Methods to Evaluate the Potential Clinical Significance of Antibodies to Red Blood Cells

Immune‐mediated red blood cell (RBC) destruction due to antibodies is an ongoing problem in transfusion medicine for the selection of the safest blood. Serological testing often revealed incompatibility with donors’ RBCs. When this incompatible blood was transfused, destruction was due mostly to extravascular‐mediated phagocytosis of the antibody‐opsonized RBCs; however, intravascular hemolysis was sometimes observed without explanation. Based on serology, antibodies with potential for clinical sequalae could not be ascertained; thus, antigen‐negative blood was usually selected for transfusion to avoid problems. Antibodies to antigens having very high frequency in the general population (>95%), however, made selection of antigen‐negative blood difficult and sometimes impossible. Some patients, who were sensitized by previous transfusions or by pregnancy, developed multiple antibodies, again creating a problem for finding compatible blood for transfusion, without the ability to discern which of the antibodies may be clinically irrelevant and ignored. Transfusion medicine scientists began searching for an in vitro means to determine the in vivo outcome of transfusion of blood that was serologically incompatible. Methods such as chemiluminescence, monocyte‐macrophage phagocytosis, and antibody‐dependent cellular cytotoxicity (ADCC) were described. Over the years, the monocyte monolayer assay (MMA) has emerged as the most reliable in vitro assay for the prediction of the clinical relevance of a given antibody. ADCC has not been fully studied but has the potential to be useful for predicting which antibodies may result in intravascular hemolysis. This article captures the protocols for the implementation and readout of the MMA and ADCC assays for use in predicting the clinical significance of antibodies in a transfusion setting. © 2022 The Authors. Current Protocols published by Wiley Periodicals LLC.


INTRODUCTION
Since the discovery of the ABO blood group system by Landsteiner (1901), the clinical discipline of transfusion medicine has grown with leaps and bounds. Indeed, the four blood group antigens, A, B, AB, and O, described by Landsteiner in the early 1900s became the first human blood group system (ABO). Since then, additional blood group systems have been described so that there are now >250 blood group antigens and 25 blood group systems (Garratty et al., 2000). Despite the serologic and genetic methods developed to identify blood group antigens and their corresponding antibodies, the question of which antibodies are important from a clinical standpoint has never been completely answered.

MONOCYTE MONOLAYER ASSAY (MMA)
MMA is an in vitro assay which is used to predict blood transfusion outcomes in patients with auto-or alloantibodies to RBCs. In this assay, anti-RBC antibodies are assessed for their Fcγ receptor (FcγR)-mediated phagocytosis. Through serological methods, compatibility testing or crossmatching is performed. Once the presence of antibodies to RBCs is detected, MMA can be further used to identify the clinical significance of the anti-RBC antibody by testing them against specific RBC antigens (e.g., opsonizing Kell positive RBCs with anti-Kell antibody). The phagocytosis results from the MMA can help reduce the risk of post-transfusion hemolysis. Cells of interest in the MMA are the peripheral blood mononuclear cells (PBMCs) due to the mononuclear phagocyte system's involvement in mediating the extravascular hemolysis of antibody-bound RBCs. Apart from predicting post-transfusion survival or clearance of RBCs, MMA can be used to study other aspects of the IgG antibody and FcγR interaction that induce phagocytosis.
Fresh whole blood collected in acid-citrate-dextrose (ACD) tubes (yellow-top tubes; minimum of two blood tubes should be collected for MMA) Fresh whole blood collected by venipuncture into red-top (no additive) serum separator tubes Rh positive (R 2 R 2 ) red blood cells (

