This study was approved by the Institutional Review Board of The University of Texas MD Anderson Cancer Center. We prospectively obtained fresh bone marrow (BM) samples from 11 patients with AML with inv(16), two patients with acute myelomonocytic leukemia (AMML) without inv(16), one patient with a myeloproliferative neoplasm (MPN), one patient with hairy cell leukemia (HCL), and one patient with CLL/SLL for gene expression profiling by real-time polymerase chain reaction. We also performed immunohistochemical studies by collecting 63 bone marrow (BM) biopsy specimens that were formalin-fixed and paraffin-embedded from the archives. These included 27 cases of AML with inv(16), 10 cases of AMML, 22 cases of other types of AML, two cases of chronic myelomonocytic leukemia, and two negative and apparently normal BM specimens. Clinical, pathological and molecular results were obtained from the medical records.

Conventional cytogenetic analysis was performed on BM aspirate samples using standard techniques described previously [19]. At least 20 metaphases were analyzed. The International System for Human Cytogenetic Nomenclature was used for reporting the results [20].

RNA was isolated from fresh BM aspirate specimens using the TRIzol^® Plus RNA Purification Kit according to the manufacturer’s protocol (Invitrogen, Grand Island, NY, USA). cDNA synthesis was performed using the QuantiTect RT Kit (Qiagen, Valencia, CA, USA) according to manufacturer’s instructions [14]. The commercially produced normal BM CD34 cDNA (AllCells, LLC, Emeryville, CA, USA) was also used in this study. Quantitative measurements of CBFβ-MYH11 (Types A, C, and E) expression levels in BM samples were measured by using Taqman Universal PCR MasterMix (Applied Biosystems Inc., Carlsbad, CA, USA). Briefly, 2–5 µL cDNA was mixed in a total 25 µL qRT-PCR reaction mixture containing primers and probes (Ipsogen, New Brunswick, NJ, USA) in which C-ABL was used as the internal control, and performed in an ABI PRISM 7700 Sequence Detection System (Applied Biosystems, Foster City, CA, USA) under standardized conditions. The PCR reaction for detection of CYP2E1, CYP1A2, and NQO1, was similar to that described above but was performed in Mx3000P real time PCR (Stratagene, La Jolla, CA, USA) using a standard 2 step qPCR cycling conditions set by the instrument. The probes and forward and reverse primers in the latter reactions were synthesized by Applied Biosystems (ABI) as follows. CYP2E1 probe (P) 5’-6FAM-(865)TGGACGGTATCACCG-MGB, Forward (F): 5’-(843)AGTGCAGAGCGCTTGTACACA, and Reverse (R): 5’-(901)AAGAACAGGTCGGCCACAGT; CYP1A2 P: 5’-6FAM-(865) AGGCCTTCATCCTGG-MGB, F: 5’-(1130) ACCCCAGCTGCCCTACTT G-3’, and R: 5’-(1187) GAAGGAGGAGTGTCGGAAGGT; NQO-1: P: 5-6FAM-AGTCTTAGAACCTCAACTGACAT-MGB, F: 5-AGAGTGGCATTCTGCATTTCTG, and R: 5-CTGGAGTGTGCCCAATGCTA; β-Actin P: 5-6FAM-ATCAAGATCATTGCTCCTCCTGAGCGC, F: 5-CCCTGGCACCCAGCAC and R: 5-GCCGATCCACACGGAGTAC. All reactions were run twice with triplicates in each run. The ratio of copy number of target gene over copy number of internal control gene was used for expression level in each specimen.

Immunohistochemical staining for CYP2E1 was performed using 4 micron thick sections of formalin-fixed, paraffin-embedded (FFPE) BM biopsy or aspirate clot sections. Briefly, sections were placed in a 60 °C oven for 1 hour, cooled, and deparaffinized and rehydrated through xylenes and graded ethanol solutions to water. All slides were quenched for 5 minutes in 3% hydrogen peroxide for endogenous peroxidase. Antigen retrieval was performed using a heat-induced method in which the specimens were placed in a citric acid solution (pH 6.1) for 25 minutes at 94 °C and cooled for 15 minutes. Slides were then placed on Autostainer (Dako, Carpinteria, CA, USA) for immunohistochemistry. The antibody, anti-human CYP2E1 (Chemicon International, Temecula, CA, USA), was used at a dilution of 1:300 and incubated for 30 minutes at room temperature. The detection system was either Mach 4 alkaline phosphatase (Biocare Medical, Concord, CA, USA) with Vulcan Fast Red to produce a bright fushia precipitate or Envision Plus HRP (Dako) with diaminobenzidine (DAB) chromogen to produce a brown precipitate. Slides were counterstained with Richard Allen hematoxylin. Hepatic tissue was used as a control, showing intense cytoplasmic staining in normal hepatocytes.

The immunohistochemistry slides were reviewed using an eyepiece reticle (grid). Any intensity of cytoplasmic reactivity for CYP2E1 in blasts was considered positive. Staining of the blasts was assessed in three random microscopic fields and averaged. The proportion of blasts staining positive was scored semiquantitatively as positive, focal/weakly positive, or negative. Bone marrow stromal cells and endothelial cells were negative.

