A Simple Radioassay to Detect Nanoscale Membrane Disruption

Understanding the mechanisms and kinetics of membrane damage is of interest to researchers in several overlapping fields of biology. In this study, we describe the development and validation of a simple 32PO43− release radioassay used to track nanometer-scale damage to the bacterial cell membrane. Nanoscale membrane damage will result in the release of small cytoplasmic molecules, such as amino acids, sugars, and osmolytes. Our radioassay tracks the release of these molecules using the release of cytoplasmic 32PO43− as a proxy. Our assay can both detect 32PO43− release and track release kinetics in the order of minutes. We demonstrate the use of our radioassay using A. baumannii treated with colistin and Ω76: two agents known to cause membrane damage. Our assay tracks greater membrane damage in A. baumannii treated with both these agents, compared to an untreated control. Our assay fills a niche that is not covered by traditional 51Cr release radioassays and fluorescent staining techniques. Furthermore, our assay can potentially be used to track membrane damage in other membrane systems such as lipid vesicles, animal cells, and organelles.


Introduction
The cell membrane is a prerequisite for life [1]. Life exists because the cell membrane concentrates biomolecules and separates them from the outside environment. Life ceases to exist when the cell membrane is irreparably damaged. Understanding the cell membrane and the mechanisms of its disruption is therefore a topic of interest across several fields, including immunology, apoptosis biology, cancer biology, and antibiotic discovery. Several methods for detecting membrane damage have been developed, but they can be classified into three major categories: radiometry, fluorometry, and microscopy.
The 51 Cr release radioassay is used to detect membrane damage via a simple procedure: target cells are labeled with 51 Cr; cytolysis of the target cells results in membrane damage, releasing 51 Cr; the degree of cytolysis can then be quantified by measuring the radioactivity of the medium [2]. 51 Cr release radioassays are considered to be the gold standard for measuring the cell-mediated cytotoxicity of T-cells and natural killer (NK) cells co-cultured with target cells [3]. 51 Cr assays have been used to detect cytotoxic reactions to rat Schwann cells [4], the recognition of influenza-infected cells by T-cells [5], and the phagocytic killing of Candida albicans [6].
However, the 51 Cr release radioassay has three major limitations. Firstly, it can only produce one reading at the end of the assay and therefore cannot measure kinetics [2,3]. Secondly, 51 Cr is a γ-emitter [7].Working with 51 Cr requires lead shielding and careful dose monitoring [8]. This has lead to the gradual phasing out of 51 Cr release radioassays in favor of newer fluorescence and bioluminescence techniques [9]. Thirdly, 51 CrO 2− 4 binds to bacterial lipopolysaccharides on the outer membrane of bacteria [10]. While useful for assaying cell death, 51 Cr cannot be used to assay damage to the inner membrane. The propidium iodide fluorescence assay has emerged as a popular alternative to the 51 Cr release radioassay [11][12][13]. The principle of fluorescence assays is simple: propiduim iodide present in the media will only enter dead cells via large perforations in the cell membrane. Once inside, the dye intercalates with DNA and fluoresces [14]. This dye will not enter and stain living cells. Intracellular propidum iodide can then be tracked using flow cytometry [15] or fluorescence microscopy [16]. The fluorescent dyes annexin-V [17] and 7-amino-actinomcin D [18] may be used as alternatives to propidium iodide.
In this work, we describe a simple radioassay to detect membrane disruption via the formation of nanometer-scale pores using 32 PO 3− 4 as a tracer. We had previously used this assay to characterize Ω76 [27], an antimicrobial peptide. We have now described the detailed protocol for use by the scientific community. Here, 32 PO 3− 4 is introduced into the bacterial cytoplasm via passive diffusion and is released upon the action of membrane disrupting agents. Unlike 51 Cr, 32 P is a β-emitter. Working with 32 PO 3− 4 only requires acrylic shielding. Further, our radioassay is capable of tracking 32 PO 3− 4 release kinetics in the order of minutes, and if required, seconds. We believe the assay described here will be of use to bacteriologists studying membrane disruption kinetics and can potentially be applied to any other membrane system as well.

