Rh, MNS, and Other Blood Group Systems

The Rh system is the most complex of the major blood group systems, with over 50 antigens. Forensically, only five are relevant: D, C, c, E, and e, encoded by two closely linked genes on chromosome 1. RHD encodes the D antigen. RHCE encodes C/c and E/e in different combinations on the same protein (Cc, cE, CE, or ce haplotypes). The two genes are so close together that they are almost always inherited as a unit.

D is by far the most immunogenic Rh antigen. A D-negative person exposed to D-positive red cells through transfusion or pregnancy will in most cases make anti-D, which can cause haemolytic disease of the newborn. This clinical importance means D typing is performed on virtually every blood sample in clinical medicine, making the reference data for D frequency well established. Approximately 85% of European-origin populations are D-positive; approximately 15% are D-negative. Frequencies differ in other groups: fewer than 1% of East Asian populations are D-negative (typically 0.3-0.5% in Chinese, Japanese, and Korean groups); around 3-5% of African populations lack D by the conventional serological definition.

Forensic typing of Rh D was performed using indirect antiglobulin tests or enzyme-enhanced agglutination methods, since direct saline agglutination with anti-D does not work well. On fresh or lightly dried bloodstains, D typing was feasible. On stains more than a few weeks old, protein degradation usually destroyed D antigen detectability. Several studies from the 1970s and 1980s showed that C, c, E, and e antigens degraded even faster than D, limiting multi-antigen Rh typing to very fresh casework stains.

The MNS system antigens sit on glycophorin A (GPA) and glycophorin B (GPB), two heavily glycosylated transmembrane proteins on red cells. GPA carries M or N antigen depending on whether position 1 of the mature protein is serine (M) or leucine (N), determined by a single nucleotide difference in the GYPA gene. GPB carries S or s antigen based on a methionine/threonine difference at position 29 of the protein, encoded by GYPB.

Because M/N and S/s segregate independently within the MNS locus (they are on different genes, GYPA and GYPB, though closely linked on chromosome 4), there are six common combined phenotypes in most populations. The MN type is the most common single M/N genotype at around 50%, which limits the discriminating power of M/N typing alone. But combined with S/s and with ABO, the addition of MNS information meaningfully narrows the matching population fraction.

The Kell, Duffy, and Kidd systems each define pairs of antithetical antigens with different population frequencies. In forensic multi-system profiling, each system contributed additional discriminating power, with the caveat that each requires an additional set of reagents and that each antigen is protein-based and subject to degradation.

• Kell system (K and k antigen): the K (Kell) antigen is present in about 9% of European populations; k (Cellano) is present in over 99%. Because K is rare and k nearly universal, K typing had modest discriminating power in most casework but provided a strong match when a K-positive stain was found. Kell antigens are on the CD238 protein and are relatively fragile.
• Duffy system (Fya and Fyb): three common Duffy phenotypes exist: Fy(a+b-), Fy(a+b+), and Fy(a-b+). Frequency data differ markedly between populations: most West African individuals are Duffy-null (Fy(a-b-)), a variant that confers resistance to Plasmodium vivax malaria. In European populations, Fy(a+b+) is the commonest at about 49%.
• Kidd system (Jka and Jkb): three phenotypes similar to Duffy in frequency distribution. Kidd antigens are on the urea transporter SLC14A1. They are notoriously difficult to type reliably even in clinical settings because anti-Jka and anti-Jkb show dosage effects and can give weak reactions with heterozygous cells.

In practice, the UK forensic laboratories in the 1970s and 1980s typically ran ABO, secretor status, Rh D, and two or three enzyme polymorphisms (phosphoglucomutase isoforms, erythrocyte acid phosphatase) as their standard profile. The enzyme systems, not the Kell/Duffy/Kidd antigens, were the preferred supplementary markers because the enzyme assays worked better on older stain material than antibody-based antigen typing.

The discriminating power of combined serological typing can be estimated by multiplying the frequencies of matching phenotypes across independent systems. The calculation assumes the systems are truly independent, which is approximately but not perfectly true (some Rh haplotypes show population-level linkage disequilibrium with some MNS haplotypes). An example calculation for a European-origin population:

1. Group B (9%) x secretor (80%) x D-positive (85%) x Fy(a+b+) (49%) x PGM subtype 1+ (approximately 35%) = roughly 1.1% of the population.
2. A 1.1% match frequency means approximately one in ninety individuals in the reference population could provide a consistent set of typing results.
3. Compare to ABO alone (group B = 9%, or one in eleven): the multi-system profile is about eight times more discriminating for this combination.

The practical limit was that only some of these markers could be typed on most crime-scene stains. A stain a few days old in warm conditions would likely yield good ABO and secretor results but equivocal or negative Rh D and completely failed MNS results. The theoretical discriminating power of a six-system profile was only achievable on fresh bloodstains collected and stored well. For most actual casework stains, two to four markers was a realistic ceiling.

The difference in stability between ABO antigens and protein-based antigens has direct consequences for what can be typed from crime-scene material. ABO antigens are carbohydrates sitting on glycolipids and glycoproteins. Carbohydrates are chemically inert relative to proteins: they are not cleaved by proteases, they resist UV damage better, and they do not unfold under heat in the way a protein does. Well-documented studies have typed ABO antigens from stains decades old.

The forensic implication is clear: a report saying a stain was Rh D-positive typed by indirect antiglobulin test on a six-week-old outdoor bloodstain should be treated with caution. The antigen may have been present but at such reduced density that a false-negative was equally possible. Responsible forensic reporting from this era documented the age and condition of the stain alongside the typing result.

Alec Jeffreys' first DNA fingerprint was published in 1984. By 1987 it was in operational forensic use in the UK. By the mid-1990s, polymerase chain reaction-based STR profiling was in use in laboratories across North America, Europe, and Australia. The random match probability for a full 13-locus CODIS STR profile is in the order of one in a quadrillion; multi-system serology, at its best, reached one in a hundred.

Multi-system blood-group typing also had practical disadvantages that became visible by comparison to DNA: it required fresh material, it consumed the sample, it needed batteries of expensive antisera with limited shelf lives, and its results were class-level even at their most informative. DNA profiling worked better on smaller and older samples, consumed less material per test, and gave near-individual-level discrimination. Forensic laboratories dismantled their multi-system serology capabilities through the 1990s and early 2000s.

• Historical case review: forensic analysts must understand multi-system typing to evaluate whether pre-DNA evidence was correctly interpreted and whether DNA retesting should be pursued.
• Interpretation of cold-case serological reports: knowing the limitations of Rh and MNS typing on aged stains helps a reviewer assess whether a historical exclusion or inclusion was reliable.
• Mass-casualty identification: ABO typing retains a triage role; Rh D typing from ante-mortem blood samples on medical records can provide a supplementary check against post-mortem samples when DNA is unavailable or incomplete.