**6. COPD as a risk factor of lung cancer**

COPD is currently the leading cause of morbidity and mortality worldwide whose prevalence and burden are projected to increase due to smoker exposure and the changing age structure of the world population, particularly in women (Lopez et al, 2003). The presence of COPD increases the risk of lung cancer of up to 4,5 fold among long-term smokers. COPD is by far the greatеst risk factor for lung cancer amongst smokers in 50- 90% of smokers with lung cancer (Young et al, 2009). Even a small reduction of FEV1 is a marker of airflow obstruction and is a significant predictor of lung cancer (Wasswa et al, 2005).

Lung cancer accounts for 12% of cancers diagnosed worldwide, making it the most common malignancy other than non-melanoma skin cancer. Approximately over one million die of lung cancer each year (Jemal et al, 2009). Worldwide, tobacco smoking is associated with more of 90% of cases of lung cancer. In less developed countries lung cancer rates are predicited to continue to increase due to endemic tobacco use. In more developed countries, the incidence and mortality rate are generally declining, reflecting previous trends in smoking prevalence (Youlden et al, 2008). Only 15% of life-time smokers develop lung cancer and 10% of lung cancers occur in never smokers especially in women and in Asiatic women in particular, which underlines the role of genetics (Scagliotti et al, 2009).

Lung cancer is also a leading cause of morbidity and mortality in patients with COPD as 33% of patients died of lung cancer over a 14,5 year follow-up (Anthonisen et al, 2005; Yao et al, 2009). Furthermore ≈60% of patients, diagnosed with lung cancer have a spirometric evidence of COPD (Molina et al, 2008). NSCLC accounts for 85% of lung cancer cases in USA and the COPD related cancer type (squamous cell lung cancer) still represents the most common histological subtype of lung cancer in European men (Papi et al, 2004). Despite significant advances in diagnostic approaches, the pathology of lung cancer is still elusive and there has been little improvement in 5-year survival rates (≈ 15% overall; <14% among males and <18% among females) (Youlden et al, 2008).

Two of the leading causes morbidity and moratlity worldwide – COPD and lung cancer are due to the environmental risk factor and cigarette smoke exposure in combination with genetic predisposition.

### **6.1 HSP and cancer**

86 Chronic Obstructive Pulmonary Disease – Current Concepts and Practice

3. There are several similarities in inflammation between ageing and COPD, such as neutrophil accumulation, NF-kB activation (Barnes, 2006) and increase in IL-6/IL-8/TNF-α. COPD patients are also corticosteroid insensitive as similar to healthy aged people. Protein turnover system in COPD is also impaired. HDAC2 is markedly reduced in COPD. This reductionis involved in enhancing inflammation is involved in

4. Furthermore, expression of antiageing molecules are reduced in COPD - SIRT1 is a major inhibitory regulator of MMP-9, and reduction in SIRT1 causes structural changes of lung, such as emphysema (Vuppusetty C, et al 2007; Rajendrasozhan S, et al. 2008). SIRT6 loss leads to abnormalities in mice that overlap with ageing-associated degenerative processes, and SIRT6 is a nuclear, chromatin-associated protein that

Analysing the presented data we can assume that COPD could be regarded as a chaperonopathy (proteinopathy), as a model of accelerated ageing, in which the organism can not keep the homeostasis under the conditions of oxidative stress. The immune system is involved, but instead of restoring the balance it augments the oxidative stress, generating a large number of reactive – oxygen species. This leads to the accumulation of modified selfproteins that are recognized as antigenic. Autoimmunity occurs as an epiphenomenon. A vicious circle is created. The environmental and the inflammatory oxidative stress leads to the accumulation of modified proteins. Chaperones are additionally depleted or also

COPD is currently the leading cause of morbidity and mortality worldwide whose prevalence and burden are projected to increase due to smoker exposure and the changing age structure of the world population, particularly in women (Lopez et al, 2003). The presence of COPD increases the risk of lung cancer of up to 4,5 fold among long-term smokers. COPD is by far the greatеst risk factor for lung cancer amongst smokers in 50- 90% of smokers with lung cancer (Young et al, 2009). Even a small reduction of FEV1 is a marker of airflow obstruction and is a significant predictor of lung cancer (Wasswa et al,

