Detection of c-MYC Gene in Micronucleated Hepatocytes from Regenerative Cirrhotic Nodules and Hepatocellular Carcinoma of Hepatitis C Virus Infected Patients

1Department of Oncology, Radiology and Oncology, Faculty of Medicine, University of Sao Paulo, Sao Paulo, Brazil 2Laboratory of Pathology Diagnóstika, Sao Paulo, Brazil 3Institute of Pediatric Oncology, University of Sao Paulo, Sao Paulo, Brazil 4Laboratory of Genetics, Instituto Butantan, São Paulo, Brazil 5Department of Clinical Gastroenterology-Hepatology, Faculty of Medicine, University of Sao Paulo, Sao Paulo, Brazil


Introduction
It is a well established fact that hepatitis C virus (HCV) is a risk factor for the development of liver cirrhosis (LC) and hepatocellular carcinoma (HCC). This type of tumor is the 7 th cause of death in the world [1]. Approximately 70% of the patients that contract acute HCV infection develop chronic hepatitis (CH). Persistent HCV infection leads to chronic inflammation with destruction of hepatocytes, and an irregular regeneration of the cells. As a result of the chronic viral infection, proliferating liver cells may become more susceptible to acquire chromosomal instability and genetic alterations, a condition known as hypercarcinogenic state [2]. Chromosome instability is one of the first events observed in HCC [3], although cytogenetic alterations can already be seen by micronucleus (MN) formation in hepatocytes of RNs and MRNs from cirrhotic patients infected or not by HCV [4,5].
Micronuclei are structures that arise from acentric fragments, or whole chromosomes, that are not included in the main nucleus during cell division [6]. Fluorescence in situ hybridization (FISH) has been used as a powerful tool for the identification of the MN content [7,8], which may be associated with the cytogenetic abnormalities observed during carcinogenesis [9][10][11]. Loss of chromosomes by MN has proved to be a good indicator of response to the action of mutagens and carcinogens, as seen in hepatoma HepG2 cells [12,13].
Imbalances of the oncogene c-MYC, which is considered a central element in the complexity of the tumorigenic process, has been investigated during the development of tumor in several human cancers [14][15][16][17], including HCC [18][19][20][21]. The amplification of c-MYC gene in HCC was also described by other authors [22][23][24][25], although the specific extrusion of its sequence by MN in LC, as well as during human hepatocarcinogenesis has not been described. The loss of c-MYC amplified sequence by MN, showed that this mechanism can be useful to moderate tumor growth [26], and it can be used as a therapeutic strategy [27].
The main objective of this study is to detect loss of c-MYC gene sequence by MN-Heps from regenerative, macroregenerative, and tumoral nodules (RNs, MRNs, and TNs). The investigation was accomplished in LC cases associated or not with HCC. The comparison among the frequency of that event in all nodules was made in relation to normal parenchyma (NP).

Procedures
All samples of liver tissue were fixed in formalin (10%), and embedded in paraffin blocks. Slices 4 µm thick were placed onto slides previously treated with 3-aminopropriltrietoxiilano (Sigma USA). Selection of study areas was based on histopathological analysis of lesions stained with hematoxylin-eosin as reference. Slices were examined for the presence of RNs, and MRNs. Nodules were classified into categories, according to morphological and histological criteria proposed by the International Working Party [28]. Diagnosis of HCC grading was done according to the standard histological criteria [29].

Detection of c-MYC gene by FISH
The detection of gain or loss of c-MYC gene by MN-Heps from RNs, MNRs, and TNs from LC, associated or not with HCC, were performed in paraffin-embedded tissue sections. Paraffin was removed as described by Almeida et al. [4], in order to accomplish fluorescence in situ hybridization (FISH) assay. The locus-specific probe (LSI) c-MYC  labeled with Spectrum Orange (Vysis Inc., Downers Grove, IL, USA), was then applied to the samples according to the manufacturer protocol. Nuclei and micronuclei were identified based on the absence or presence of fluorescence signal. Loss of c-MYC gene by MN corresponds to MN +c-MYC in photographs. Labeling was considered positive when more than 70% of hybridization was observed. For each nodular area (RNs, MRNs and TNs) frequency of MN in 2,000 interphase mononucleated hepatocytes (Mono-Heps) was assessed. This gives a total number of 20,000 cells (group I), 30,000 cells (group II), and 10,000 cells (group III). Nuclear gain of c-MYC gene copies was surveyed in 100 Mono-Heps of the same area. Analysis of +c-MYC gene in nuclei and MN was done under an epifluorescence Zeiss III light microscope equipped with UV epi-illumination, triple filter (DAPI, FITC and Texas Red) and immersion objective (100X). Fluorescence was evaluated after 475-495 nm excitation and 520 nm filtering. Kodak Gold 100 ASA film was employed for image documentation.

