Expression levels of asparagine synthetase in blasts from children and adults with acute lymphoblastic leukaemia


Sally A. Coulthard, NICR, Paul O'Gorman Building, Medical School, Framlington Place, Newcastle Upon Tyne, NE2 4HH, UK. E-mail:


l-asparaginase is active in the treatment of acute lymphoblastic leukaemia (ALL) through the depletion of serum asparagine. Here we report that median asparagine synthetase (AS) mRNA levels were higher in acute myeloid leukaemia (AML) than ALL blasts in both children and adults, with intermediate levels in normal peripheral blood mononuclear cells (NPBMC). NPBMC versus child ALL (Tukeys multiple comparison test, P < 0·05); child ALL versus child AML (P < 0·001) and adult ALL versus adult AML (P < 0·01) were all significant and support the hypothesis that selectivity to treatment with l-asparaginase is due, at least in part, to lower AS expression.

l-asparaginase (l-asp) is an important component in the treatment of childhood acute lymphoblastic leukaemia (ALL). Recently the assumption that leukaemic cells are sensitive to l-asp because of a relative lack of asparagine synthetase (AS) has been questioned (Stams et al, 2003; Krejci et al, 2004). Using a real-time quantitative reverse transcription polymerase chain reaction (RQ RT-PCR) assay with transcription binding protein IID/TATA (TBP) as the internal control we provide evidence to support the original supposition that lymphoblasts have significantly lower levels of expression of AS than normal peripheral blood mononuclear cells (NPBMC).

Materials and methods

Following ethical approval, diagnostic bone marrow samples were obtained from 59 children (median 5 years; range 0·6–15·5 years) and 23 adults (median 29 years; range 17·7–71 years) with ALL and eight children (median 5·2 years; range 1–16 years) and 22 adults (median 57; range 17–84 years) with acute myeloid leukaemia (AML). The presence of >80% blasts were confirmed by morphology. NPBMC were obtained from 14 healthy volunteers. Mononuclear cells were isolated by density gradient centrifugation using Lymphoprep (Nycomed Pharma, Oslo, Norway). K562, MOLT-4 and NALM-6 cell lines were grown under standard conditions for use as controls.

Total mRNA was extracted using the RNeasy mini kit (Qiagen, Crawley, UK) according to the manufacturer's protocol. Reverse transcription of 200 ng total RNA was performed using a Taqman reverse transcription kit (Applied Biosystems, Warrington, UK) with random hexomers. Primers and Taqman probe for AS RQ RT-PCR, were designed using Primer-Express (Applied Biosystems). Primer and probe sequences were as follows; forward –5′-TCAGCCCGCCACATCAC-3′– (in exon 1), reverse –5′-CAATGAAGCTATAAGCTTTCTTCAAGTG-3′ (spanning exon 2 and 3), probe, –5′-CTGACCTGCTTACGCCCAGATTTTCTTCAA-3′ (spanning exon one and two). Primers and probe for TBP were purchased as a Taqman pre-developed control reagent.

Standards containing 75, 18·75, 4·69, 1·17, 0·29 and 0·07 ng of K562 total RNA with reverse transcribed Escherichia coli tRNA as a carrier were assayed for AS and TBP in triplicate on 27 separate occasions. Computerised tomography (CT) values (the fractional cycle numbers at which the fluorescence passed the fixed threshold) for each standard were obtained using an ABI Prism 5700 sequence detection system (Applied Biosystems, Warrington, UK) and the average CT for each point plotted against log standard RNA concentration.

Reaction mixtures (82·5 μl) contained 41·25 μl Taqman PCR universal master mix, 900 nmol/l AS forward primer, 900 nmol/l AS reverse primer, 225 nmol/l AS probe or 4·13 μl of ready synthesised TBP primers and probe and either 7·3 μl of sample cDNA, 22·5 μl of standard cDNA or 3·2 μl of calibrator cDNA. Aliquots (25 μl) were added to wells of a microplate in triplicate. The thermal cycling conditions were 95°C for 10 min, 40 cycles at 95°C for 15 s and 60°C for 1 min.

A calibrator sample was quantified for AS and TBP in each assay. AS gene expression relative to the TBP gene were calculated as follows:


Results and discussion

The reliability of the assay was determined by quantifying AS mRNA levels in MOLT-4, NALM 6 and K562 cell lines. Mean rAS values were 0·04, 0·50 and 0·84, respectively (n = 7). Intra-assay coefficients of variation (CV) were 16·5% 13·4% and 10·1%, respectively, (n = 3) and inter-assay CV values were 32·1%, 9·04% and 9·97% respectively (n = 7). For childhood ALL, rAS values ranged from 0·013 to 0·84 (65-fold, median = 0·17, n = 59) and adult ALL 0·07 to 1·55 (22-fold, median = 0·19, n = 23) (Fig 1). For childhood AML, rAS values ranged from 0·19 to 9·4 (49-fold, median = 0·47, n = 8) and adult AML 0·19 to 2·30 (12-fold, median = 0·50, n = 22). Little variation was observed in NPBMC, where values ranged from 0·27 to 0·49 (twofold, median = 0·40, n = 14).

Figure 1.

Relative AS mRNA levels (rAS) of samples from children and adults with leukaemia and peripheral mononuclear cells from healthy individuals. Results are the average of three independent experiments with each sample quantified in triplicate. The significance of the comparisons indicated was calculated using Tukeys multiple comparison test. ALL, acute lymphoblastic leukaemia; AML, acute myeloid leukaemia.

There was a higher median rAS in the AML patient samples than the ALL patient samples tested for both children and adults. The median rAS for NPBMC was higher than that for children and adults with ALL but less than that for children and adults with AML. These results were significant [one way analysis of variance (anova) following log10 transformation, P < 0·0001; Fig 1]. No significant difference in median rAS was observed when adults and children with either ALL or AML (P > 0·05 in both cases) were compared.

No significant difference in rAS was detected between gender, immunophenotype, age at presentation or presenting white cell count for either adults or children with ALL or AML. No relationship was observed between rAS and cytogenetics, including the presence of the t(12;21) translocation or percentage blasts in the bone marrow at day 8 of therapy for children with ALL (data not shown).

Horowitz and Meister (1972) suggested that leukaemic cells lack AS activity compared with normal cells and therefore rely on circulatory asparagine for survival. l-asp depletes asparagine from blood and bone marrow, starving the leukaemic cells of asparagine, causing them to undergo cell cycle arrest at G1 and, ultimately, apoptosis (Ueno et al, 1997). In vitro, resistance to l-asp has been shown to be highly correlated with cellular AS activity, mRNA and protein content (Hutson et al, 1997) and sensitive cell lines made resistant by repeated sub-culturing in sub-lethal doses of l-asp display increased AS expression (Kiriyama et al, 1989).

Our results contradict those reported by both Stams et al (2003) and Krejci et al (2004) in that AS mRNA levels in NPBMC were found to be significantly higher than those in lymphoblasts, in line with the expectation from many in vitro studies. As previously discussed (Krejci et al, 2005a,b, Stams et al, 2005) this may be due to the choice of the endogenous control used to normalise AS expression. We selected TBP as a suitable control as the variation of TBP expression between samples of lymphoid origin has been shown to be lower than either GAPDH or β2microglobulin (Lossos et al, 2003) and, in our RQ RT-PCR assay, the abundance of TBP was similar to that of AS.

In summary, our observations support the original suggestion that AS expression is lower in lymphoblasts than in NPBMC and that this may explain their increased sensitivity to l-asp therapy.


Research support was kindly provided by the Leukaemia Research Fund and North of England Children's Cancer Research.