Lung cancer accounts for 12% of cancers diagnosed worldwide, making it the most common malignancy other than non-melanoma skin cancer. Approximately over one million die of lung cancer each year (Jemal et al, 2009). Worldwide, tobacco smoking is associated with more of 90% of cases of lung cancer. In less developed countries lung cancer rates are predicited to continue to increase due to endemic tobacco use. In more developed countries, the incidence and mortality rate are generally declining, reflecting previous trends in smoking prevalence (Youlden et al, 2008). Only 15% of life-time smokers develop lung cancer and 10% of lung cancers occur in never smokers especially in women and in Asiatic women in particular, which underlines the role of genetics

with the severity of airways obstruction (FEV1 percentage of predicted).

enhancing inflammation and corticosteroid insensitivity.

promotes resistance to DNA damage (Meyer et al, 1988).

modified. Ubiquitin proteasome system is overloaded or also modified.

**6. COPD as a risk factor of lung cancer** 

2005).

(Scagliotti et al, 2009).

in healthy adults,(Kirkril et al, 2008) although this is still controversial (Nadeem et al, 2005).COPD patients also have reduced total antioxidant capacity. Furthermore, ferricreducing antioxidant power is lower in COPD patients, and it had a positive correlation

> There is a cascade of molecular events that mediate the transformation of a normal cell to a cancerous one. Several etiological factors and events are recognized as triggering mechanism of cancerogenesis – viruses, radiation, hereditary and non-hereditary mutations, carcinogenic compounds. Most tumors are formed by stepwise progression of normal cell into a cancer one by using alterations in cell physiology, described by (Hanahan and Weinberg, 2000) – self sufficiency in growth signals, insensitivity to growth inhibiton, evasion from apoptosis, limitless replicative senescence, sustained angiogenesis and tissue invasion and metastasis. Heat shock reponse participates in cancerogenesis of both up – and downregulation of specific heat shock proteins. Variations in HSP expression could be found in many tumors and preneoplastic lesions as well. At a histological level the transition from a normal tissue to tumor is accompanied by the increase in HSP expression. HSP are involved in the cancerogenesis and are up-regulated to protect cells from apoptosis and induce drug resistance. Their role in cancer cells is to protect other proteins against aggregation, to solubilize initial protein aggregates, to assist in folding of nascent proteins or refolding of damaged proteins; to target severely damaged proteins to degradation. Overexpression of HSP in cancer cells is beneficial to their survival because they inhibit apoptosis and induce drug resistance. They act as a double-side sword. Some of them - HSP90 maintains chaperoning function in a number of oncogenic molecules and promotes tumor survival. Others – HSP60,70 and HSP72 may sensitize cancer cells to immune attacks by two mechanisms. They may be expressed on tumor cell surface and enhance their recognition by NK-cells or induce antitumor immunity by HSP related tumor vaccines (Calderwood et al, 2006).

#### **6.1.1 HSP and non-small cell lung cancer**

Bonay et al, 1994 are the first to study the expression profile of HSP in the normal lung as well as the effect of cigarette smoke on their expression. They provide detailed description of HSP distribution in lung carcinoma, applying both immunohistochemical and immunoflourescent techniques for their investigation. In lung tissue from non-smokers, lung epithelial cells are positive for HSP90 and the inducible form of HSP70. There was also a weak expression of HSP60. Macrophages also expressed these HSPs but weaker in

Chronic Obstructive Pulmonary Disease - Chaperonopathology 89

analysis on a tissue microarray slide, containing samples from 146 NSCLC patients - 96 squamous cell lung tumours, 10 adenosquamous carcinomas, 35 adenocarcinomas and five broncho-alveolar carcinomas. Tumors were of different grade of differentiation (29 - well differentiated, 56 -moderate and 36 - poor differentiation) and different clinical stage - 37

αB-crystallin was not detected in the normal alveolar pneumocytes; a few of the peribronchial glands, however, stained faintly but only in the cytoplasm. Although partially, there were a few areas where basal epithelial cells of the normal ciliated bronchial epithelium also showed weak cytoplasmic staining and no nuclear staining. In contrast, the basal layer of the tumours showed intensive cytoplasmic staining and a lack of nuclear staining. Lymphoid cells infiltrating the tumour stroma as well as the macrophages showed no cytoplasmic staining, but the nuclear staining varied from intensive to a lack of staining. Apoptotic and necrotic cells had faint cytoplasmic and intensive nuclear staining. Intensive

