Exosomal Hsp70 in liquid biopsies-a biomarker for prediction and response monitoring in cancer

In contrast to normal cells, tumor cells of multiple entities overexpress the Heat Shock Protein 70 (Hsp70) not only in the cytosol, but also present it on their plasma membrane in a tumor-specific manner. Furthermore, membrane-Hsp70 positive tumor cells actively release Hsp70 into lipid microvesicles termed exosomes into the blood. Due to conformational changes of Hsp70 in the lipid environment, most commercially available antibodies fail to detect membrane-bound and exosomal Hsp70. To fill this gap and to assess the role of exosomal Hsp70 in the circulation as a potential tumor biomarker, we established the novel complete Hsp70 (compHsp70) sandwich ELISA using two monoclonal antibodies (mAbs) that are able to recognize both, free and lipid-associated Hsp70 on the cell surface of viable tumor cells and exosomes. The epitopes of the mAbs cmHsp70.1 (aa 451-461) and cmHsp70.2 (aa 614-623) that are conserved among different species reside in the substrate-binding domain of Hsp70, with measured affinities of 0.42 nM and 0.44 nM, respectively. Validation of the compHsp70 ELISA revealed a high intraand inter-assay precision, linearity in a concentration range of 1.56 to 25 ng/ml, high recovery rates of ‘spiked’ liposomal Hsp70 (>84%), comparable values between human serum and plasma samples, and no interference by food intake or age of the donors. Hsp70 concentrations in the circulation of patients with glioblastoma, squamous cell or adeno non-small cell lung carcinoma (NSCLC) at diagnosis were significantly higher than those of healthy volunteers. Hsp70 concentrations dropped concomitantly with the decrease in viable tumor mass on irradiation of patients with approximately 20 Gy (range 18 – 22.5 Gy) or after completion of radiotherapy (60 70 Gy). In summary, the compHsp70 ELISA presented herein provides a highly sensitive and reliable tool for measuring free and exosomal Hsp70 in liquid biopsies of tumor patients, levels of which can be used as a predictive tumor-specific biomarker, risk assessment and for monitoring therapeutic outcome.


Introduction
Lung cancer is the major cause of cancer-related deaths and the second most common cancer in men and women, worldwide [1]. Due to its nonspecific symptoms, lung cancer is frequently diagnosed at a late disease stage [2]. A relevant proportion of patients with locally advanced or metastasized tumors does not show an improvement in progression-free and overall survival following radical surgery, simultaneous chemo-and radiotherapy and immune checkpoint inhibitors [3,4]. Like NSCLC, Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 21 May 2021 doi:10.20944/preprints202105.0524.v1 glioblastoma multiforme (GBM) is a devastating disease of the central nervous system with symptoms that present at a very late disease stage. Despite multimodal treatment strategies consisting of surgery, radiotherapy and a temozolomide-based chemotherapy, overall survival remains poor at 15-18 months [5]. These examples underline the high medical need for tumor-specific biomarkers that improve the detection of tumors at an earlier stage and the monitoring of therapeutic responses. The development of such biomarkers will increase therapeutic success and the life expectancy of patients with highly aggressive tumors. Another challenge in clinical practice are the potential side effects of image-guided medical diagnosis. The availability of minimally invasive methods such as liquid biopsies for assessing the presence of tumor-specific biomarkers will have broad applicability and enable repetitive sampling with a good tolerability. Herein, we present an ELISA-based quantification of free and exosomal Hsp70 (HSPA1A) as a reliable approach for detecting tumors and monitoring therapeutic responses.
Members of the 70 kDa chaperone family support folding of nascent polypeptides, prevent protein aggregation and assist transport of proteins across membranes [6,7], and they reside in nearly all subcellular compartments of nucleated cells [8]. The importance of Hsp70 is documented by its high abundance, evolutionary conserved amino acid (aa) sequence [9,10] and functional similarities such as maintenance of protein homeostasis across different species [11]. Transgenic rodent models have revealed that Hsp70 of Drosophila melanogaster can substitute the activity of murine Hsp70 [12][13][14], and human Hsp70 expressed in myocardial cells of transgenic rats can protect the heart from ischemic stress, in vivo [15].
In contrast to normal cells, tumor cells frequently overexpress the major stress-inducible Hsp70 [16] in the cytosol and present it on the plasma membrane in a tumor-specific manner [17]. A global profiling of cell surface-bound proteins revealed a high abundancy of Hsp70 and other intracellular chaperones, like GRP78, GRP75, HSP60, HSP54, HSP27 on the plasma membrane of different tumor cells [18]. It is assumed that Hsp70 trafficking to the plasma membrane is enabled by an alternative non-ER/Golgi endo-lysosomal pathway [19]. Since major changes in extracellular salt concentrations and pH fail to deplete Hsp70 from the plasma membrane, a (trans-) membrane receptor mediated anchorage of Hsp70 is highly unlikely. Lipid profiling and artificial lipid copellation assays revealed that Hsp70 can directly interact with glycosphingolipids such as globoyltriaosylceramide (Gb3) which localize in cholesterol-rich microdomains (rafts) in the membrane of tumor cells. Since normal cells lack this tumorspecific lipid composition in their plasma membrane, Hsp70 resides strictly in the cytosol of normal cells. Furthermore, stress triggers an interaction of Hsp70 with the apoptosis-related membrane lipid component phosphatidylserine (PS) [19].
Tumor cells presenting Hsp70 on their cell membrane are more resistant to radiotherapy and chemotherapy compared to their membrane-Hsp70 negative counterparts [23]. After exposure to environmental stress, the synthesis and membrane expression of Hsp70 is further upregulated in tumor cells. A high Hsp70 content contributes to an aggressive tumor phenotype, mediates protection against apoptosis, promotes invasion/migration and mediates resistance to standard therapies [24]. Moreover, viable tumor cells expressing Hsp70 on their plasma membrane actively release exosomes, whereas 'free' Hsp70 generally originates from dying cells [25][26][27]. Since exosomes are created by a double invagination, the protein content in exosomal membranes reflects that of the tumor cell membrane from which they originate [28]. As a result, membrane-Hsp70 positive tumor cells release exosomes presenting Hsp70 on their exosomal surface [27], and the lumen of exosomes contains proteins of the tumor cytosol [29].
We have previously shown that serum Hsp70 levels in patients with tumors are higher than in patients with infectious diseases [30] or healthy volunteers [31]. A high Hsp70 serum content correlates with an increased malignancy and resistance to chemo-and radiotherapy [22,32,33].

