High doses of laser phototherapy can increase proliferation in melanoma stromal connective tissue
- 42 Downloads
It is well established that laser phototherapy (LP) is contraindicated directly over cancer cells, due to its bio modulatory effects in cell and blood vessel proliferation. The aim of the present study was to analyze the influence of typical low-level laser therapy (LLLT) and high intensity laser therapy (HILT) and an in-between dose of 9 J on collagen fibers and blood vessels content in melanoma tumors (B16F10) implanted in mice. Melanoma tumor cells were injected in male Balb C mice which were distributed in four groups: control (no irradiated) or irradiated by 3, 9, or 21 J (150; 450, or 1050 J/cm2). LP was performed in daily sessions for 3 days with a InGaAlP—660 nm (mean output: 50 mW, spot size: 2 mm2). Tumor volume was analyzed using (1) picrosirius staining to quantify collagen fibers content and (2) Verhoeff’s method to quantify blood vessels content. Tumor growth outcome measured in the 3-J group was not significantly different from controls. Nine and 21-J groups, presented significant and dose-dependent increases in tumor volume. Quantitative analysis of the intensity of collagen fibers and their organization in stroma and peri-tumoral microenvironment showed significant differences between irradiated and control group. Blood vessels count of 21-J group outnumbered the other groups. High doses (≥ 9 J) of LP showed a dose-dependent tumor growth, different collagen fibers characteristics, and eventually blood vessel growth, while a typical LLLT dose (3 J) appeared harmless on melanoma cell activity.
KeywordsPhotobiomodulation LLLT Melanoma Collagen Bloob vessels Growing tumor
In the last few years, laser phototherapy (LP) has gained increasing popularity as a therapeutic modality. Initially, LP was given several names like “soft laser” or “cold laser,” but from 2001 low-level laser therapy (LLLT) was introduced as a MESH term for non-surgical LP. LLLT is applied transcutaneously with typical doses ranging from 1 to 6 J when targeting injuries or pathology in joints, tendons, and muscles. The proliferative potential of LP irradiation has attracted some negative attention, due to possible undesirable enhancement of tumor growth in neoplastic diseases. LP irradiation of cultured tumor cells has also generated some conflicting results, due to the great diversity in the types of cultivated tumor cell lines, and also differences in wavelengths and other irradiation parameters in published papers [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]. The most notable feature of the available LP studies in tumors is that there are few published studies in vivo [11, 12, 13]. Recently, more powerful lasers and higher irradiation doses have been advocated for LP and the new term high intensity laser therapy (HILT) has been introduced with typical doses ranging from around 10 J or more.
In vivo studies are essential to analyze the disease development, and they should be the main tool for studying tumor cells behavior. The complexity of the multicellular environment in an ongoing disease makes it hard to predict tumor behavior and because these isolated cell culture studies are inadequate to the assessment of tumor responses.
Tumor mass is frequently surrounded by a capsule and an invaded trabeculae (stroma) made of connective tissue (CT), rich in fibroblasts which are the main CT cells producing the extracellular matrix (ECM).
ECM is a complex of proteins and carbohydrates which provides important cues for surviving, differentiation, migration, and proliferation of surrounding cells . ECM may regulate cell proliferation at least by two different mechanisms: (1) acting as reservoir of growth factors and their availability to cell receptors or (2) through integrins, providing proper conditions to cell adhesion and signaling for cell growth. Integrins are able to induce cytoskeleton reorganization, activation of focal kinase (p125), and focal adhesion kinase (p125FAK). In addition, phosphatidyl-3 kinase regulator could be activated inducing changes in gene expression of cell cycle regulatory proteins [14, 15]. Some others ECM components such as fibronectin, laminin, vitronectin, and hyaluronic acid may induce cell proliferation as well [16, 17].
Tumor invasion and metastasis are multi-step processes which include tumor-host mediated by cell–cell and cell-matrix interactions. In order to invade surrounding tissue the tumor cells must overcome several local obstacles such as ECM, basal membranes, and blood vessels .
During the process of tumor invasion, ECM undergoes modifications, enhancing the collagen production: mainly types I and III and referred to as a desmoplastic reaction. In addition to desmoplastic reaction, both angiogenesis and immune system response are the main host reactions.
