Plasma Proteins and Electrophoresis

Protein studies measure the total amount, show its distribution among fractions, or identify the molecule responsible for an abnormal result. Total protein and albumin measure quantity, electrophoresis shows distribution, and immunochemical and mass spectrometric methods identify individual proteins. Interpretation depends on the method, specimen, pattern, and clinical setting.

Amino acids

An amino acid contains an amino group, a carboxyl group, a hydrogen atom, and a variable side chain attached to an alpha carbon. The side chain determines charge, polarity, and many of the chemical interactions that shape a protein. At physiologic pH, most free amino acids contain both a positively charged amino group and a negatively charged carboxyl group. This dipolar form is a zwitterion. The pH at which a molecule has no net electrical charge is its isoelectric point, or pI.¹

A peptide bond joins the carboxyl group of one amino acid to the amino group of another and releases water. Two joined residues form a dipeptide; longer chains form polypeptides. Protein digestion reverses these bonds by hydrolysis. Gastric acid unfolds dietary proteins, and pepsin begins proteolysis. Secretin promotes pancreatic bicarbonate secretion to neutralize acid in the duodenum, while cholecystokinin promotes secretion of pancreatic proteases. Trypsin, chymotrypsin, elastase, and carboxypeptidases produce smaller peptides. Brush-border and intracellular peptidases complete digestion, and transporters carry amino acids and small peptides into enterocytes for delivery to the portal circulation.¹

Nutritional requirements vary with age, growth, illness, and metabolic defects, so the adult categories below are only a starting point.¹

Leucine, isoleucine, and valine are the branched-chain amino acids. Their shared early catabolic steps help produce the characteristic profile in maple syrup urine disease.

Selenocysteine is a specialized selenium-containing amino acid. Cells synthesize it on its transfer RNA from serine and insert it during programmed recoding of selected UGA codons. It does not fit the ordinary dietary classification used for the 20 standard amino acids.¹

Quantitative plasma amino acids

Quantitative amino acid profiles support the investigation and monitoring of inherited metabolic disorders. The pattern matters more than one isolated concentration. Interpretation also requires the patient’s age, fasting state, diet, medications, treatment, and condition at collection.

Aminoacidopathies result from defects in an enzyme, cofactor, or transporter. A profile may show a deficient product, an accumulated substrate, or metabolites produced through alternate pathways. Prompt recognition and treatment are necessary when accumulated compounds are neurotoxic. Cells continue to metabolize amino acids after collection, so delayed separation or transfer of the buffy coat can change the measured pattern.^1,2

Collection and handling are method-specific. One current liquid chromatography with tandem mass spectrometry procedure requires a 4-hour fast for older children and adults, sodium-heparin plasma, wet ice after collection, separation immediately or within 4 hours under refrigeration, transfer without buffy coat, and frozen transport. That procedure does not measure total homocysteine, so a separate assay is required when a homocysteine disorder is under investigation. The performing laboratory’s instructions govern the actual specimen.²

Protein structure, synthesis, and classification

Protein structure is described at four levels.

Denaturation changes secondary, tertiary, or quaternary structure and usually impairs function. Peptide-bond cleavage changes primary structure and is proteolysis or degradation. This distinction matters because a denatured protein can retain its amino acid sequence while losing activity or solubility.¹

Ribosomes translate messenger RNA into polypeptides. Folding, cleavage, glycosylation, phosphorylation, and other post-translational changes produce the mature protein. Hepatocytes make albumin, transferrin, fibrinogen, and many complement and coagulation proteins. Plasma cells make immunoglobulins, and tissue-specific proteins enter blood from other organs.

Proteins can be classified by shape or composition. Globular proteins include albumin, enzymes, and immunoglobulins. Fibrous proteins include collagen and fibrin. A simple protein yields amino acids when completely hydrolyzed. A conjugated protein also contains a nonprotein component. Hemoglobin is a hemoprotein, lipoproteins carry lipid, metalloproteins bind a metal, glycoproteins and proteoglycans contain carbohydrate, and nucleoproteins contain nucleic acid. Immunoglobulins are glycoproteins.¹

Protein solubility often reaches a minimum near the pI because reduced net charge decreases electrostatic repulsion among molecules. In isoelectric focusing, a protein migrates through a stable pH gradient until it reaches the pH that matches its pI. The method’s high resolution allows it to resolve cerebrospinal fluid oligoclonal bands.

