin et al., 1999; Zhang et al.,
1999).
With the age-related decline seen in thymic output and the proposed expansion
of memory T cells to compensate for this decline and maintain T cell numbers
within defined limits one would anticipate that the number of TREC molecules
within the total T cell pool in the blood should decline with age. Several studies
have shown this to be the case. TREC levels do show a progressive decline with age
(Douek et al., 1998).
In the analysis of TREC levels one can choose whether to analyse the TREC
generated from the TCRa (Douek et al., 1998) or TCRb (Poulin et al., 1999) chain
since excision circles of DNA would be generated in their production. Here we have
chosen to analyse signal joint TCR rearrangement excision circles (sjTRECs)
generated during the process of aTCR gene rearrangement and recombination
(Douek et al., 1998; Kong et al., 1998; Livak and Schatz, 1996). There are several
reasons for this choice. The first is that TCRa chain gene rearrangement occurs
after TCRb chain rearrangement. Any excision circles generated from TCRb chain
rearrangement could be more diluted within the recent thymic migrant population
than the TCRa chain excision circles because of the number of cell divisions
undergone by the cells between TCRb chain production and expression and TCRa
chain production and expression. Secondly these sjTRECs all possess an identical
DNA sequence, spanning from the dREC gene segment to the cJa gene segment

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

61

which allows their use as specific markers for abTCR expressing recent thymic
migrants (Douek et al., 1998). Finally, sjTREC’s have been shown to be stable and
detectable in phenotypically naıve abTCR T cells (CD45RA+) but are undetectable
¨
in gd TCR T cells, and B cells (Douek et al., 1998). The aim of this work was to
determine whether we could discover a rapid and sensitive method for the quantitation of sjTREC’s.

2. Materials and methods

2.1. Blood collection and DNA purification
Samples of peripheral blood were taken from healthy female volunteers chosen
on the criteria that they were healthy at the time of sample extraction. 30 ml of
volunteers’ peripheral blood was collected into EDTA-K3 containers (Becton
Dickinson, Dorset, UK) with their informed consent and in accordance with the
guidelines set by the Riverside Ethics Committee. Peripheral blood mononuclear
cells (PBMCs) were isolated by density gradient centrifugation using Histopaque
(Sigma, Dorset, UK). DNA was extracted from 5 × 107 PBMCs using the Puregene DNA purification kit (Gibco, UK). b actin polymerase chain reaction (PCR)
amplification and agarose gel analysis was performed for each sample in order to
determine the quality of the DNA. The concentration of the DNA was determined
by spectroscopy.

2.2. sjTREC PCR amplification and quantification
sjTREC DNA were detected according to the technique of Douek et al. (1998).
The sjTREC bands were visualised on 1.2% agarose gels containing 0.005%
ethidium bromide (Sigma) and analysed using the Scion Imager (Meyer Instruments). The light intensity values of the bands were used to calculate the number of
sjTREC molecules from the standard curve constructed (details below). Results
were used if the light intensity value of the positive controls’ DNA band differed to
the actual values on the standard curve by no more than 10%.
To determine the sjTREC levels in volunteers’ DNA samples, a standard curve
was constructed via the PCR amplification of known starting numbers of standard
sjTREC molecules (i.e. 103, 104, 105, 106, 107, 108). PCR products were then
separated on 1.2% agarose gels containing 0.005% ethidium bromide and analysed
using the Scion imager software. An area was defined around each band and the
brightness was read. The size and shape of the area was kept constant throughout
all of the measurement. For the final analysis a background reading was taken from
the negative control lane at an equivalent position of the sjTREC band in a positive
lane of the gel. This background reading was subtracted from the readings obtained
from each PCR product band. A best fit straight-line graph (R 2 = 0.97) composed
of the starting amount of sjTREC molecules used for the PCR against DNA band
intensity was constructed.

62

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

2.3. sjTPEC analysis using the Lightcycler
DNA samples and known starting copy numbers of standard sjTREC DNA
molecules (from 103 to 108) were amplified using the Lightcycler (Roche, UK) and
the Sybr Green Light Cycler kit (Roche, UK) in accordance with the manufacturers
instructions. Values for the volunteers’ sjTREC levels were considered valid when
the sjTREC standard curve had an R 2 value of 0.99–1.0.