Membrane-Disrupting Agents
This study used Colistin sulfate salt (Sigma C4461-100MG, lot no. SLBT0851, St. Louis, MO, USA) and Ω76 (synthesized by Genscript Inc., Hong Kong, China) to disrupt bacterial cell membranes; Ω76 may also be purchased from NovoPro Bioscience Inc. (catalogue number: 318759) or requested from the authors. You may test any known or putative membrane disrupting agent using this protocol. However, we recommend using Ω76 as a positive control.

Bacterial Culture
This study tested membrane disrupting agents against A. baumannii (P1270). This culture can be purchased from the Microbial Type Culture Collection (MTCC), Chandigarh (MTCC culture number: 12889). You may test a known membrane disrupting agent against any bacterial or eukaryotic cell culture.

Radiation Protection
An acrylic radiation shield, appropriate personal protective equipment (PPE), and a Geiger-Müller counter are needed while handling 32 PO 3− 4 . Store, handle, and discard radioisotopes as per your institutional guidelines. The Practical Radiation Technical Manual (IAEA) [28] provides detailed instructions on precautions needed while handling radioisotopes. In the event of a radiation spill, stop work immediately, notify personnel in the area of the spill, clean the spill with absorbent paper while wearing disposable gloves, dispose of your gloves and absorbent paper into the radioisotope waste container, survey yourself and the area to ensure that radiation levels have dropped to background levels, and inform your radiation safety officer (RSO) before resuming work.

Cold Room (4 • C)
All steps in this protocol need to be performed in a cold room to keep the cells being assayed metabolically inactive. Alternatively, an ice bath may be used for all steps following 32 PO 3− 4 uptake.

Centrifuge
The centrifuge must be capable of reaching speeds of at least 12,000 rpm and with rotors to accommodate 1.5 mL and 5 mL microcentrifuge tubes. Note that 5 mL microcentrifuge tubes may be substituted with 10 or 50 mL centifuge tubes if the appropriate rotor is unavailable.

Gel-Rocker
A gel-rocker is required for the gentle rocking of cells to aid the passive diffusion of 32 PO 3− 4 .

Aseptic Environment
A laminar flow hood or bunsen burner is required to create an aseptic environment while inoculating your culture. An aseptic environment is not required for further steps in this protocol.

Radiolabeled Phosphate Uptake
Note that 32 PO 3− 4 is very easily introduced into the bacterial cytoplasm via passive diffusion after incubation for 24 h. Care must be taken to incubate your culture at 4 • C to suspend bacterial metabolism and prevent the incorporation of phosphate into biomolecules.

1.
Inoculate your culture in 10 mL of Muller Hinton broth. Incubate at 37 • C/24 h, on a shaker incubator at 180 rpm.

2.
Pipette 2 mL of this culture into a suitable container (preferably a 5 mL microcentrifuge tube) and centrifuge at 10,000 rpm for 10 min. Collect the pellet and discard the supernatant.

4.
Add 100 µCi 32 PO 3− 4 to tube A1. CAUTION: Place an acrylic radiation shield between you and the radiation source whenever handling radioisotopes. Wear appropriate PPE.

5.
Incubate tube A1 on a gel rocker at 4 • C for 24 h. The 32 PO 3− 4 uptake occurs via passive diffusion across the cell membrane in metabolically inactive cells.
All the steps described above are illustrated in Figure 1.

5.
Incubate tube A1 on a gel rocker at 4 • C for 24 hours. 32 PO 3− 4 uptake occurs via 150 passive diffusion across the cell membrane in metabolically inactive cells.

Radiolabeled Phosphate Retention Check
After incubation, it is essential to verify that 32 PO 3− 4 entered, and is firmly retained within, the bacterial cytoplasm. This can be confirmed using a series of washing and pelleting steps.

1.
Pipette 500 µL of the incubated culture in tube A1 into an empty centrifuge tube (tube A2). The remaining culture in tube A1 can be refrigerated and used for further experiments.