Nuclear staining was detected in 133 tumours (95 squamous cell histology and 38 adenocarcinomas). Cytoplasmic staining was detected in 127 tumours (95 squamous cell histology and 32 adenocarcinomas). Lack of nuclear staining was detected in 44 (33%) cases and intensive nuclear staining was observed in 89 (67%). A total of 26 tumours strongly expressed αB-crystallin in both nucleus and cytoplasm. Most of the tumours showed homogeneous cytoplasmic staining; more than 60% of the cells of the tumour had the same intensity of staining. The heterogeneity was detected up to the level of nuclear staining. The cytoplasmic staining was not of statistically significant correlation to histology. In contrast, the nuclear staining proved to be characteristic for the adenocarcinomas (p < 0.001,

αB-crystallin was significantly overexpressed in NSCLC. In these tumors, the cytoplasmic expression of αB-crystallin was statistically significantly related to the tumour size (Tfactor). This might be due to the fact that αB-crystallin has been reported to serve as a chaperone under stress conditions for other oncogenic molecules (beta-catenin, cyclin D1

In comparison to breast, renal and colorectal cancers, where only cytoplasmic and membraneous staining was reported, in NSCLC a nuclear staining was observed. The nuclear relocalisation is a characteristic feature for the whole group of small heat-shock proteins and in most cases is triggered under stress conditions (Klemenz R et al 1991, Voorter Ch et al 1992). In the nucleus, αB-crystallin is claimed to be responsible for the stabilisation of the speckled architecture of lamin A/C and is thus involved in the splicing factor compartment (Adhikari A, et al 2004). IJssel et al, 2003 discuss that its fundamental role in the nucleus (transcription, splicing and genomic stability) is difficult to be discerned

The precise biologic function of both cytoplasmic and nuclear localisation of the protein is obscure and needs other approaches for elucidation. Moreover, the variability of cellular compartment expression is complicated by the fact that in many epithelial tumours the protein is down-regulated and lacks cytoplasmic expression – buccal cancer and head and neck cancer. The importance of αB in NSCLC may be due to the fact that the nuclear staining was characteristic for adenocarcinoma histology and was significantly related to the

patients were in stage I, 27 in stage II, 65 in stage III and 17 in stage IV.

nuclear staining was also detected in cells undergoing mitosis.

Contigency Coeff Cramer 0.369).

and VEGF) or is itself oncogenic (Ghosh J, 2007).

from its chaperone function in that cellular compartment.

comparison to bronchial epithelium. However no other parenchymal, immune or inflammatory cell was positive for these heat shock proteins.

Cigarette smoking modifies neither the distribution, nor the intensity of staining in bronchial epithelium in smokers. Macrophages also expressed one or more of the HSP but in low levels. Bonay et al, 1994 have shown considerable heterogenity in the expression of HSP by cells in a given tumor. They explained their observations by different degree of differentiation state of the cells.

HSP90 is required for conformational maturation as well as stability of many proteins, involved in signalling pathways. It is responsible for the functional activity of a lot of oncogenic kinases that drive the signal transduction and proliferation of lung cancer cells. It seems to be upregulated in lung cancer and recently it has been connected with the stabilization of the mutant form of EGFR and one of the mechanisms for the resistance to tyrosine kinase inhibitors.(Shumamura et al, 2005, 2008)

HSP70 is another chaperone of interest in lung cancer. Volm et al, 1995 studied the resistance of lung cancer and its associatation to HSP70 expression. Tumor samples of 90 patients with NSCLC were investigated by immunohistochemistry, and no association between HSP70 and doxorubicin resistance was found. However a trend for an association between glutathione-S-transferase positivity and HSP70 positivity was observed. In addition there was a strong positive association between catalase positivity and HSP70 positivity. These observations show that both heat shock and stress promote intracellular oxidative damage and catalases are necessary for their protection.