Protein turnover continually releases amino acids for reuse or catabolism. Nitrogen balance is the difference between nitrogen intake and nitrogen loss. Transamination transfers an amino group to an alpha-keto acid, while oxidative deamination can release ammonia. The liver converts much of that nitrogen to urea. The remaining carbon skeleton is glucogenic, ketogenic, or both according to the pathway it enters.¹

Plasma proteins by electrophoretic fraction

At the alkaline pH commonly used for serum protein electrophoresis, most major serum proteins have a net negative charge. They migrate toward the anode at rates determined mainly by charge, with size, shape, support medium, and electro-osmotic effects also contributing. The familiar fractions are prealbumin, albumin, alpha-1, alpha-2, beta, and gamma.¹

Immunoglobulin chain structure, classes, class switching, and antibody function are covered in Immune System Physiology and Immunoglobulins. That biology guides the interpretation of quantitative immunoglobulin, electrophoresis, immunofixation, and free light-chain results.

Albumin and transthyretin concentrations reflect inflammation, distribution, synthesis, and loss. They should not be used as stand-alone measures of body protein stores or muscle mass.³ Low albumin can result from inflammation, reduced hepatic synthesis, renal or gastrointestinal loss, increased catabolism, dilution, or movement into the extravascular space. Hyperalbuminemia usually reflects reduced plasma water rather than increased synthesis.

Glycated albumin reflects shorter-term glycemia but remains method dependent. Myoglobin is interpreted with heme proteins and muscle injury. Cystatin C and beta-2-microglobulin require renal filtration and tubular context, which is covered in Renal Physiology, Function Testing, and Disease Patterns. Cardiac troponins and natriuretic peptides require tissue-injury and disease-marker context. Alpha-fetoprotein, fetal fibronectin, and bone-turnover proteins also require indication-specific methods and decision limits; their concentration alone does not define a diagnosis.

Total protein and albumin: methods and interpretation

Measurement methods

The validated reagent and platform procedure specifies measurement wavelength and reaction timing. In one current biuret reagent, tartrate prevents copper hydroxide precipitation and iodide prevents copper autoreduction.⁶ Ultraviolet absorbance near 280 nm detects aromatic amino acids but lacks selectivity in a mixture such as serum. Protein stains, including amido black, Ponceau S, and Coomassie dyes, visualize separated bands. Immunoassays add molecular specificity when the target is a named protein. Bilirubin, hemoglobin, globulin composition, renal disease, and specimen matrix can produce method-dependent albumin dye bias, so the procedure’s interference data govern result review.^1,5

Serum and plasma total protein are similar, but plasma retains fibrinogen. Serum is preferred for serum protein electrophoresis because residual fibrinogen can create a band between the beta and gamma regions. Prolonged tourniquet use and standing can concentrate proteins through loss of plasma water. Intravenous fluid contamination can dilute them. The laboratory’s interference and specimen rules determine the action for hemolysis, lipemia, and icterus.⁶

Reference intervals depend on age, specimen, assay, and laboratory population. Interpret results against the interval supplied with the report. A high total protein result can reflect dehydration or increased globulins, including polyclonal or monoclonal immunoglobulins. A low result can reflect dilution, reduced synthesis, protein loss, or increased catabolism. Total protein can remain within its interval when one fraction rises as another falls.

Calculated globulin and the albumin/globulin ratio use total protein and albumin from the same specimen in the same units:

calculated globulin = total protein − albumin

albumin/globulin ratio = albumin ÷ calculated globulin

A zero or negative calculated globulin makes the ratio invalid and prompts verification of specimen, units, and analytical flags. The full calculation and a worked example are in Core Chemistry Calculations.