Statistical analysis
Data were subjected to statistical analysis by using both commercial (SPSS, Inc. Chicago, IL: USA) and public domain software [30]. They were considered statistically significant when p was ≤ 0.05.

Detection of c-MYC gene in MN-Heps
Hybridization of Mono-Heps of all paraffin-embedded liver tissue samples from groups I, II and III with fluorescent c-MYC probe was at least 70% efficient. The extrusion +c-MYC gene by MN-Heps from nodules occurred in all cases of cirrhosis, and in those related to HCC (groups I and II). It was not observed in MN-Heps from the control group (III).
The vast majority of MN-Heps from nodules of groups I and II, shows a +c-MYC gene double signal in MN. This does not depend on its size and signal number of this gene seen on the main nucleus (Figures 1a and 1e). Few MN-Heps of cases from group II displaying a large number of +c-MYC signals in their nuclei, have more than two copies of this gene per MN (Figure 1c).
MN-Heps with two or more +c-MYC signals in the main nucleus, may not have in MN (Figures 1b and 1d). In MN-Heps from MRNs, more than two +c-MYC signals were found. MN of different sizes have or not the +c-MYC signal (Figure 1e). In RNs and TNs, some MN-Heps have two MN with only one signal in both main nucleus, and MN ( Figure 1f).
Comparison of the frequencies of +c-MYC gene in MN-Heps from RNs, MRNs from groups I and II, and the TNs (group II) showed that they were higher and significant in relation to control group (p<0.05). This fact was also evident when compared with frequencies in TNs and RNs from group I. Comparison of +c-MYC gene frequency between RNs (I) and MRNs (II) in MN-Heps, showed no significant differences (Table 1). Furthermore, the comparison between RNs and MRNs (I and II) also showed no significant difference among themselves (Table  1). Similarly, comparison between TNs and RNs (II), and with those of MRNs (II) were not different, ( Table 1).
As is shown in Table 1, all scored groups have c-MYC gene losses (p< 0.05) when compared to NP. The internal comparison among these groups indicates a similar variation of the +c-MYC gene (p~1.000) The increased number of +c-MYC signals in nuclei was detected in 28% of the Mono-Heps from RNs and MRNs, and 46% of TNs of all cases studied. In RNs from groups I and II, hepatocytes with +c-MYC were observed, ranging from 3-5 signals per nucleus in more than 10% of Mono-Heps. This was also evident in MRNs and TNs, but in a higher frequency. In the same area, some nuclei showed 6 to 9 +c-MYC gene signals (Figure 2), and a small proportion of hepatocytes with more than 9 copies of the gene. In the control samples, there was no increased number of +c-MYC signals, 98.6% of liver nuclei displaying two +c-MYC gene signals ( Figure 2). The results of this analysis are shown in Table 2.

Discussion
For the first time, we identify signals of copies of c-MYC gene lost by MN-Heps from cirrhotic liver samples of patients infected by HCV, with or without HCC. A retrospective analysis was efficient and enabled us to obtain an evaluation of MN-Heps of archived samples. Using interphase cytogenetic FISH method, we evaluated the frequency of elimination of c-MYC gene by MN-Heps from ordinary RNs and MRNs, and TNs respect to NP. We verified that the frequency of elimination of c-MYC gene by MN-Heps from RNs and MRNs of the cirrhotic groups (I and II) was not different among them (P>0.05), suggesting that this type of cytogenetic changes, occurs in early stages of cirrhotic process and, it does not depend on the presence of HCC. In our previous data, we had already considered the ordinary cirrhotic nodules (RNs and MRNs) cytogenetically abnormal as evaluated by MN formation [4].

