**3. Morphological changes connected to cellular events**

#### **3.1. Cell death of adherent cells**

**1. Introduction**

320 Holographic Materials and Optical Systems

**2. Technique**

(www.phiab.se).

Digital holographic (DH) microscopy is a digital high-resolution holographic imaging technique with the capacity of quantification of cellular status without any staining orlabeling of cells [1–3]. Various cellular parameters can be visualized and calculated from the particular hologram, including individual cell area, thickness, volume, and population confluence and cell counts [4–10]. One of the advantages of studying cells with DH microscopy is that they can be grown and analyzed in their normal growth medium during the entire study. The culture vessel will be placed on the microscope for imaging and then replaced in the 37°C incubator, or placed on a heating plate to retain 37°C, during the analysis. Since the first studies on living cells, DH microscopy has been used to study a wide range of different cell types, e.g., protozoa, bacteria, and plant cells, mammalian cells such as nerve cells, stem cells, various

tumor cells, bacterial-cell interactions, red blood cells (RBC), and sperm cells [11–15].

DH microscopy is based on the interference between two, preferably coherent, beams that differ in phase (**Figure 1**). The beams usually originate from the same source, which are split before the sample. One of the beams, the reference beam, will remain undisturbed, while the other, the object beam, will be shifted in phase by the sample. The optical set-up can be either transmissive or reflective, providing no difference in the principle only in the configuration of the optical elements [16]. When the object beam has traveled through or been reflected by the object, the two beams merge. A light detector (e.g., a CCD-sensor) will capture the interference pattern and computer algorithms will convert the signal into a holographic image based on the light phase shifting properties of the cells, the refractive index [17]. The three-dimensional holographic image is then a representation of the real objects [18]. The technique is cell friendly, fast, and simple to use and has unique imaging capabilities for time-lapse investigations on

**Figure 1.** Schematic view of the DH microscopy technique. A digital holographic setup with a laser beam is split into two identical beams. The sample beam passes through the cells, while the reference beam travels undisturbed. The two beams merge and the image sensor will capture an interference image, which display a 3D-image after reconstruction Cell volume changes, resulting from cytotoxic treatments or apoptosis events, have recently been investigated with DH microscopy [11, 21–25]. When cells go in to early apoptosis, the first discernible indication is an increase in the cell phase shift followed by a decrease as the cell eventually dies [21]. Pavillon et al. recognized early apoptotic cells within minutes by their DH phase signal, while it took several hours to identify dead cells using trypan blue staining [21].

We have previously demonstrated that death-induced cells can be distinguished from untreated cells by the use of DH microscopy [25]. Morphological analyses of the two adherent cell lines L929 and DU145, treated with the anti-tumor agent etoposide for 1–3 days, were performed in cell culture flasks. Etoposide causes errors in the DNA synthesis and promotes apoptosis of the cancer cell by forming a ternary complex with DNA and the enzyme topoisomerase II [26]. Measurements revealed significant differences in the average cell number, the confluence, cell volume, and cell area when comparing etoposide-treated cells with untreated cells. The cell volume of the treated cell lines was initially increased at early time points. By time, cells decreased in volume, especially when treated with high doses of etoposide [25]. Moreover, this analysis was confirmed by a MTS assay. With DH microscopy, small differences between the two cell lines were clarified. Mouse fibroblast L929 cells showed a lower sensitivity for etoposide at the lowest concentrations, while for the human prostate cancer cell line DU145, the confluence, cell area, and volume increased at first, and then decreased over time.

In another study, we selected two suspension cell lines, a diffuse large B-cell lymphoma (DLBCL) cell line, U2932 and the T-cell acute lymphoblastic leukemia cell line Jurkat. The cell lines were treated with dimethyl sulfoxide (DMSO), 100 μM etoposide or left untreated as a control, and were incubated for 24 h. Unpublished work by us shows that the average cell area and cell volume decreased significantly for the Jurkat cell line compared with control. For the U2932 cell line, the average cell area did not change, whereas the average cell volume significantly decreased after etoposide treatment (**Figure 2**). Interestingly, the results may indicate cell line sensitivity, which also has been shown in the earlier studies [25, 27]. In conclusion, cell death experiments performed with DH microscopy reveal that small differ‐ ences between two cell lines can be clarified.