Isolate PBMCs
1. Obtain human whole blood from donor or patient in ACD tubes and red-topped serum tubes. Store whole blood in ACD tubes at room temperature (18°C to 22°C) and red-topped serum tubes at 4°C to allow separation of serum.
It is recommended, for optimal function, that the PBMCs from whole blood ACD tubes be isolated and used within 36 hr of collection.
2. Transfer room temperature blood from whole blood ACD tubes to 50-ml Falcon tubes (approximately one 50-ml Falcon tube for every two ACD tubes). Add equal volume of the room temperature RPMI-1640 complete medium to the blood (1:1 ratio of whole blood to medium), for a final volume of 35 ml.
3. Add 15 ml room temperature Ficoll Paque Plus to a new 50-ml Falcon tube.
4. Carefully layer the 1:1 diluted whole blood on top of the Ficoll Paque Plus density gradient to minimize amount of mixing at the interface for optimal separation of blood.
The 50-ml Falcon tube can be tilted at a 45°angle and blood can be added dropwise by placing pipet tip close to the density gradient and enabling blood mixture to run down the side of the tube very slowly. Typically, 10 ml of whole blood yields 10 million PBMCs with some donor-to-donor variation.
5. Centrifuge the layered mixture at 700 × g for 30 min with brakes OFF.
The density gradient centrifugation will separate the mixture into (from top to bottom; Fig. 1): General experimental flow diagram of the monocyte monolayer assay (MMA). PBMCs isolated from whole human blood are seeded onto chamber slides to allow monocyte adherence for 1 hr. Opsonized RBCs are then added to the monocytes for the phagocytosis to occur. After 2 hr incubation, the chamber slides are washed, and the phagocytic index is determined by counting phagocytosis events under the microscope. Results are considered significant when the phagocytic index exceeds 5 phagocytosed RBCs in 100 monocytes. Abbreviations: PBMCs, peripheral blood mononuclear cells; RBCs, red blood cells.
6. Aspirate and discard the majority of the topmost layer (supernatant) which is plasma, leaving 1-2 ml remaining above the buffy coat layer, and carefully retrieve the buffy coat (PBMCs) layer. (Using plastic or glass Pasteur pipets with a suction bulb while performing circular motions is recommended.) Transfer the retrieved PBMCs into a new 15-ml tube.
7. Wash the isolated buffy coat layer two times with pH 7.4 PBS solution for 10 min at 350 × g with full brakes ON in between washes.
8. Lyse any RBCs carried over with ACK lysis buffer. Add 5-10 ml ACK lysis buffer, depending on pellet size, and incubate at room temperature for 5 min. After incubation, top up with pH 7.4 PBS and centrifuge 10 min at 350 × g with full brakes ON and wash one more time.
9. Reconstitute PBMC pellet in 3-7 ml (depending on the size of the pellet) RPMI-1640 complete medium.
10. Count PBMCs using a hemocytometer in a 1:1 staining ratio with trypan blue by only counting the cells not stained with trypan blue. Reconstitute PBMCs to 1,750,000 cells/ml in RPMI-1640 complete medium.
11. Seed 400 μl (700,000 cells) using a micropipet into each well of the 8-well chamber slide and incubate at 37°C, 5% CO 2 for 1 hr in a humidified tissue culture incubator.

Pre-treatment of adhered monocytes
Steps 12-13 are only necessary if trying to inhibit or enhance phagocytosis.
12. Adhered monocytes can be pre-treated with any drug(s) or compound(s) of interest. Reconstitute the test material in RPMI-1640 complete medium to the desired concentration.
For example, 200 μg/ml of IVIg can be used to achieve a 95% to 100% of inhibition of phagocytosis when using human monocytes.

of 13
Current Protocols 13. Aspirate any non-adhered PBMCs from the wells using a micropipet and discard. Add 400 μl of test material and incubate at 37°C, 5% CO 2 for 1 hr.

Opsonization of Rh(D)+ R 2 R 2 red blood cells
R 2 R 2 RBCs are used as a positive control for FcγR-mediated phagocytosis. R 2 R 2 RBCs can be centrifuged at 870 × g for 15 min with brakes ON and reconstituted in ID-CellStab (RBC storage/stabilization solution) and stored at 4°C for up to a month.
Although we use D+ R 2 R 2 RBCs for opsonization with anti-D for our positive control, any Rh+ RBCs could be used and do not have to be phenotyped.