Recombinant Hsp70
Human recombinant Hsp70 protein was produced in an optimized SF9 insect cell line (Orbigen,

Peptide SPOT synthesis and analysis
An array of consecutive 14-mer peptides with 12-residue overlap, covering the amino acid (aa) sequence 382-641 of human Hsp70, was synthesized according to the SPOT method, as previously described [38,39], on a Gly-PEG500-derivatised cellulose membrane using a MultiPep RS instrument (Intavis, Cologne, Germany). After N-terminal acetylation and de-protection of the peptide side chains,

Collection of human plasma and serum samples
Blood samples (7.5 ml each) were taken from healthy donors (n = 108), and patients with non-small cell lung cancer (NSCLC; n = 166) and glioblastoma multiforme (n = 34). Blood samples were also collected from patients with lung cancer at diagnosis (n = 80), during radiotherapy (after 20 Gy; n = 58) and after finishing radiotherapy (after 60 -70 Gy; n = 56). All study participants provided informed, written consent. Approval of the study was obtained by the local ethical committees of the Klinikum rechts der Isar, Technical University of Munich and the University Hospital Halle a.d. Saale. Plasma was prepared from EDTA blood (S-Monovette, Sarstedt, Nümbrecht, Germany) by centrifugation at 1500 g for 15 min at room temperature. Serum was obtained after clotting of the blood for 30 min at room temperature in a serum separator tube with clotting activator (S-Monovette, Sarstedt, Nümbrecht, Germany), followed by centrifugation at 750 g for 10 min. Serum and plasma were stored in aliquots (150 μl) at -80°C. To test the influence of food intake as a potential interference factor, serum samples were collected from healthy human individuals before and 2 h after intake of a high fat diet.

Results
Epitope mapping of the cmHsp70. 1  species are conservative and nonpolar [41,42]. The epitope similarities of the cmHsp70.1 and Hsp70.2 mAbs indicates that, in addition to humans, the compHsp70 ELISA system is likely to be capable of measuring free and exosomal Hsp70 in the blood of different mammalian species, including dog, bovine, horse and pig.     (Table 2).  Table 2).

Impact of interference factors on Hsp70 levels in the blood determined by the compHsp70 ELISA
To investigate the robustness of the data obtained with the compHsp70 ELISA using cmHsp70. Hsp70 concentrations before and after food intake were the same (Figure 5b). Since the age of the donors might have an impact on the Hsp70 concentrations in the circulation, plasma samples of 108 volunteers in different age groups ranging from 21-77 years (Table 3) were analyzed. As shown in Figure 5c, there was no significant correlation between plasma Hsp70 concentrations and the age of the donors, as determined using the Pearson correlation test (R 2 = 0.0781).   Table 3.