The frequency of c-MYC gene elimination by MN-Heps from TNs
is not different from that detected in RNs from LC cases, and also in relation to RNs, and MRNs from LC/HCC associated cases (Table  1). These frequencies of loss of c-MYC gene by MN-Heps, probably should be related to the chronic HCV infection that leads to cell death, irregular regeneration, or chromatin degradation by nucleases from MN [26].
The elimination of c-MYC gene by MN-Heps, persistently present in RNs, MRNs, and TNs did not occur in healthy liver tissue. These differences might be related to chronic HCV infection, since the proliferation of hepatocytes in regenerative response to any injury, is an important event, especially in cirrhotic livers of patients infected with HCV [31,32], predisposing to the formation of MN. In the recent past, the size of MN has been considered as an indicator of the extent of chromosomal changes [6,33]. The long period of inflammation leading to complex cirrhosis process, may have contributed to the accumulation of genetic damage, justifying the formation of different MN sizes.
In our study, we detected MN-Heps with c-MYC gene represented by one or two copies of signals of this gene, and other cells with one signal in MN, and another in the main nucleus, in addition to hepatocytes with a couple of signals in the nucleus, and another in MN. These events may result from chromosomal rearrangements (as isochromosomes), and aneuploidy induced by deregulation of c-MYC gene [34]. Increased number of copies of the c-MYC gene has been reported in the hepatocyte nuclei of HCC samples [20,21,24,25], as well as in other types of cancer, and the amplification of c-MYC gene has been associated with advanced stages of tumor development [14,15,17]. In the present study, we also found increased signals of c-MYC gene (up to 9), not only in hepatocytes of HCC, but also of RNs and MRNs, considered as benign lesions, a histopathogical point of view. In a rare study, evaluating c-MYC gene amplification in multiple steps of breast cancer evolution, the authors did not found amplification, neither in hiperplasia, nor in normal adjacent cells. So it appears that c-MYC gene amplification is a late event, being detected only in invasive cells, in the breast cancer progression [35]. In all RNs and MRNs, we found increased signals of c-MYC gene, and we speculated on the possibility that the overexpression of c-MYC protein found in the same samples of RNs and MRNs (unpublished data), may be related to the alteration of  number of copies of c-MYC gene in these nodules in relation to normal parenchyma. During the hepatitis viruses, chromosomal instability is a common event mediated by recombinogenic proteins [2]. According to other authors [34,36,37], c-MYC protein has been considered a structural modifier of the genome that affects nuclear organization. Deregulation of the expression of its protein can mediate genomic instability, initiating both, intra-and extra-chromosomal amplification mechanisms.
Based on the patterns of signals of c-MYC gene found in hepatocytes, independent of cirrhotic group (with and without HCC), in this study, the high number of amplified signals of the c-MYC gene in MN or in the main nucleus are represented by double minutes (DMs). Since DMs are autonomous structures that do not have centromeres, and have a random distribution in the nucleus, they might provide their expulsion of hepatocyte nuclei. These extrachromosomal elements can be segregated in a stable form to daughter cells during mitosis, but the mechanism by which DMs are stick to the chromosome arm is not known. In addition, DM-MN type can be generated, both at mitosis and interphase of the cell cycle [8,33,38,39]. These authors suggest that the amplified DNA can be reduced by in situ degradation or elimination by MN. The excess of amplified DNA, by an active process, concentrates on the peripheral of the nucleus, and it may be expulsed by MN in a bud formation mechanism. The MN is extruded from the cytoplasm, during S phase of the cell cycle. We believe that such mechanisms might occur, also in hepatocytes. Nevertheless, none of the mechanisms of extrusion of amplified gene are well understood.
Similar events were shown in blast cells from patients with acute myeloid leukemia expulsion of clusters of signals of amplified c-MYC gene by MN, probably as DM [17]. Also, the induced or spontaneous elimination of amplified c-MYC gene by MN in Colo320DM and HL60 cell lines, suggests a strategy to reduce the tumorigenic potential and reversion of the malignant phenotype [26,40].
Our data are not sufficient to determine, with accuracy, the mechanisms by which c-MYC sequences were lost by MN-Heps. Nevertheless, we believe that in cirrhotic process, the formation of the MN as chromosome-type and/or DM-type by genetic and/or epigenetic mechanisms [39] is a likely process.
Considering the role of c-MYC in human hepatocarcinogenesis [41], it is important to note that no data are forthcoming about the elimination of this sequence (amplified or not), by MN and their fate in the liver tissue. From what we know, the MN is commonly observed in cancer, and it is considered an indicator of DNA damage, representing genetic instability, that contributes to the risk of development of degenerative diseases [8]. As the amplification of c-MYC has been related to the degree of aggressiveness in HCC [18,20,42], it is possible that the expulsion of that sequence by spontaneous MN formation, may have implications for the clinical prognosis of the patient. Cytogenetic and molecular mechanisms in the cirrhosis development have not yet been elucidated, so the importance of nuclear loss of a specific gene such as c-MYC, by MN, needs further studies to establish its prognostic impact.
Our data showed for the first time that, loss of c-MYC gene by MN-Heps, is an early event in hepatocarcinogenesis. Evaluation of c-MYC gene loss by MN-Heps associated to FISH technique has been proven to be useful to identify the type of cytogenetic instability in archived samples of cirrhotic liver tissue, and to estimate the potential risk of development of HCC.