2.4. Separation of T cell subsets
5 × 106 purified PBMCs were treated with 1 mg of mouse IgG monoclonal
antibody (mAb) to CD45RA or CD45RO and incubated for 30 min at 4°C. The cells
were then washed with PBS/0.1% BSA. Dynabeads pan mouse IgG were added to
the amount of approximately 4 beads/cell (beads were first washed with PBS/0.1%
BSA 3 times prior addition to the cells) to the samples and incubated for 20 min with
gentle mixing. The cells were placed on a Dynal MPC magnetic column and left for
2 min. The fluid in the tubes was removed by pippetting and the cells resuspended
in PBS/0.1% BSA, then left for 2 min. The fluid was again removed by pippetting
and the beads resuspended in PBS/0.1% BSA and left on the Dynal MPC for 2 min
(this step was repeated two times). The purified CD45RA or CD45RO cells were then
resuspended in PBS/0.1% BSA. The purified CD45RA and CD45RO cells were then
treated with 1 mg of mouse IgM mAb to either CD4 or CD8 and incubated for 30
min at 4°C and then washed with PBS/0.1% BSA. Dynabeads pan mouse IgM were
added to the cells and the subsequent procedure used for the CD45RA and CD45RO
cell purification was undertaken to obtain CD4 cells expressing CD45RA or CD45RO
and CD8 cells expressing CD45RA or CD45RO.

2.5. FACS analysis of CD3 and CD45RA, CD62L T lymphocyte populations
100 ml of peripheral blood obtained from subjects were treated with the following
monoclonal antibodies; mouse anti-human CD3 conjugated to FITC (Sigma), mouse
anti-human CD45RA conjugated to Quantum red (Pharmingen, Dorset, UK) and
mouse anti-human CD62L conjugated to PE (Pharmingen) or to the appropriate
isotype matched negative controls and incubated on ice for 30 min. The red cells were
then lysed by the addition of 2 ml of Ortholyse (Ortho Diagnostic Systems,
Buckinghamshire, UK) and the samples were analysed using the BD FACSCalibur.

3. Results

3.1. Detection of sjTREC in naı6e but not memory T cells
¨
We separated T cell subsets to determine firstly whether the technique we used
could discriminate in the detection of sjTREC in phenotypically defined naıve and
¨

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

63

memory cells and secondly whether there was a difference in the CD4 and CD8
subset expression of these sjTREC. Our results, shown in Fig. 1 reveal that sjTREC
were detectable in naıve (CD45RA+) cells in both CD4 and CD8 subpopulations.
¨
However we failed to detect sjTREC in the memory (CD45RO+) populations of
either CD4 or CD8 subset using this same PCR conditions which provided positive
results for the naıve T cell subpopulations.
¨

3.2. Quantitation by scanning the image
The results of the PCR analysis using differing initial amounts of the sjTREC
positive control molecule for the construction of a standard curve is shown in Fig.
2. The results seen here are the means and standard deviations from four separate
experiments. The line of best fit drawn on the points obtained had an R 2 value of
0.97 indicating the close correlation between the brightness of the band and the
initial sjTREC copy number. Inset in the graph is an example of the results from
one of the gels containing the PCR products arising from different starting sjTREC
numbers.

3.3. Comparison with Lightcycler results
The results of the sjTREC levels in blood from female donors of three different
ages is shown in Fig. 3(a) and (b). Fig. 3(a) shows the results obtained using the
Lightcycler and Fig. 3(b) shows the results from the Scion Image scanning method.
Comparison of the two results indicates that both clearly show an age-related

Fig. 1. Detection of sjTREC in T cell subsets. The gel shows the presence of the sjTREC band in DNA
derived from CD4+CD45RA+ (Lane 2) and CD8+CD45RA+ (Lane 7), but not in CD4+CD45RO+
(Lane 4) and CD8+CD45RO+ (Lane 6) cells. The molecular weight markers are in Lane 1 and the
shorter sjTREC molecule used as a positive control is in Lanes 3 and 5.

64

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

Fig. 2. Standard curve generated from the sjTREC positive control. Brightness analysis of the bands
produced from PCR amplification of different starting copy numbers, ranging from 103 to 108, of the
shortened sjTREC gene. Each point shown is the mean 9 1 S.D. from 4 readings. The result from one
gel is shown above the points.

decrease in sjTREC numbers with both detecting comparable numbers of sjTREC
within the samples. The results showing a decline in sjTREC levels with age are in
agreement with those observed in previous studies as well as reflecting the results
obtained from phenotypic analysis. Furthermore as expected these samples show
the number of true naıve T cells (CD3+CD45+CD62L+) to decline with age.
¨
4. Discussion
The measurement of episomal DNA circles in the nucleus of T cells is a sensitive
method for the determination of thymic output with age. The use of sjTREC
measurement either in relation to the DNA content of the sample (Douek et al.,
1998 ) or the total CD3+ T cell content of the sample as here provides a means of
assessing the naıve T cell subpopulation of the sample.
¨
This paper describes the detection of sjTREC in CD45RA+ cells in both the CD4
and CD8 T cell subsets, but the failure to detect these products in CD45RO+ T
cells in either the CD4 or CD8 subsets. This result would be anticipated from
previous work, which shows that the sjTREC episomal circles are not replicated
with the cell and so are subsequently diluted out by cell division. It is not that the
sjTREC molecules are absent from the CD45RO+ populations, but that they are
present at such a low frequency per cell that the PCR conditions used failed to
detect them. The CD45RA+ T cells population contains the naıve T cell emigrants
¨
from the thymus which are a non-dividing population with the highest frequency of
TRECS per T cell. Naıve T cells only enter division once activated by antigen and
¨
enters the memory/activated T cell pool (CD45RO+).