Malusecka et al, 2001 studied the expression profile of both HSP70 and HSP27 proteins in 106 patients with NSCLC. They found in the majority of patients (95/106) both cytoplasmic and nuclear positivity to HSP70. In stage I tumors and dysplastic lesions however there was an enhanced nuclear positivity. As for HSP27 a positive cytoplasmic immunostaining in 70% of cases with the highest score for squamous cell lung cancer was found. A positive association for the expression levels of both proteins was also described. HSP27 and HSP70 were indicated as important factors for lung tumor transformation process, as well as for factors for chemoresistance.

Capello et al, 2006 studied the role of HSP60 and HSP10 in lung cancerogenesis. They described the level of expression of these chaperones in 35 patients with spirometirically proven COPD and compared them to the levels of expression in 10 adenocarcinomas and adenosquamous cell lung cancers. In normal bronchial epithelium that was found in 10 of the COPD patients HSP60 and HSP10 were positive in 23% of cases. In basal hyperplasia lesions they were positive respectively in 29% and 26%. Only 3% of squamous metaplastic lesions were positive for HSP60 and only 2% positive for HSP10. Of the dysplastic lesions 3% were positive for HSP60; 2% - for HSP10. Adenosquamos cell lung cancer was negative for both chaperones. The authors showed that HSP60 and HSP10 loss is related to the development and progression of bronchial cancer in COPD patients.

Recently Jackson and Garcia Rojas investigated the role of HSP27 in cellular resistance in lung cells, and reported that HSP27, which is phosphorylated by MAPK pathway protect epithelial cells from oxidant stress.

The role of another small heat shock protein in NSCLC was described by Cherneva et al, 2010.The expression of alpha-B crystalline was explored applying immunohistochemical

comparison to bronchial epithelium. However no other parenchymal, immune or

Cigarette smoking modifies neither the distribution, nor the intensity of staining in bronchial epithelium in smokers. Macrophages also expressed one or more of the HSP but in low levels. Bonay et al, 1994 have shown considerable heterogenity in the expression of HSP by cells in a given tumor. They explained their observations by different degree of

HSP90 is required for conformational maturation as well as stability of many proteins, involved in signalling pathways. It is responsible for the functional activity of a lot of oncogenic kinases that drive the signal transduction and proliferation of lung cancer cells. It seems to be upregulated in lung cancer and recently it has been connected with the stabilization of the mutant form of EGFR and one of the mechanisms for the resistance to

HSP70 is another chaperone of interest in lung cancer. Volm et al, 1995 studied the resistance of lung cancer and its associatation to HSP70 expression. Tumor samples of 90 patients with NSCLC were investigated by immunohistochemistry, and no association between HSP70 and doxorubicin resistance was found. However a trend for an association between glutathione-S-transferase positivity and HSP70 positivity was observed. In addition there was a strong positive association between catalase positivity and HSP70 positivity. These observations show that both heat shock and stress promote intracellular oxidative

Malusecka et al, 2001 studied the expression profile of both HSP70 and HSP27 proteins in 106 patients with NSCLC. They found in the majority of patients (95/106) both cytoplasmic and nuclear positivity to HSP70. In stage I tumors and dysplastic lesions however there was an enhanced nuclear positivity. As for HSP27 a positive cytoplasmic immunostaining in 70% of cases with the highest score for squamous cell lung cancer was found. A positive association for the expression levels of both proteins was also described. HSP27 and HSP70 were indicated as important factors for lung tumor transformation process, as well as for

Capello et al, 2006 studied the role of HSP60 and HSP10 in lung cancerogenesis. They described the level of expression of these chaperones in 35 patients with spirometirically proven COPD and compared them to the levels of expression in 10 adenocarcinomas and adenosquamous cell lung cancers. In normal bronchial epithelium that was found in 10 of the COPD patients HSP60 and HSP10 were positive in 23% of cases. In basal hyperplasia lesions they were positive respectively in 29% and 26%. Only 3% of squamous metaplastic lesions were positive for HSP60 and only 2% positive for HSP10. Of the dysplastic lesions 3% were positive for HSP60; 2% - for HSP10. Adenosquamos cell lung cancer was negative for both chaperones. The authors showed that HSP60 and HSP10 loss is related to the