Cell proliferation increases in wound healing were observed after LP already in the pioneering work of Mester et al. in 1960s . Enhanced collagen production after LLLT was further confirmed and extended to other ECM components [20, 21, 22], as well as type I of collagen mRNA expression is enhanced by LP . Procollagen production stimulation by LLLT may occur at pre- or post-translational level and it is influenced by other factors present in serum containing medium . Extracellular collagen processing may be modified by LLLT through enhancement of metalloproteinase (MMP-2) activity and MMP-1 gene expression upregulation. In addition, other ECM components may be stimulated by LLLT such as, elastin , fibronectin , type III collagen, tenascin, and fibronectin .
Through this perspective, the aim of this study was to investigate the effect of a commonly used LLLT dose of 3 J and two higher doses of 9 J (in-between) and 21 J (HILT) respectively, in the stroma of melamona tumor (B16F10) injected in Balb C mice. For this purpose, we decided to use picrosirius staining to study collagen fibers characteristics and Verhoeff staining to quantify blood vessels.
Male isogenic male Balb C mice (n = 28), which were randomized into one of four experimental groups: 3 J dose group, 9 J dose, 21 J dose group, and a control group. The animals were provided by Central Animal House of Cruzeiro do Sul University and placed in a common box, housed under standard conditions of temperature (22–25 °C), relative humidity (40–60%), 12/12-h of light/dark cycle, and access to food and water ad libitum.
To induce tumor in experimental groups, we followed a similar method previously described by Kikuchi et al.  and Kuwano et al. : all animals received subcutaneous injection of a solution containing 2 × 106 murine melanoma B16F10 cells. After 15 days of tumor growth the irradiation was initiated according the parameters described below.
Laser device, doses, and treatment procedures
Experimental groups and irradiation parameters
Energy delivery (J)
Energy Density (J/cm2)
irradiation time (s)
Irradiation was started 15 days after the injection of the melanoma cells (period chosen to achieve significant growth characteristics of melanoma tumors). Irradiation was performed each 24 h for three consecutive days. During irradiation, mice were placed in a container where they stood still during the laser application.
In the day before the first irradiation, in all the days where irradiation was performed and on the day of sacrifice, animals were weighed and had their lesions measured with digital calipers, always by the same operator blinded to group allocation. One week after the last irradiation, mice were sacrificed by cervical dislocation for the removal of the lesion and subsequent histological analysis.
Tumor volume analysis
Tumor diameters were measured every day with a slide caliper (Digimess, 100,166 h model, São Paulo, Brazil) and tumor volume calculated using the following formula: volume (mm3) = width2 (mm2) × length (mm) × 0.52.
Histological tumor analysis
After tumor size measurements, animals were anesthetized with inhaled halothane (Sigma-Aldrich, MO, USA) and sacrificed by cervical dislocation. The tumor mass was immediately removed and immersed in a 4% phosphate buffered paraformaldehyde (Sigma-Aldrich, MO, USA) solution for 48 h. Specimens were dehydratated and embedded in paraffin prior the 5 μm microtome (Leica RM 2125 RT, Wetzlar, Germany) sections. For picrosirius (Sigma-Aldrich, MO, USA) staining, histological sections were initially immersed in xylene (Sigma-Aldrich, MO, USA) (2 × 20 min) and rehydrated in baths of alcohol (Synth, São Paulo, Brazil) in decreasing concentrations (100 to 70%) and finally washed in running water. Posteriorly, the sections were bathed for 15 to 30 min in sirius red solution (Sigma-Aldrich, MO, USA) (1%), washed in running water, dried, and counterstained with a solution of methylene blue (Sigma-Aldrich, MO, USA) (3%). After drying, were mounted with cover slips in mounting medium Enthellan® (Sigma-Aldrich, MO, USA). Verhoeff’s iron hematoxylin stain was achieved by 5% solution of hematoxylin (Sigma-Aldrich, MO, USA) in absolute alcohol (Synth, São Paulo, Brazil), and 10% aqueous solution of ferric chloride (Sigma-Aldrich, MO, USA) in Lugol’s iodine (Sigma-Aldrich, MO, USA) and, finally, after drying specimens were mounted with cover slips in mounting medium Enthellan® (Sigma-Aldrich, MO, USA). Photomicrographs were carried out in a Nikon-YS100 (Nikon, Tokyo, Japan) photomicroscope, and quantification was performed through the use of the Image J software (https://imagej.nih.gov). Obtained images were processed according method previously reported .