Protein electrophoresis

Separation and measurement

Agarose gel electrophoresis commonly uses an alkaline buffer near pH 8.4 to 8.8 and separates serum proteins into bands, which are stained and scanned by a densitometer. Capillary zone electrophoresis separates proteins in narrow capillaries and measures absorbance directly as they pass a detector. Both systems commonly report albumin, alpha-1, alpha-2, beta, and gamma fractions, but their fraction boundaries, resolution, and interference patterns differ. High-resolution agarose and capillary methods can resolve additional bands; the number is platform-dependent. Isoelectric focusing separates proteins by pI and is especially useful for restricted immunoglobulin bands in cerebrospinal fluid. Laboratories validate the method, review flagged traces, and establish their own reference intervals.¹

An absolute fraction concentration can be calculated from its densitometric percentage and the total protein measured on the same specimen. If albumin is 60% of a total protein concentration of 6.8 g/dL:

albumin fraction = 0.60 × 6.8 g/dL = 4.08 g/dL

This estimate inherits error from the total-protein assay, trace integration, and peak-boundary selection.

Electrophoretic mobility relates migration distance to time and electric field strength:

mobility (cm²/[V·s]) = migration distance (cm) ÷ [time (s) × field strength (V/cm)]

For a band that travels 4.2 cm in 4,200 seconds at 10 V/cm:

mobility = 4.2 ÷ (4,200 × 10) = 0.00010 cm²/(V·s)

The calculation describes movement under the stated conditions. It does not make results from different buffers, temperatures, support media, or instruments interchangeable.

Pattern recognition

Electrophoretic patterns narrow the differential but require the specimen, quantitative results, and clinical context for interpretation.^1,7

An inherited or acquired bisalbuminemia can split the albumin peak. The paired peaks have little clinical effect by themselves, but recognizing the pattern prevents an albumin variant from being mistaken for a restricted globulin.¹

Artifacts can resemble disease. Plasma fibrinogen may form a restricted band near the beta-gamma interface. Hemolysis can add hemoglobin-related bands and complicate haptoglobin interpretation. High C-reactive protein can form a restricted band on some systems. Radiographic contrast, certain antimicrobials, and therapeutic monoclonal antibodies can also alter a trace. Previous results, specimen type, medication history, and confirmatory testing help resolve these findings.^1,7

Monoclonal protein studies

A monoclonal protein is an immunoglobulin or immunoglobulin component produced by one clone. The laboratory finding can accompany monoclonal gammopathy of undetermined significance, plasma-cell neoplasms, lymphoplasmacytic disorders, amyloid light-chain amyloidosis, and other conditions. A monoclonal IgM result occurs in several of these disorders, so isotype alone does not assign the diagnosis. The immune disease module covers the clinical categories.

Current College of American Pathologists guidance supports this initial laboratory approach when a monoclonal gammopathy is suspected:⁷

1. Perform serum protein electrophoresis and serum free light chains.
2. If serum electrophoresis shows a monoclonal protein or the serum free light-chain ratio is abnormal, perform serum immunofixation or a validated equivalent.
3. For suspected amyloid light-chain amyloidosis, order serum protein electrophoresis, serum free light chains, serum immunofixation, and urine immunofixation together because the monoclonal protein may be very small.
4. Correlate all results because no single method detects every monoclonal protein.

Immunofixation electrophoresis separates proteins in parallel lanes and applies antisera to heavy and light chains. A restricted band that reacts with one heavy-chain reagent and one light-chain reagent identifies the isotype. Immunofixation is more sensitive than routine serum protein electrophoresis for small proteins but is primarily qualitative. Immunosubtraction can provide an equivalent identification step on validated capillary systems. Some laboratories use mass spectrometry to distinguish endogenous monoclonal proteins from therapeutic antibodies or to characterize difficult findings. If a restricted kappa or lambda band has no associated IgG, IgA, or IgM band, IgD and IgE testing is required before the protein is reported as light-chain only.⁷

Serum free light-chain testing measures unbound kappa and lambda light chains and calculates a kappa/lambda ratio. Renal impairment can raise both concentrations and change the expected ratio. Inflammation can also raise both. Assays differ in calibration and antibody reactivity, so the method’s reference interval and renal interpretation rules apply. Antigen excess, nonlinearity, polymerization, and lot-to-lot behavior are additional analytical concerns.⁸

When a restricted peak can be measured reliably, densitometry estimates the M-protein concentration. Co-migration with transferrin, complement, or another protein can make that estimate inaccurate, especially in the beta region. A small peak may still be clinically important, and a large peak can be analytically difficult. Trend interpretation uses the same method whenever possible.