Comparative analysis of Hsp70 concentrations in the blood of cancer patients and healthy donors
In a first clinical evaluation, serum Hsp70 concentrations in patients with non-small cell lung carcinoma (NSCLC; n = 166) and high grade gliomas (HGG; n = 34; 26 primary, 8 relapse) were determined using the compHsp70 ELISA and compared to those in healthy volunteers (n = 108). The mean serum Hsp70 concentrations in patients with HGG (91.8 ± 21.3 ng/ml) and NSCLC (332.2 ± 37.9 ng/ml) were significantly higher than those in healthy volunteers (35.1 ± 4.0 ng/ml; Figure 6a). Receiver Operating Characteristic (ROC) curve analysis compared serum Hsp70 concentrations of healthy individuals with those of NSCLC and HGG patients (Figure 6b). The Area Under the Curve (AUC), the CI 95% value, the sensitivity and the specificity for a cut-off value was 114 ng/ml for NSCLC and 6 ng/ml for HGG patients, as determined by calculating the Youden-Index (Table 3). . For both tumor entities, Hsp70 concentrations were higher in patients with stage IV disease than those with stage III disease (*p<0.05). Furthermore, Hsp70 concentrations were higher in patients with adeno than squamous cell carcinoma histology, although these differences did not reach statistical significance (Figure 6c).
In addition to measuring circulating Hsp70 concentrations in tumor patients at diagnosis, Hsp70 levels were determined in responding patients with NSCLC before, during and after completion of radiotherapy. For this, blood samples were collected from patients before radiotherapy, during radiotherapy (after approximately 20 Gy; range 18 -22.5 Gy) and after completion of radiotherapy (60 -70 Gy). Hsp70 concentrations before radiotherapy (494.1 ± 72.2 ng/ml; n = 80), during radiotherapy (310.5 ± 36.8 ng/ml; n = 58) and after completion of radiotherapy (380.0 ± 51.8 ng/ml; n = 56) were significantly higher than those of healthy individuals (35.1 ± 3.99 ng/ml; n = 108) when measured using the compHsp70 ELISA (Figure 6d). After receiving a radiation dose of approximately 20 Gy (range 18 -22.5 Gy) Hsp70 levels dropped significantly from 494.1 ± 72.2 to 310.5 ± 36.8 ng/ml. After completion of radiotherapy (60 -70 Gy) the Hsp70 concentration was 380.0 ± 51.8 ng/ml.  is reasonable to assume that extracellular Hsp70 might serve as a universal tumor biomarker in a broad range of cancer entities.
As previously reported [30,57], blood Hsp70 levels correlate with the intracellular Hsp70 levels and match the membrane-Hsp70 status of the tumor cells from which they originate. In this study, we observed a significant decrease in the extracellular Hsp70 levels in patients with lung carcinoma during (after approximately 20 Gy) and after completion of radiotherapy (60 -70 Gy). In contrast, no significant drop in Hsp70 values was detected after radiotherapy when a commercial Hsp70 ELISA was used, and the values measured with the novel compHsp70 ELISA were more than 100-fold higher. The decrease in exosomal Hsp70 in the peripheral blood likely indicates a reduction in viable tumor mass in response to ionizing radiation. The minor, but not significant, increase of Hsp70 after completion of radiotherapy most likely attributes to an increased presence of free Hsp70 in the circulation derived from dying tumor cells and radiation-induced inflammation.
In line with our findings, a prospective clinical study including patients with solid tumors (breast and NSCLC) has also demonstrated exosomal Hsp70 levels to inversely correlate with therapeutic response [58]. In this study, a protocol that allows the isolation of exosomes from plasma samples of The prognostic value can be further increased by combining the data on exosomal Hsp70 with that of other biomarkers in liquid biopsies such as circulating DNA and/or mircoRNA. This hypothesis is in line with a report of Tomita et al. which has shown that the combined monitoring of CYFRA 21-1 and CEA, as relevant biomarkers in patients with NSCLC, improves prognostic relevance [61].
In conclusion, the novel compHsp70 ELISA presented herein provides a reliable and robust tool to quantify free and liposomal Hsp70 in the serum and plasma of cancer patients, levels of which reflect the presence and risk characteristics of tumors, their membrane-Hsp70 status, and therapeutic response.