Obtained data was first plotted for analysis of normal distribution, and statistical analysis was then performed with parametric tests if the data were normally distributed. The statistical level of significance was set at p < 0.05. After confirmation, obtained data was tested statistically by ANOVA test. The mean value and its standard error (SE) were calculated, and differences between control group data and the irradiated group data were tested statistically with Tukey–Kramer test.
All data analyses showed normal distributions of data, and consequently these analyses were performed with parametric statistical tests.
Histologic tumor analysis
Blood vessel count analysis using Verhoeff staining method showed a reduced vessel count in the 3 and 9 J group. However, 21 J group has shown a significant increase of vessels count in comparison with the other irradiated groups (3 and 9 J) (Fig. 5).
Tumor volume analysis
In the present study, we analyzed the effects of low-level laser irradiation on stroma of malignant melanoma tumor injected in mice by collagen characteristics and vessels count to identify potential unwanted proliferative effect of low-level laser irradiation. Picrosirius staining is a specific colorimetric method to analyze collagen fiber length, thickness, and chemical affinity to the staining [29, 30].
Briefly, we observed that laser irradiation presented opposite effects when applied to each distinct situation. Considering the cultured melanoma cells, we surprisingly observed that laser irradiation presented a non-significant effect or even an inhibitory effect and cell death depending with the energy dose of the laser used. On the other hand, when we used the in vivo melanoma model in mice, we observed an unwanted significant increase in tumor mass volume and considerable histological alterations.
B16F10 melanoma cells are in a pigmented tumor lineage with a suppression of the CDK4/cyclin D inhibition due p16 mutation, permitting a free G1 → S cell cycle progression and less than 24 h duplication time . The high metabolism is important to consider once cell response to laser irradiation depends on cell metabolic state .
In this study, our findings are in line with our previous results . A commonly used LLLT dose of 3 J did not modify melanoma tumor volume significantly, whereas higher laser doses of 9 and 21 J (HILT) induced significant tumor volume growth. It is worth to mention that LLLT dose guidelines for pain, antiflammatory effect, and fibroblast stimulation do not recommend higher laser doses like 21 J (HILT). In fact, recommendations by World Association for Laser Therapy (REFERENCE www.walt.nu) are typically ranging up to 6 J (Chow et al. Lancet 2009 neck, Bjordal 2011 oral mucositis) when targeting joints, tendons, muscle, and mucosa. In this context, it can be assumed that the most commonly doses used in LLLT (1–6 J) are probably harmless even in the presence of an undiscovered in situ melanoma.
Tumors are complex entities containing many cell types in addition to the grossly abnormal tumor cells. In fact, the development of tumors relies on a cross-talk between tumor cells and the surrounding stroma which supports it. Tumor stroma provides the framework for tumor cells and is composed of connective tissue cells (fibroblasts, endothelial cells, pericytes, smooth muscle cells, and inflammatory white blood cells) and extracellular matrix (ECM) (collagen, proteoglycans and adhesive glycoproteins). As tumor progresses, tumor cells induce changes in stroma by secreting signal proteins that alter the behavior of stromal cells and their secretory products that can modify ECM. Stromal cells secrete proteases which further remodel ECM [32, 33, 34].
Our data suggests ECM altered the remodeling process, and more specially the collagen fibrils. In all irradiated groups, we observed a small amount of collagen content as compared to non-irradiated controls.
LLLT collagen and procollagen stimulation are well established in the literature [20, 21, 22, 23, 24, 25, 35]. However, this may be the first time that it is observed in a study employing a tumor model.
In addition to collagen content, our data suggests different chemical constitution from naïve collagen, as evidenced by higher staining affinity. Ignatieva et al.  detected significant and specific effects of the biochemical composition and material properties on the response of AF collagen network on laser irradiation.
In order to have a greater detailing of the ECM remodeling, we analyzed three tumor regions separately: (a) tumor core mass, (b) tumor capsule in the hypodermis, and (c) tumor capsule in the dermis.
In the tumor mass core, it was observed in 9 and 21 J (HILT) groups, lesser amounts of collagen and more intense staining of the fibrils. The same groups presented minor collagen content in hypodermic part of the capsule and these were less intense colored, less lengthy and less thickly. In the dermic part of the capsule, we observed in 21 J (HILT) group higher quantities of fibers, longer, thicker, and more intensely stained. However, In 9 J (in-between) group, we observed the opposite.
It is important to notice that collagen characteristics different from naïve occurred only in 9-(in-between) and 21-J (HILT) groups that also experienced tumor growth. In this perspective, we suggest that only 9 J (in-between) and 21 J (HILT) may alter collagen turnover in collagen stroma.