Proteins in urine

Urine protein testing distinguishes albumin-predominant glomerular loss, low-molecular-weight tubular protein loss, overflow of a filtered protein, and protein from postrenal inflammation or bleeding. The reagent-strip module covers the protein error of indicators, the strip’s albumin preference, and confirmatory testing. Quantitative studies add concentration, creatinine normalization, and protein identity.

Automated urine total-protein assays commonly use dye-binding or turbidimetric reactions. Pyrogallol red-molybdate provides sensitivity at urine concentrations. Precipitation methods such as sulfosalicylic acid detect several protein classes but are affected by specimen turbidity, drugs, and radiographic contrast. Modified biuret and Folin-based methods also exist. These reactions have unequal responses to albumin, immunoglobulin light chains, and low-molecular-weight proteins, and no current higher-order system makes results from all manufacturers interchangeable.⁹

Urine protein electrophoresis can support the mechanism of proteinuria. An albumin-dominant trace supports glomerular loss. Multiple low-molecular-weight bands support impaired tubular reabsorption. A restricted free light-chain band supports overflow proteinuria and requires immunofixation for identity. Mixed and postrenal patterns occur, so the trace does not establish the diagnosis by itself.^1,7

KDIGO classifies albuminuria by ACR as A1 below 30 mg/g, A2 from 30 through 300 mg/g, and A3 above 300 mg/g. The equivalent SI decision points are below 3, from 3 through 30, and above 30 mg/mmol. A random ACR at or above 30 mg/g should be confirmed when appropriate, preferably with a subsequent first-morning void, because exercise, infection, menstruation, hydration, and day-to-day biological variation can affect a result.^10,11

Worked ACR example. Urine albumin is 66 mg/L and urine creatinine is 1.50 g/L.

ACR = 66 mg/L ÷ 1.50 g/L = 44 mg/g

The liters cancel, leaving milligrams of albumin per gram of creatinine. This result is in category A2. The category describes albumin excretion and must be combined with persistence, estimated glomerular filtration rate, and the clinical setting.

Worked PCR example. Urine total protein is 52 mg/dL and urine creatinine is 104 mg/dL.

PCR = 52 ÷ 104 = 0.50 mg/mg = 500 mg/g

The local method and clinical pathway determine interpretation of PCR. Timed excretion calculations and collection checks are in Timed Urine and Renal Clearance Calculations.

Serum protein electrophoresis, serum immunofixation, and serum free light chains detect most monoclonal gammopathies more effectively than a urine screen alone. Urine immunofixation is used in selected settings, including suspected amyloid light-chain amyloidosis and some monoclonal kidney disorders. A 24-hour urine collection may be requested when quantitative urinary M-protein excretion is needed.⁷

Proteins in cerebrospinal fluid

Cerebrospinal fluid contains far less protein than serum. Concentration varies with age and collection site; lumbar fluid usually contains more protein than ventricular fluid. Blood contamination can add albumin and immunoglobulins. Quantitative methods therefore need a measuring interval and blank appropriate for low concentrations. Laboratories may use reflectance, turbidimetric, dye-binding, or immunochemical methods, with method- and age-specific reference intervals. Increased total protein is nonspecific and can accompany blood-CSF barrier dysfunction, inflammation or infection, hemorrhage, impaired CSF flow, tumor, or demyelinating injury. Exam-only intervals are listed separately in ASCP BOC Examination Reference Ranges.^1,12

After concentration, a normal CSF electrophoretic pattern can show prealbumin, albumin, alpha-1, alpha-2, beta, and gamma proteins. Albumin remains the largest fraction. Beta-1 transferrin is the serum form. Beta-2 transferrin, also called the tau fraction in older electrophoresis terminology, is a carbohydrate-deficient transferrin form found in CSF and used as a marker of CSF in nasal, otic, or wound fluid. Immunoglobulins contribute to the gamma region. Fraction names and migration positions depend on the platform.^1,13

The albumin quotient evaluates transfer of a liver-derived protein across the blood-CSF barrier. Convert the paired CSF and serum albumin concentrations to the same units before division:

albumin quotient (QAlb) = CSF albumin ÷ serum albumin

For CSF albumin 180 mg/L and serum albumin 40,000 mg/L:

QAlb = 180 ÷ 40,000 = 0.0045 = 4.5 × 10⁻³

QAlb decision limits are age- and procedure-dependent. An increased quotient supports impaired blood-CSF barrier function or altered CSF flow. It does not identify a cause.