**Figure 2.** Etoposide induces a loss of cell area and volume in the cell lines Jurkat and U2932. Jurkat and U2932 cells were treated with etoposide (100 μM), or left untreated, for 24 h, and holograms were captured by the Holomonitor™ M4. The mean cell area (A) and the mean cell volume (B) were calculated using holographic microscopy images. Error bars are based on the total number of images.

Human α‐lactalbumin made lethal to tumor cells (HAMLET) is based on a natural protein present in human breast milk [28]. HAMLET induces cell death in tumor cells and immature cells, but not in normal differentiated cells [29]. As monitored by DH, 15 min of incubation, with 35 μM of HAMLET, was enough for a reduction in cell area and increase in thickness, with evidence of membrane blebbing in human lung carcinoma A549 cells [26]. After 60 min, the cells became even smaller in area and thicker [30]. Puthia et al. examined how HAMLET affects β‐catenin and Wnt‐signaling in the treated human colon cancer cell line DLD1. Already after 30 min, the cell morphology changed with HAMLET treatment. A time‐dependent decrease in cell area and an increase in maximum thickness was seen [31]. DH microscopy was also used to analyze the effect of the cell death‐inducing curcumin analog C‐150 [32]. Four different glioblastoma cell lines were treated with 1 μM of the analog for 24 h. The results showed significantly increased cell volume and average thickness and decreased cell area for the cell lines investigated.

DH microscopy has been applied for the analysis of chemokinetic responses, selective cytotoxic, adhesion, and migration modulator effects on two different melanoma cell lines, HT168‐M1 and A2058, after treatment with di‐ and trihydroxyanthraquinones [33]. Alizarin and purpurin have been reported to have activity against cancer cells. Their results showed that no basic parameter was influenced by alizarin or purpurin as a cytotoxic or apoptotic substance in HT168‐M1 cells. In the case of A2058 cells, alizarin could induce positive effects in the average cell area and volume as measured by DH.

#### **3.2. Cell cycle of adherent cells**

Our earlier results on etoposide, colcemid, or staurosporine treated cells showed changes in average cell volume [34]. Mouse fibroblasts were treated and analyzed with DH micro‐ scopy after 24 h of incubation. The results showed comparable accuracy to flow cytometry measurement of cell cycle distribution, where staurosporine induced G1 arrest and colcemid or etoposide induced G2/M arrest. The results with DH microscopy showed that the cells decreased in cell size in response to staurosporine treatment, while the cell size increased in response to colcemid or etoposide treatment. Etoposide was further used in a dose-dependent manner in order to investigate how well DH microscopy was able to record a change in the cell cycle profile, as compared to flow cytometry. Etoposide reduced average cell number, decreased average cell confluence, and increased average cell volume. Indeed, using immortalized murine fibroblast cells, this first proof-of-concept study suggests that DH microscopy is a possible alternative tool for analysis of cell cycle alterations [34]. In another study, treated SKOV3-TR cells were visualized for 24 h until cell cycle arrest and characterized by the presence of rounded cells that were unable to complete mitosis [35]. After 44 h, the cells had undergone apoptosis. Higher concentrations of treatment

indicate cell line sensitivity, which also has been shown in the earlier studies [25, 27]. In conclusion, cell death experiments performed with DH microscopy reveal that small differ‐

**Figure 2.** Etoposide induces a loss of cell area and volume in the cell lines Jurkat and U2932. Jurkat and U2932 cells were treated with etoposide (100 μM), or left untreated, for 24 h, and holograms were captured by the Holomonitor™ M4. The mean cell area (A) and the mean cell volume (B) were calculated using holographic microscopy images. Error