Recently Jackson and Garcia Rojas investigated the role of HSP27 in cellular resistance in lung cells, and reported that HSP27, which is phosphorylated by MAPK pathway protect

The role of another small heat shock protein in NSCLC was described by Cherneva et al, 2010.The expression of alpha-B crystalline was explored applying immunohistochemical

inflammatory cell was positive for these heat shock proteins.

tyrosine kinase inhibitors.(Shumamura et al, 2005, 2008)

damage and catalases are necessary for their protection.

development and progression of bronchial cancer in COPD patients.

differentiation state of the cells.

factors for chemoresistance.

epithelial cells from oxidant stress.

analysis on a tissue microarray slide, containing samples from 146 NSCLC patients - 96 squamous cell lung tumours, 10 adenosquamous carcinomas, 35 adenocarcinomas and five broncho-alveolar carcinomas. Tumors were of different grade of differentiation (29 - well differentiated, 56 -moderate and 36 - poor differentiation) and different clinical stage - 37 patients were in stage I, 27 in stage II, 65 in stage III and 17 in stage IV.

αB-crystallin was not detected in the normal alveolar pneumocytes; a few of the peribronchial glands, however, stained faintly but only in the cytoplasm. Although partially, there were a few areas where basal epithelial cells of the normal ciliated bronchial epithelium also showed weak cytoplasmic staining and no nuclear staining. In contrast, the basal layer of the tumours showed intensive cytoplasmic staining and a lack of nuclear staining. Lymphoid cells infiltrating the tumour stroma as well as the macrophages showed no cytoplasmic staining, but the nuclear staining varied from intensive to a lack of staining. Apoptotic and necrotic cells had faint cytoplasmic and intensive nuclear staining. Intensive nuclear staining was also detected in cells undergoing mitosis.

Nuclear staining was detected in 133 tumours (95 squamous cell histology and 38 adenocarcinomas). Cytoplasmic staining was detected in 127 tumours (95 squamous cell histology and 32 adenocarcinomas). Lack of nuclear staining was detected in 44 (33%) cases and intensive nuclear staining was observed in 89 (67%). A total of 26 tumours strongly expressed αB-crystallin in both nucleus and cytoplasm. Most of the tumours showed homogeneous cytoplasmic staining; more than 60% of the cells of the tumour had the same intensity of staining. The heterogeneity was detected up to the level of nuclear staining. The cytoplasmic staining was not of statistically significant correlation to histology. In contrast, the nuclear staining proved to be characteristic for the adenocarcinomas (p < 0.001, Contigency Coeff Cramer 0.369).

αB-crystallin was significantly overexpressed in NSCLC. In these tumors, the cytoplasmic expression of αB-crystallin was statistically significantly related to the tumour size (Tfactor). This might be due to the fact that αB-crystallin has been reported to serve as a chaperone under stress conditions for other oncogenic molecules (beta-catenin, cyclin D1 and VEGF) or is itself oncogenic (Ghosh J, 2007).

In comparison to breast, renal and colorectal cancers, where only cytoplasmic and membraneous staining was reported, in NSCLC a nuclear staining was observed. The nuclear relocalisation is a characteristic feature for the whole group of small heat-shock proteins and in most cases is triggered under stress conditions (Klemenz R et al 1991, Voorter Ch et al 1992). In the nucleus, αB-crystallin is claimed to be responsible for the stabilisation of the speckled architecture of lamin A/C and is thus involved in the splicing factor compartment (Adhikari A, et al 2004). IJssel et al, 2003 discuss that its fundamental role in the nucleus (transcription, splicing and genomic stability) is difficult to be discerned from its chaperone function in that cellular compartment.