K562 (chronic myelogenous leukemia in erythroid blast crisis) cells were cultured at 37 °C with 2% CO2 in complete culture medium (10% FBS in RPMI-1640). Twenty-four hours following plating, hydroquinone (HQ, benzene-1,4-diol or quinol, Sigma, St. Louis, MO, USA), diluted by DMSO and (BQ, 1,4-benzoquinone, Sigma-Aldrich, diluted by ethanol) were added to the plates. The cells were exposed to these chemicals at two different concentrations, 0.1 and 1.0 uM. mRNA quantification of enzymes CYP2E1, CYP1A2 and NQO1 was performed at two different time points, twenty-four and forty-eight hours after exposure to HQ and BQ. RNA was isolated from K562 cells using TRIzol^® Plus RNA Purification Kit according to the manufacturer’s instructions (Invitrogen). This was followed by cDNA reverse transcription using SuperScript™ II Reverse Transcriptase (Invitrogen) with 1–10 μg RNA. Quantitative RT-PCR was performed using the primers as described above. Fold changes were used for expression differences between treated and control groups, and calculated as described by Pfaffl et al. [21].

The Chi square test was used to compare distinct groups. A p-value ≤ 0.05 was considered significant.

Eleven cases of AML with inv(16) were selected for gene expression profiling. The diagnosis of AML with inv(16)/CBFβ-MYH11 was established by demonstrating inv(16)(p13q22)/CBFβ-MYH11. Based on the 2008 World Health Organization (WHO) classification, this type of AML is defined by the presence of recurrent balanced chromosomal rearrangement via inv(16)(p13.1q22) or t(16;16)(p13.1;q22), irrespective of the blast count. Morphologically, this type of AML shows myelomonocytic differentiation and eosinophilia with abnormal granulation. The clinicopathologic features of these cases are summarized in Table 1. There were nine men and two women, with a median age of 42.5 years (range, 10–65). All patients presented de novo, without prior treatment. Eight of 11 patients had ≥20% blasts. The median BM blast count at presentation was 52% (range, 1–88). Five of 11 (45%) patients had eosinophilia (>4%). None of these patients had received any known CYP2E1 inducing drugs including isoniazid and ethanol.

Cases 1–8 with ≥20% blasts showed morphological evidence of AML in the BM aspirate samples available for testing. Case 9 had involvement by minimal residual AML, supported by the detection of inv(16) abnormality by conventional cytogenetic analysis. Cases 10 and 11 had a history of AML with inv(16) but were in complete remission as shown by the absence of CBFβ/MYH11 transcripts in the current BM sample.

CYP2E1 levels correlated well with CBFβ-MYH11 transcript levels in all cases of AML with inv(16), and were proportional to the BM blast count. As a control, we tested CYP2E1 mRNA from two cases of newly diagnosed untreated AMML without inv(16) and pooled RNA from 20 untreated AMML without inv(16). Of these, only one case of AMML showed minimal expression of CYP2E1 while the remaining cases were negative. RNA specimens from patients with other diseases including MPN, HCL and CLL/SLL cases were either negative or had minimal CYP2E1 expression (Figure 1).

Furthermore, in two patients with AML with inv(16), BM samples collected before and after standard chemotherapy with cytarabine were available. CYP2E1 mRNA expression was assessed before treatment, and subsequently at 1 and 3 months after starting conventional chemotherapy with cytarabine consolidation. In the first patient, the BM blast count and CBFβ-MYH11 transcript levels dramatically decreased within the first month of therapy with a corresponding decrease in CYP2E1 by almost 3 logs. In the second case, the patient did not show a significant decrease in the blast count and CBFβ-MYH11 transcript levels in the first month. However, with continued therapy, the patient showed a significant decrease in both of the above parameters; the patient went into complete remission at the subsequent bone marrow examination. Although the profiles of the two cases are different, in both of these cases, CYP2E1 expression levels decreased in parallel with the CBFB-MYH11 expression levels and blast counts following treatment (Figure 2A, case 1; and Figure 2B, case 2). In comparison, CYP1A2 expression levels did not change in either patient.

In seven of the 11 unsorted BM specimens, the expression levels of CYP2E1, CYP1A2 and NQO1 from each individual were compared. Over-expression of CYP2E1 correlated positively with expression of NQO1, (R² = 0.934, n = 7) and correlated negatively with CYP1A2 (R² = −0.506), (Table 2).

In 27 cases of AML with inv(16), CYP2E1 was strongly positive in the cytoplasm of 20 (74.0%), focally/weakly expressed in five (18.5%), and negative in two (7.4%) cases. In contrast, in 10 cases of AMML without inv(16), CYP2E1 was strongly positive in one (10%), focally/weakly positive in three (30%), and negative in six (p < 0.001). We also assessed CYP2E1 expression in other types of AML (n = 22), CYP2E1 was clearly positive in seven (31.8%), focally/weakly in three (13.6%) and negative in 12 (54.6%), but the overall intensity of expression was significantly less than that of cases of AML with inv(16) (p < 0.001). CYP2E1 was negative in two cases of CMML, and in two normal BM controls. The staining pattern and positivity in AML with inv(16), AMML without inv(16) and other types of AML are demonstrated in Figure 3 and the data are summarized in Table 3.

K562 cells were exposed to benzene analogues/metabolites 24 hours after plating. Benzene analogues/metabolites included the active benzoquinone/BQ) and inactive (hydroquinone/HQ). RNA expression levels of CYP2E1, CYP1A2 and NQO1 were measured at two different time points, 24 and 48 hours after exposure. As shown in Figure 4, exposure to the active form (BQ) of benzene induced elevated levels of CYP2E1, CYP1A2 and NQO1. Similarly, exposure to HQ induced elevated levels of CYP2E1, CYP1A2 and NQO1. Furthermore, exposure resulted in significantly higher levels of CYP1A2 compared with either CYP2E1 or NQO1 in these cell lines.