Human α‐lactalbumin made lethal to tumor cells (HAMLET) is based on a natural protein present in human breast milk [28]. HAMLET induces cell death in tumor cells and immature cells, but not in normal differentiated cells [29]. As monitored by DH, 15 min of incubation, with 35 μM of HAMLET, was enough for a reduction in cell area and increase in thickness, with evidence of membrane blebbing in human lung carcinoma A549 cells [26]. After 60 min, the cells became even smaller in area and thicker [30]. Puthia et al. examined how HAMLET affects β‐catenin and Wnt‐signaling in the treated human colon cancer cell line DLD1. Already after 30 min, the cell morphology changed with HAMLET treatment. A time‐dependent decrease in cell area and an increase in maximum thickness was seen [31]. DH microscopy was also used to analyze the effect of the cell death‐inducing curcumin analog C‐150 [32]. Four different glioblastoma cell lines were treated with 1 μM of the analog for 24 h. The results showed significantly increased cell volume and average thickness and decreased cell area for

DH microscopy has been applied for the analysis of chemokinetic responses, selective cytotoxic, adhesion, and migration modulator effects on two different melanoma cell lines, HT168‐M1 and A2058, after treatment with di‐ and trihydroxyanthraquinones [33]. Alizarin and purpurin have been reported to have activity against cancer cells. Their results showed that no basic parameter was influenced by alizarin or purpurin as a cytotoxic or apoptotic substance in HT168‐M1 cells. In the case of A2058 cells, alizarin could induce positive effects

Our earlier results on etoposide, colcemid, or staurosporine treated cells showed changes in average cell volume [34]. Mouse fibroblasts were treated and analyzed with DH micro‐ scopy after 24 h of incubation. The results showed comparable accuracy to flow cytometry

ences between two cell lines can be clarified.

322 Holographic Materials and Optical Systems

bars are based on the total number of images.

the cell lines investigated.

**3.2. Cell cycle of adherent cells**

in the average cell area and volume as measured by DH.

**Figure 3.** The drug PLX4032 influences cell area and volume differently in the human melanoma cell lines WM-266-4 and CHL-1. WM-266-4 and CHL-1 cells were treated with PLX4032 (100 nM and 1 μM), or left untreated, and holograms were captured by Holomonitor™ M4 after 24 and 48 h, respectively. The mean cell area (A) and the mean cell volume (B) were calculated using holographic microscopy images. Error bars are based on the total number of images.

showed early cell cycle arrest, and the cell population started to undergo apoptosis after 14 h. Indeed, by using time-lapse DH imaging, cell cycle arrest followed by progression to apoptosis was clearly visualized.

**Figure 4.** 3D-images of WM-266-4 cells. The human skin melanoma cell line WM-266-4 treated with PLX4032 (100 nM and 1 μM), or left untreated. Holograms of the cells were captured using holographic microscopy after 24 and 48 h, respectively. Hologram pictures show the morphology changes over time with the different concentrations of PLX4032.

**Figure 5.** 3D-images of CHL-1 cells. 3D holograms showing the human skin melanoma cell line CHL-1 treated with PLX4032 (100 nM and 1 μM), or left untreated. Holograms of the cells were captured using holographic microscopy after 24 and 48 h, respectively. Hologram pictures show the morphology changes over time with the different concentrations of PLX4032.

We have treated two human melanoma cell lines, WM-266-4 and CHL-1, with different sensitivity for the drug PLX4032, also called vemurafenib [36]. PLX4032 induces cell cycle arrest in lower doses and apoptosis in higher doses in melanoma cells with a certain *BRAF* gene mutation. It has previously been shown that CHL-1 cells are not inhibited by the drug due to lack of this mutation. In our study, PLX4032 treatment increased the average cell area and cell volume for the WM-266-4 cell line compared with untreated control cells, while the CHL-1 cell line was less affected (**Figure 3**). Morphological cell changes are presented in 3D holograms for WM-266-4 cells (**Figure 4**) and CHL-1 cells (**Figure 5**). The cells become more flat, with increased area, after PLX4032 treatment. Indeed, with DH microscopy, even small differences between treated cells were clarified.