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

65

The identification of recent thymic migrants on the basis of phenotype has been
possible in some rodent studies because of the use of intrathymic labeling with
fluorescein (Scollay et al., 1980) or the identification that the recent thymic migrant
has distinctive phenotype (Hosseinzadeh and Goldschneider, 1993). However in
human systems this has proved to be somewhat more complex. The problem has
arisen from studies showing an ability of cells with some of the phenotypic
characteristics of naıve T cell to arise from the memory T cell pool (Nociari et al.,
¨
1999). These populations have been shown to contain virus specific clones after the
resolution of the specific infection (Wills et al., 1999) and in addition it has been
claimed that this change in phenotype has been linked to the loss of CD28 (Nociari
et al., 1999). The use of TREC’s as an additional marker to identify recent thymic

Fig. 3. Levels of sjTREC per 5 × 107 of each of volunteers peripheral blood mononuclear cells as
derived by the Lightcycler (a) or by the scanning method (b). The age and number of naıve
¨
(CD3+CD45RA+CD62L+) T cells in the sample of 5 × 107 PBMC was (i) 21 years old and 11 373 000
naıve T cells, (ii) 33 years old and 5 397 600 naıve T cells and (iii) 55 years old and 3 250 000 naıve T
¨
¨
¨
cells.

66

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

migrants has provided a clearer means of identification of recent thymic migrants
and assessing thymic output (Al Harthi et al., 2000; McFarland et al., 2000; Poulin
et al., 1999; Zhang et al., 1999).
One of the major problems with the use of TREC measurements is that previous
studies have used either radioisotopes (Douek et al., 1998), ELISA based methods
(Al Harthi et al., 2000) or expensive equipment to measure TREC levels with real
time PCR (Zhang et al., 1999). The use of radioisotopes was the first method to be
used to quantitate TREC levels and although sensitive and effective has some cost
and safety implications and in some cases problems with availability. The ELISA
method of detection uses PCR in the presence of Dig-UTP, and the product is
captured on a 96 well plate coated with streptavidin by a biotin conjugated probe
recognizing an internal sequence and an ELISA is performed using an anti-Dig
peroxidase antibody. The resulting analysis is sensitive and quantitative, but as with
all multi-step processes is dependant on all stages and reagents working correctly.
The measurement of product in a real time PCR is a fast direct method of
assessment, but the initial outlay to purchase the equipment is prohibitive in for
some laboratories.
This paper describes a rapid, relatively safe and simple method of measuring the
amount of sjTREC’s in a sample of T cells which does not depend upon using
radioisotopes. The method uses the brightness of the DNA band in the gel
measured on a captured image using an analysis programme freely available on the
Internet. As with all methods there are some critical stages to the assay. The
exponential shape of the graph for different amounts of positive control sample
used clearly shows that the correct number of PCR cycles are being used. For a
sample to be assessed correctly it is important that it falls within the limits of the
graph used to produce the standard curve. Clearly it is important that the
conditions used for the analysis of the DNA under test be as close to those used to
determine the standard curve. So considerations such as the thickness and running
conditions of the gel, amount of ethidium bromide and settings on the image
capture equipment must be kept constant. Repeated running of the positive control
standards and the construction of a standard curve as we have done here must be
a first priority and is a good measure of the repeatability of the system.
We have compared our method with results we have obtained using the Lightcycler which measures real time PCR products from the reaction. The results within
this small sample reveal a remarkable similarity from the two methods, indicating
that that the gel image scanning method is a valid and reliable method of sjTREC
analysis.

Acknowledgements
We would like to thank D. King for help with the cell separation studies, and Dr
D. Douek for providing help with the TREC analysis. This work was supported by
the Luard family (J.P. is the Luard scholar) and the Welcome Trust (grant number
051541).