of 13
Current Protocols 22. Submerge slide into the beaker with the discarded supernatants and wash slide by moving it back and forth slowly (20-30 strokes) to remove any remaining nonphagocytosed RBCs.
23. Using the other beaker without any discarded supernatants, submerge slide and wash slowly for 20-30 strokes more.
24. Remove slide from PBS and dab off excess on paper towel. Air dry slide.
25. When almost dry, fix by submerging in 100% methanol for 45 s then air dry.
26. Mount slide using an in-house made Elvanol mounting medium and add coverslips (24 × 75 mm).
27. Allow mount to dry overnight before quantification.
Quantification of phagocytosis 28. Using a phase contrast microscope and 40× objective lens, quantify phagocytosis using a manual cell counter.
Although we use phase-contrast microscopy to count the RBCs phagocytosed in 100 monocytes, hematological stains such as Wright-Giemsa can also be used (Fig. 1).
Have two manual cell counters in each hand to count phagocytosed RBCs in one and total number of monocytes in the other (300 monocytes should be counted per well).
29. Calculate average phagocytic index (PI) per test (across triplicates) by dividing the number of phagocytosed RBCs by the number of total monocytes counted and multiplying by 100: (Number of phagocytosed RBCs/Number of total monocytes counted) × 100. Express data as average PI ± the standard error of the mean (SEM).

Interpretation of results
We and others have found that a PI >5 is correlated with clinically significant antibodies. The MMA may not correlate with serologic results as shown in Figure 1. In this result, the antibody (anti-Cartwright (Yt a )) reacts by IAT serology with the Yt(a+) RBCs but not with Yt(a-) RBCs. This would suggest that this antibody is clinically significant and only Yt(a-) blood should be selected for transfusion. Despite the IAT reactivity, the MMA assay indicates that the PI is <5 and the antibody is, thus, considered clinically insignificant and that all these serologically incompatible donor bloods can be transfused into this patient without sequalae. Indeed, this patient was transfused with Yt(a+) blood and did not have any problems.

ANTIBODY-DEPENDENT CELLULAR CYTOTOXICITY ASSAY (ADCC)
Natural killer (NK) cells are classical mediators of ADCC through the interaction of their low affinity Fcγ receptor CD16 with IgG antibodies present in circulation. Some of these antibodies (alloantibodies) can lead to unwanted reactions in the case of patients receiving a transfusion thus matching between recipient and blood donor is required. Therefore, as a complementary method to determine the clinical significance of antibodies in transfusion reactions, and test for compatibility between donor and recipient, we evaluate the capacity of NK cells to mediate ADCC against red blood cells, when the latter have been exposed to a specific human serum.

of 13
Current Protocols Fresh whole blood collected into acid-citrate-dextrose (ACD) tubes (yellow-top tubes; minimum of nine to ten blood tubes should be collected for NK cell-mediated ADCC assays; store blood at room temperature up to 48 hr) Fresh whole blood collected by venipuncture into red-top (no additive) serum separator tubes Isolation buffer (see recipe) 1× PBS, pH 7.4, without Ca 2+ /Mg 2+ (Wisent Bioproducts, 311-425-CL) Patient's isolated serum (for opsonization), or utilize patient's plasma or a purified antibody of your interest 51 Cr (Na 2 CrO 4 , sodium chromate), 1 mCi (

Optimization of number of RBCs to use in the ADCC assay
Optimization steps should be done just once at the beginning of the study.
1. Wash RBCs with PBS three times by centrifugating at 350 × g for 5 min at room temperature each time.
8. After 4 hr, take plate out of the incubator and carefully collect 50 μl cell-free supernatant from each well (with a multichannel pipet), and transfer to LumaPlates. 9. Let the LumaPlates dry overnight.
Other radioactivity counters may be used.
11. Calculate the ratio of dead versus alive cells [divide the counts per minute (cpm) of the dead cells by the cpm of the alive cells] and plot in an xy graph against the corresponding cell number (Fig. 2A). Select the most adequate number of cells for your subsequent experiments.
Considering the effector cells' availability, the number of RBCs to use should be determined individually for the specific assay. In this case, we used the lowest number of RBCs that yielded the highest radioactivity readouts (Fig. 2A).
Purify NK cells 12. Perform PBMC isolation as described in monocyte monolayer assay (Basic Protocol 1).
13. Count isolated PBMCs and adjust to 50 × 10 6 cells/ml in isolation buffer to proceed with the NK purification.
14. Purify NK cells using the NK isolation kit or the NK enrichment kit (Stemcell Technologies; EasySep TM ) according to the manufacturer's instructions (Fig. 3).
If NK cell isolation kits are not available, the assay can be done using the whole PBMC fraction. However, the level of cytotoxicity obtained when using PBMC is significantly lower than the cytotoxicity observed with purified NK cells (Fig. 2B).
15. Count purified NK cells (ready to use) and resuspend in complete RPMI at the desired concentration.