The effect of the novel anti-cancer drug ISA-2011B, etoposide, and docetaxel was monitored with DH microscopy in real-time for up to 48 h by Semenas et al. ISA-2011B treatment led to a reduction in cell size and changes in morphology, which was also achieved by docetaxel treatment [37].

#### **3.3. Cell death of suspension cells on antibody-based microarray**

showed early cell cycle arrest, and the cell population started to undergo apoptosis after 14 h. Indeed, by using time-lapse DH imaging, cell cycle arrest followed by progression to

**Figure 4.** 3D-images of WM-266-4 cells. The human skin melanoma cell line WM-266-4 treated with PLX4032 (100 nM and 1 μM), or left untreated. Holograms of the cells were captured using holographic microscopy after 24 and 48 h, respectively. Hologram pictures show the morphology changes over time with the different concentrations of PLX4032.

**Figure 5.** 3D-images of CHL-1 cells. 3D holograms showing the human skin melanoma cell line CHL-1 treated with PLX4032 (100 nM and 1 μM), or left untreated. Holograms of the cells were captured using holographic microscopy after 24 and 48 h, respectively. Hologram pictures show the morphology changes over time with the different concen-

We have treated two human melanoma cell lines, WM-266-4 and CHL-1, with different sensitivity for the drug PLX4032, also called vemurafenib [36]. PLX4032 induces cell cycle arrest in lower doses and apoptosis in higher doses in melanoma cells with a certain *BRAF* gene mutation. It has previously been shown that CHL-1 cells are not inhibited by the drug due to lack of this mutation. In our study, PLX4032 treatment increased the average cell area and cell volume for the WM-266-4 cell line compared with untreated control cells, while the CHL-1 cell line was less affected (**Figure 3**). Morphological cell changes are presented in 3D holograms for WM-266-4 cells (**Figure 4**) and CHL-1 cells (**Figure 5**). The cells become more flat, with increased area, after PLX4032 treatment. Indeed, with DH microscopy, even small

apoptosis was clearly visualized.

324 Holographic Materials and Optical Systems

trations of PLX4032.

differences between treated cells were clarified.

We have introduced antibody-based microarrays [38] to the experimental DH microscopy setup. By using single-chain variable antibody fragments (scFv) [39], directed against some of the most common cell membrane proteins on T- and B-lymphocytes, suspension cells can be analyzed with DH. Antibody-based microarray techniques have been used to determine phenotypic protein expression profiles for human B cell sub-populations [39] and to detect soluble antigens [40]. In our study [27], we combined DH microscopy and antibody-based microarray to introduce a powerful tool to measure morphological changes in specifically etoposide-treated antibody-captured cells, U2932, and Jurkat (**Figure 6**). We demonstrated that the cell number, mean area, thickness, and volume could be noninvasively measured by using DH microscopy. The cell number was stable over time, but the two cell lines used showed changes of cell area and cell irregularity after treatment. The cell volume in etoposide-treated cells was decreased, whereas untreated cells showed stable volume [27]. In conclusion, cell death of suspension cells investigated with the help of antibody-based microarrays and DH demonstrated that morphological parameters can be investigated of different cell lines and treatments.

**Figure 6.** Jurkat cells on an antibody-based microarray captured with DH microscopy. A 200,000 Jurkat cells in 100 μl of phosphate-buffered saline (PBS)-0.5% bovine serum albumin were applied to an antibody array and incubated at room temperature for 30 min. The array was thereafter washed manually, until the cell binding areas were clearly visible. A 3D-image was thereafter captured with DH microscopy.

In conclusion, several independent studies now show the feasibility of DH microscopy to demonstrate cell-death induced morphological changes after compound addition.