3.
Resuspend P1 in 500 µL physiological saline. NOTE: Do not use phosphate-buffered saline at any step in this protocol. Unlabeled phosphate may compete with radiolabeled phosphate. 4-9. Repeat Steps 2-3 three more times. Over the course of this protocol, your pellet should be resuspended in physiological saline four times (P1→P4), resulting in four centrifuge tubes containing different supernatants at every step of the washing process (S1→S4). 10. Use a scintillation counter to enumerate the disintegration rates of tubes S1→S4 and P4.
• Disintegration rates are expected to fall approximately 10→100-fold from tubes S1→S3. This indicates that excess 32 PO 3− 4 is being washed out from the media. • Disintegration rates are expected to remain within the same order of magnitude between tubes S3 and S4. This indicates that all the excess 32 PO 3− 4 has been washed out. • Finally, the ratio of disintegration rates for P4:S4 is expected to be approximately 100:1. This ratio indicates the proportion of 32 PO 3− 4 firmly retained within the cytoplasm vs. the proportion of 32 PO 3− 4 released from the cytoplasm upon resuspension and centrifugation.
All the steps described above are illustrated in Figure 2. Table 1 contains experimental values for all the variables discussed in this section. After incubation, it is essential to verify that 32 PO 3− 4 entered, and is firmly retained 155 within, the bacterial cytoplasm. This can be confirmed using a series of washing and 156 pelleting steps.

1.
Pipette 500 µL of the incubated culture in tube A1 into an empty centrifuge tube 159 (tube A2). The remaining culture in tube A1 can be refrigerated and used for further 160 experiments.

175
• Disintegration rates are expected to fall approximately 10→100-fold from tubes 176 S1→S3. This indicates that excess 32 PO 3− 4 is being washed out from the media. 177 • Disintegration rates are expected to remain within the same order of magnitude 178 between tubes S3 and S4. This indicates that all the excess 32 PO 3− 4 has been 179 washed out.  All the steps described above are illustrated in Figure 2. Table 1 Figure 5. Three replicates per condition were performed, and the data for each replicate is provided in the columns rep-1 rep-3. All values are in disintegrations/min.

1.
Transfer 333 µL of the suspension from tube P4 to a 50 mL centrifuge tube containing 9.667 mL saline, bringing the total volume to 10 mL.

3.
Release 250 µL of the contents in the syringe into an empty microcentrifuge tube (Tube C). This tube serves as the pre-reaction total radiation check. The disintegration rate of this tube represents the total disintegration rate from 32 PO 3− 4 in both the cells and the saline medium.

4.
Carefully remove and discard the needle. Attach a 0.2 µm syringe filter to the syringe. Attach a new needle to the syringe filter. The filter will separate the saline filtrate from cells, allowing for the measurement of 32 PO 3− 4 released from the cells while ignoring 32 PO 3− 4 still present within the cells.

5.
Release 250 µL of the contents in the syringe into an empty microcentrifuge tube (tube T0). This tube's baseline disintegration rate indicates the amount of 32 PO 3− 4 present in the saline medium (the filtrate) before the addition of your membrane disrupting agent (at time = 0).

6.
Draw 250 µL of a pre-made stock solution of your membrane disrupting agent into the syringe. Note your stock solution will be diluted 40-fold within the syringe. Prepare your stock concentration accordingly. Replace your stock solution with saline for your negative control condition. Start timing your experiment from this point onwards. 7.
At predetermined timepoints, release 250 µL of the contents in the syringe into microcentrifuge tubes (tubes T1→Tn). 8.
Use a scintillation counter to enumerate the disintegration rates of tubes C, T0, T1→Tn. The percentage of 32 PO 3− 4 released at any timepoint (tube Tx) can be calculated using Equation (1).
All the steps described above are illustrated in Figure 3. Table 1 contains experimental values for all the variables discussed in this section.  Figure 3. The 32 PO 3− 4 release assay. All the steps required to determine whether the chosen membranedisrupting agent (agent X) disrupts membranes leading to the release of cytoplasmic small molecules, tracked using 32 PO 3− 4 .