The precise biologic function of both cytoplasmic and nuclear localisation of the protein is obscure and needs other approaches for elucidation. Moreover, the variability of cellular compartment expression is complicated by the fact that in many epithelial tumours the protein is down-regulated and lacks cytoplasmic expression – buccal cancer and head and neck cancer. The importance of αB in NSCLC may be due to the fact that the nuclear staining was characteristic for adenocarcinoma histology and was significantly related to the

Chronic Obstructive Pulmonary Disease - Chaperonopathology 91

could be that the major characteristic of this pathology is the increased oxidative stress and chronic systemic inflammation, predominantly localized in the lungs.This may provoke reactive αB-crystallin protein overexpression in COPD patients, as one of its function is

Analysing the levels of antibodies of αB-crystallin in the plasma of patients with NSCLC and their clinicopathological chracteristic, Cherneva et al, found no significance between pathological parameters and this biological marker. This however was not the issue when concerning the lymphogenic spread of the disease. The levels of antibodies in patients with lymph node metastases was higher compared to those without them. The reason for this remains elusive and requires further investigation. The ROC curve analysis showed decent characteristics in discerning patients with and without metastatic spread –AUC 0.667 (95%CI – 0.515-0.820) sensitivity- 60% and specificity –70% at a cut-off 0.381. It should be however carefully taken in consideration that the clinical staging does not envisage the molecular one and the presence of already spread micrometastatic disease in N0 patients is obscure. Whether the higher rate of antibodies in patients with lymph metastases corresponds to a better immune reactivation and host defence remains a matter of question,

Summarizing, the expression of heat shock proteins in NSCLC patients is of limited significance as a diagnostic approach, either alone or in a combination panel. The presence of αB-crystallin antibodies in plasma of NSCLC patients seems to be due the reactivation of the immune system and is unspecific as far as it is provoked under various stress conditions. The higher levels of the antibodies detected in patients with lymphogenic metastatic spread could be of clinical application as far as they could be used as markers for risk of recurrence

Although a lot of research is done, unraveling the signalling pathways and the intimate mechanisms of innate and adopted immunity, COPD remains a leading cause of morbidity and mortality worldwide. It seems that the current concepts of COPD are not leading to a solution that can be applied in clinical practice. It refers to both pathogenesis and treatment.

Oxidative stress does exist in the lungs and it is inevitable, having in mind its physiology. Keeping its balance is a kind of self-protection. COPD is obviously a disease of disbalance oxi/antioxidant, protease/antiprotease, apoptosis/proliferation, acetylases/deacetylases. In most cases COPD develops in smokers that makes us think that it necessarily is related to smoking, but why only 20% of smokers develop COPD? Moreover how can we explain the presence of non-smoker 's COPD? Is this another disease? If not can we say that it is a proteinopathy? A proteinopathy that sets a disbalance in lung cells even in the absence of exogenous noxa; a reason for sending a "danger" signal to the innate immunity that acts on default and instead of helping additionally stresses the already "stressed cells". Could it be that COPD in smokers is also proteinopathy, but acquired? There are not enough studies that let us make certain conclusions about the role of chaperones, particularly of heat shock

**7.1 "Stressing" the oxidative stress, inflammation and autoimmunity** 

proteins in COPD. There is only data that let us make hypotheses.

antioxidation (Aggeli et al, 2008).

since patients should be followed up.

and patients' prognosis.

**7. Future research in COPD** 

tumour stage (p = 0.042). Patients whose tumours had nuclear staining had shorter overall survival time in comparison to those that lacked staining (log-rank test p = 0.002). This supports the hypothesis that the nuclear positivity of the tumours refers to a more aggressive tumour biology. The nuclear positivity of αB in NSCLC stratifies patients from II and III stage in risk subgroups. Keeping in mind that more than 75% of patients are diagnosed at stage III, the introduction and validation of prognostic markers at this stage would undoubtedly help in predicting recurrence and improving clinical prognosis.

 To sum up the role of heat shock proteins in NSCLC we can say that the high molecular chaperones are important molecular mechanisms in lung cancerogenesis, probably contributing by their chaperoning abilities related to other oncogenic molecules - (HSP90,70) as well as by performing their role in apoptosis – (HSP60,70). They could be used in cancer treatment as their inhibition (HSP90) is associated with overwhelming of chemoresistance (Shimamura et al, 2005, 2008) – or their induction (HSP70) as a way of sensitizing tumors to chemo- and radiotherapy (Gehrmann, 2006) The small heat shock proteins are probably related to the regulation of apoptosis, cytoskeletal stability, chaperoning of antioxidant enzymes and prevention of oxidative stress. Their clinical significance is related to their applicaton as markers for chemoresistance - (HSP27), or risk stratification and survival (αBcrystallin).