R. Aspinall et al. / Mechanisms of Ageing and De6elopment 121 (2000) 59–67

67

References
Al Harthi, L., Marchetti, G., Steffens, C.M., Poulin, J., Sekaly, R., Landay, A., 2000. Detection of T cell
receptor circles (TRECs) as biomarkers for de novo T cell synthesis using a quantitative polymerase
chain reaction-enzyme linked immunosorbent assay (PCR-ELISA). J. Immunol. Meth. 237, 187 –
197.
Bertho, J.M., Demarquay, C., Moulian, N., Van De Meeren, A., Berrih-Akhnin, S., Gourmelon, P.,
1997. Phenotypic and immunohistological analyses of the human adult thymus: evidence for an
active thymus during adult life. Cell. Immunol. 179, 30 – 40.
Cossarizza, A., Ortolani, C., Paganelli, R., Barbieri, D., Monti, D., Sansoni, P., Fagiolo, U., Castellani,
G., Bersani, F., Londei, M., Franceschi, C., 1996. CD45 isoforms expression on CD4+ and CD8+
T cells throughout life, from newborns to centenarians: implications for T cell memory. Mech.
Ageing Dev. 86, 173–195.
Douek, D.C., McFarland, R.D., Keiser, P.H., Gage, E.A., Massey, J.M., Haynes, B.F., Polis, M.A.,
Haase, A.T., Feinberg, M.B., Sullivan, J.L., Jamieson, B.D., Zack, J.A., Picker, L.J., Koup, R.A.,
1998. Changes in thymic function with age and during the treatment of HIV infection [see
comments]. Nature 396, 690–695.
Hosseinzadeh, H., Goldschneider, I., 1993. Recent thymic emigrants in the rat express a unique antigenic
phenotype and undergo post-thymic maturation in peripheral lymphoid tissues. J. Immunol. 150,
1670–1679.
Hulstaert, F., Hannet, I., Deneys, V., Munhyeshuli, V., Reichert, T., De Bruyere, M., Strauss, K., 1994.
Age-related changes in human blood lymphocyte subpopulations. II. Varying kinetics of percentage
and absolute count measurements Age-related changes in human blood lymphocyte subpopulations.
II. Varying kinetics of percentage and absolute count measurements. Clin. Immunol. Immunopathol.
70, 152–158.
Kendall, M.D., Johnson, H.R., Singh, J., 1980. The weight of the human thymus gland at necropsy. J.
Anat. 131, 483–497.
Kong, F., Chen, C.H., Cooper, M.D., 1998. Thymic function can be accurately monitored by the level
of recent T cell emigrants in the circulation. Immunity 8, 97 – 104.
Livak, F., Schatz, D.G., 1996. T-cell receptor alpha locus V(D)J recombination byproducts are
abundant in thymocytes and mature T cells. Mol. Cell Biol. 16, 609 – 618.
McFarland, R.D., Douek, D.C., Koup, R.A., Picker, L.J., 2000. Identification of a human recent thymic
emigrant phenotype. Proc. Natl. Acad. Sci. USA 97, 4215 – 4220.
Nociari, M.M., Telford, W., Russo, C., 1999. Postthymic development of CD28−CD8+ T cell subset:
age-associated expansion and shift from memory to naıve phenotype. J. Immunol. 162, 3327 – 3335.
¨
Poulin, J.F., Viswanathan, M.N., Harris, J.M., Komanduri, K.V., Wieder, E., Ringuette, N., Jenkins,
M., McCune, J.M., Sekaly, R.P., 1999. Direct evidence for thymic function in adult humans. J. Exp.
Med. 190, 479–486.
Scollay, R.G., Butcher, E.C., Weissman, I.L., 1980. Thymus cell migration. Quantitative aspects of
cellular traffic from the thymus to the periphery in mice. Eur. J. Immunol. 10, 210 – 218.
Steinmann, G.G., 1986. Changes in the human thymus during aging. Curr. Top. Pathol. 75, 43 – 88.
Wills, M.R., Carmichael, A.J., Weekes, M.P., Mynard, K., Okecha, G., Hicks, R., Sissons, J.G., 1999.
Human virus-specific CD8+ CTL clones revert from CD45ROhigh to CD45RAhigh in vivo:
CD45RAhighCD8+ T cells comprise both naıve and memory cells. J. Immunol. 162, 7080 – 7087.
¨
Zhang, L., Lewin, S.R., Markowitz, M., Lin, H.H., Skulsky, E., Karanicolas, R., He, Y., Jin, X.,
Tuttleton, S., Vesanen, M., Spiegel, H., Kost, R., van Lunzen, J., Stellbrink, H.J., Wolinsky, S.,
Borkowsky, W., Palumbo, P., Kostrikis, L.G., Ho, D.D., 1999. Measuring recent thymic emigrants
in blood of normal and HIV-1-infected individuals before and after effective therapy. J. Exp. Med.
190, 725–732.

.