Collagen is the major ECM protein component produced by fibroblasts. It is secreted as procollagen molecule that is further processed by procollagen peptidases in order to form collagen fibrils. On the other hand, collagen degradation starts at extracellular space through metalloproteinase activity in order to recycle ECM. Turnover time depends on type of connective tissue considered .
Extracellular collagen degradation is achieved by matrix mineralloproteinase (MMP) enzyme family. MMP-1 (collagenase-1), MMP-8 (collagenase-2), and MMP-13 (collagenase-3) are responsible for initial collagen degradation and their genic expression in fibroblasts can be enhanced by inflammatory chemical modulators, like interleukin-1 (IL-1) and tumor necrosis factor-α (TNF-α). In previous studies, we observed that low-level laser therapy could directly modulate MMPs gene expression and enzyme levels . MMPs could be indirectly modulated by IL-1 and TNF-α, which also can be modulated by LLLT [38, 39].
Collagen biosynthesis/degradation in low-level laser irradiation tumor model is worthwhile to consider due its possible modulation of cell growth. Collagen structure may inhibit or stimulate cell proliferation. Fibrillar structure of collagen restricts cell proliferation, as described gallbladder smooth muscle cells; however, heat denatured collagen enhance five times cell proliferation. The mitogenic effect was mediated by protein kinase signaling pathway .
On the other hand, cancer progression induces overexpression of matrix metalloproteases (MMP) and related proteases which disrupts the balance between them and also their natural inhibitors. Extracellular matrix imbalance (protein degradation, including collagen) activates downstream molecular events which may lead to a permissive environment for tumor progression, invasion, and dissemination .
Vascular smooth muscle cells remain arrested in G1 phase in the presence of fibrillar collagen but proliferate when grown on nonfibrillar collagen. A negative growth regulatory control involves an increase in the levels of p21 and p27 . In addition, oxidized collagen stimulates vascular smooth muscle cells proliferation . M24 met human melanoma cells are inhibited to proliferate when contacting to fibrillar collagen . These findings may explain the tumor growth experienced by 9 J (in-between) and 21 J groups as they presented a small amount of collagen fibrils and more intense staining of it.
To their growth, tumors must recruit a suitable blood supply to ensure that it gets enough oxygen and nutrients. Thus, angiogenesis is required for tumor growth beyond certain size. This task is accomplished vascular endothelial growth factor (VEGF) .
Low-level laser therapy can stimulate microcirculation and angiogenesis in 3–8 J/cm2 power densities range [46, 47]. Our findings have shown that the 3 J (LLLT) group, reduced the blood vessel count, and the 9 J (in-between) group experienced a reduction in blood vessels count even greater than in the 3 J (LLLT) group. Only the 21 J (HILT) group showed a significant rise in blood vessels numbers, more in accordance with literature that points out laser enhancement of vasculature [46, 47].
In the current study, it has been shown that a commonly recommended LLLT dose of 3 J has negligent effects on melanoma tumor growth and cancer cell proliferation. Higher laser doses in the HILT range of 9 J and particularly 21 J has shown a dose-dependent increase in tumor growth and melanoma cell proliferation. All the doses showed different collagen fiber characteristics and eventually blood vessel growth.
Compliance with ethical standards
All experiments were carried out in accordance with the guidelines from Cruzeiro do Sul University Bioethical Council for human and animal care, PROTOCOL 011/07.
Conflict of interest
The authors declare that they have no conflict of interest.
Professor Ernesto Cesar Pinto Leal-Junior received research support from Multi Radiance Medical (Solon, OH), a laser device manufacturer.
All authors agree to the submission of this manuscript.
- 13.Frigo L, Luppi JSS, Favero GM et al (2009) The effect of low-level laser irradiation (In-Ga-Al-AsP—660 nm) on melanoma in vitro and in vivo. BMC Cancer 9, article 404Google Scholar
- 28.Rich L, Whittaker P (2005) Collagen and picrosirius red staining: a polarized light assessment of fibrillar hue and spatial distribution. Braz J Morphol Sci 22(2):97–104Google Scholar
- 29.Barbosa-Júnior AA (2001) Morphological computer-assisted quantitative estimation of stained fibrous tissue in liver sections: applications in diagnosis and experimental research. J Bras Patol 37(3):197–200Google Scholar