The IgG index compares the CSF-to-serum IgG ratio with the paired albumin ratio:

IgG index = (CSF IgG ÷ serum IgG) ÷ (CSF albumin ÷ serum albumin)

For CSF IgG 4.0 mg/dL, serum IgG 1,000 mg/dL, CSF albumin 20 mg/dL, and serum albumin 4.0 g/dL (4,000 mg/dL):

IgG index = (4.0 ÷ 1,000) ÷ (20 ÷ 4,000) = 0.004 ÷ 0.005 = 0.80

One current laboratory procedure uses 0.70 as its upper reference limit, so 0.80 is increased for that procedure. An increased index supports intrathecal IgG production but does not establish a specific neurologic diagnosis.¹⁴

Oligoclonal band testing uses isoelectric focusing followed by IgG-specific immunoblotting on paired CSF and serum. The serum comparison separates intrathecal bands from circulating bands that crossed into CSF.¹⁵

Intrathecal IgG synthesis occurs in multiple sclerosis and in other inflammatory or infectious neurologic disorders. Myelin basic protein is a nonspecific marker of myelin injury. Alzheimer disease biomarker panels use procedure-specific combinations and decision limits. These results require the validated neurologic testing pathway. Cell counts, appearance, and infectious patterns are covered in Body Fluid Cell Counts and Cerebrospinal Fluid.

References

1. Bishop ML, Fody EP, Van Siclen C, Mistler JM, Moy M. Clinical Chemistry: Principles, Techniques, and Correlations. 9th ed. Jones & Bartlett Learning; 2023.
2. Mayo Clinic Laboratories. Amino Acids, Quantitative, Plasma. Accessed August 30, 2026.
3. Evans DC, Corkins MR, Malone A, et al. The use of visceral proteins as nutrition markers: an ASPEN position paper. Nutr Clin Pract. 2021;36(1):22-28. doi:10.1002/ncp.10588.
4. Joint Committee for Traceability in Laboratory Medicine. C1RMS118: total protein in blood serum, blood plasma. Updated August 29, 2025. Accessed August 30, 2026.
5. Panteghini M, Auclair G, Thelen M, Wu L, Wielgosz R. A roadmap for standardization of serum albumin measurements. Clin Chem Lab Med. Published online August 24, 2026. doi:10.1515/cclm-2026-1215.
6. Mayo Clinic Laboratories. Protein, Total, Serum. Accessed August 30, 2026.
7. Keren DF, Bocsi G, Billman BL, et al. Laboratory detection and initial diagnosis of monoclonal gammopathies. Arch Pathol Lab Med. 2022;146(5):575-590. doi:10.5858/arpa.2020-0794-CP.
8. Mayo Clinic Laboratories. Immunoglobulin Free Light Chains, Serum. Accessed August 30, 2026.
9. Miller WG, Seegmiller JC, Bachmann LM, Panteghini M. Measuring urine albumin and total protein. Nephrol Dial Transplant. 2026;41(suppl 2):ii4-ii11. doi:10.1093/ndt/gfag060.
10. Kidney Disease: Improving Global Outcomes CKD Work Group. KDIGO 2024 clinical practice guideline for the evaluation and management of chronic kidney disease. Kidney Int. 2024;105(4S):S117-S314. doi:10.1016/j.kint.2023.10.018.
11. Seegmiller JC, Bachmann LM. Urine albumin measurements in clinical diagnostics. Clin Chem. 2024;70(2):382-391. doi:10.1093/clinchem/hvad174.
12. Mayo Clinic Laboratories. Protein, Total, Spinal Fluid. Accessed August 30, 2026.
13. Mayo Clinic Laboratories. Beta-2 Transferrin: Detection of Spinal Fluid in Other Body Fluid. Accessed August 30, 2026.
14. Mayo Clinic Laboratories. Multiple Sclerosis Profile, Serum and Spinal Fluid. Accessed August 30, 2026.
15. Mayo Clinic Laboratories. Oligoclonal Banding, Serum and Spinal Fluid. Accessed August 30, 2026.