1.
Transfer 333 µL of the suspension from tube P4 to a 50 mL centrifuge tube containing 189 9.667 mL saline, bringing the total volume to 10 mL.

4.1.
Rationale for the Development of the 32 PO 3− 4 Release Radioassay We had previously developed the 32 PO 3− 4 release radioassay to understand the nature and kinetics of membrane disruption caused by Ω76, an antimicrobial peptide, against the cell membranes of E. coli (K-12 MG1655) and A. baumannii (P1270) [27]. The motivation for developing this radioassay arose from MBC assays, time-kill curves, mouse models, scanning electron microscopic experiments, and fluorescent confocal microscopic experiments performed on these organisms.
We noted that Ω76 possessed an MBC 50 of 4 µg/mL against both E. coli and A. baumannii [27]; Ω76 is rapidly bactericidal, causing a ≥10 5 -fold reduction in A. baumannii c.f.u. counts over the course of 10 min [27]. Moreover, Ω76 displayed efficacy against A. baumannii in a mouse peritoneal model of infection, improving survival outcomes compared to controls [27]. Fluorescent, FITC-labeled Ω76 is incorporated into the cell membranes of both E. coli and A. baumannii ( Figure 4A). However, upon treating E. coli and A. baumannii with Ω76, only E. coli displayed large-scale membrane disruption and the release of cytoplasmic contents ( Figure 4B), while the cell membrane of A. baumannii appeared intact. . FITC-labeled Ω76 (green channel) was observed colocalizing with Nile red (red channel), which stains the cell membrane for both E. coli and A. baumannii. This was confirmed by Jaccard similarity coefficients of 0.64 and 0.74, respectively. FITC-labeled Ω76 (green channel) did not colocalize with DAPI (blue channel) that stains the bacterial chromosome for both E. coli and A. baumannii. This was confirmed by Jaccard similarity coefficients of 0.27 and 0.30, respectively. Therefore, Ω76 localizes within the cell membrane. Scale bar: 2 µm. The full method has previously been described in [27]. Note that the all images have been digitally magnified 3× after acquisition for clarity. (B) Scanning electron microscopy experiments for Ω76 (128 µg/mL) against E. coli (K-12 MG1655) and A. baumannii (P1270); Ω76 causes large-scale membrane disruptions and the release of cytoplasmic contents in E. coli. However, Ω76 causes no visible membrane disruptions on A. baumannii. Scale bar: 2 µm. The full method has previously been described in [27].
Since Ω76 possesses in vitro and in vivo efficacy against A. baumannii, and since Ω76 is incorporated into the bacterial cell membrane, we hypothesized that Ω76 may cause nanoscale membrane disruptions (possibly with toroidal pore or barrel-stave architectures [29]) that are too small to be visualized using scanning electron microscopy. The results of the 32 PO 3− 4 release radioassay described below validated this hypothesis ( Figure 5).

Expected
Results for the 32 PO 3− 4 Release Radioassay Nanoscale membrane disruptions are expected to cause the release of cytoplasmic small molecules into solution. The larger or more numerous the disruptions, the greater will be the release rate of these molecules. We had previously used 32 PO 3− 4 as a small molecule tracer to assay membrane disruption in A. baumannii under three conditions: untreated (negative control), colistin-treated (positive control), and Ω76 treated [19].

•
The untreated condition displayed the least phosphate release. Only 10% of 32 PO 3− 4 was released after 60 min ( Figure 5A). The rate of phosphate release remained fairly constant throughout this period, ranging from 0.06-0.3%/min. • Colistin interacts with lipopolysaccharides (LPS) on both the outer and inner membranes, leading to membrane disruptions and the release of cytoplasmic contents [30]. A total of 25% of 32 PO 3− 4 was released after 60 min ( Figure 5B). The rate of phosphate release peaked at 2.4%/min at t = 4 min.