For the quantification and correct interpretation of results, different effector/target ratios of cells should be used. Serial dilution of effector cells should be made in triplicate on the 96-well round-bottom microtiter plate of the assay.
Target cell labeling 16. Patient RBCs are washed and opsonized as described in Basic Protocol 1, steps 14-17.
17. Count RBCs and add 50 μCi of 51 Cr per 1 × 10 7 cells. Incubate for 1 hr in a 37°C incubator, 5% CO 2 . Resuspend every 15 min by tapping the tube on the side.
19. Resuspend washed RBCs in complete RPMI and adjust cell concentration to 30,000 cells in 100 μl.
Considering the effector cells availability, the number of RBCs to be use should be determined individually for the specific assay. In this case, we used the lowest number of RBCs that yielded the highest radioactivity readouts (Figure 2A). 26. Count (1 min/sample) in a radioactivity counter (e.g., MicroBeta 2 microplate counter).
Other radioactivity counters may be used.
27. Determination of specific 51 Cr release assay and calculation: The percent of specific 51 Cr release (equivalent to specific lysis) is calculated as: (Experimental valuespontaneous release)/(Maximum release-spontaneous release) × 100. Each value is calculated as the mean of triplicates.

Background Information
Evaluation of the clinical significance or insignificance of antibodies to RBC antigens when deciding on the selection of blood in anemic patients requiring RBC transfusion support has historically been based on the specificity of the antibodies using serological methods. In complicated transfusion cases, such as when patients have alloantibodies against high-prevalence antigens of uncertain clinical significance or multiple alloantibodies whereby it is difficult to find crossmatch compatible blood, the in vitro method most tested and proven to be predictive of in vivo antibody clinical significance is the monocyte monolayer assay (MMA; Hadley, 1998;Noums, Billingsley, & Moulds, 2015;Tong, Cen, & Branch, 2019;Zupanska, 1993). However, cell-mediated cytotoxicity-the direct lysis of RBCs-may be an important mechanism of antibody-dependent (ADCC) or antibodyindependent RBC lysis (Flegel, 2015;Garratty, 2008). ADCC, although proposed to be a mechanism of RBC lysis (Barcellini, 2015), has not been as well studied as other methods; but due to cases of brisk intravascular lysis seen in some instances (Michelis et al., 2014), ADCC would be a method of RBC lysis that should be further evaluated (Garratty, 2008). We have described the MMA and ADCC assays used in our laboratory in great detail so that they can be used to predict the potential clinical significance of RBC auto-and alloantibodies.

Critical Parameters
For NK assays and MMA, buffy coat can be used.
Best results are obtained with a control antibody having a known concentration. Although any anti-D could be used, we use WinRho® SDF CDN (Saol Therapeutics, 1003092) Rh immune globulin as its concentration of anti-D is known. Any Rh immune globulin could be used for the positive control to ensure reproducibility.

Troubleshooting
See Tables 1 and 2 for common problems encountered when performing these protocols and suggested solutions.

Understanding Results
As a general rule, ≥5% killing is considered significant.

Time Considerations
The described assays are time consuming and it is important to have a dedicated person well trained to perform these assays. The MMA assay takes ∼4 to 5 hr to get to a stopping point, when slides are finalized for reading using phase-contrast microscopy or looking at hematological stained slides. The reading of the phagocytosis is subjective; thus, well trained individuals should be reading the slides and readings should be compared between individuals to insure similar results. The reading can take a half-day if looking at 300 monocytes per slide. The ADCC is also time consuming and purification of the NK cells can take a couple of hours as it involves isolating the PBMCs using Ficoll-Hypaque and then utilizing an NK purification kit. Opsonization of the RBCs takes about 1.5 hr and setting up the plates for reading in the scintillation counter can take a couple of hours. The plates also need to be left overnight prior to reading the radioactivity. Thus, the entire ADCC assay takes about 2 days, similar to the MMA assay.