Lonza Bioscience blog

Albumin is the most abundant plasma protein synthesized by hepatocytes and plays central roles in maintaining oncotic pressure, transporting molecules, and supporting metabolic homeostasis. In vitro, albumin secretion is commonly used as a functional marker of hepatocyte phenotype, reflecting the ability of cultured cells to maintain differentiated hepatic functions over time. In liver culture systems—including spheroid culture, liver-on-a-chip and other micro physiological models—albumin production is frequently monitored alongside metabolic and toxicological endpoints to assess long-term culture stability and hepatic functionality.
 

Studies have shown that preservation of albumin secretion is often associated with improved maintenance of liver-specific functions, particularly in three-dimensional (3D) culture systems that more closely recapitulate aspects of the in vivo liver microenvironment.^1,2

To better understand donor variability, albumin secretion was evaluated across 26 primary human hepatocyte (PHH) donors cultured in both two-dimensional (2D) sandwich and 3D spheroid formats. Albumin production was assessed on day 5 in 2D cultures and day 7 in 3D spheroids. Considerable variability in albumin secretion was observed among donors in both culture systems. Notably, albumin output did not show a clear association with commonly used quality-control attributes such as post-thaw viability, attachment efficiency, or CYP3A4 activity. These observations are consistent with recent efforts to define critical quality attributes for PHHs, which emphasize that functional parameters may provide complementary information rather than serving as standalone indicators of overall cell quality.^3,4 
 

To investigate whether baseline albumin secretion influences toxicological responses, four representative donors spanning a range of albumin-production levels were exposed to acetaminophen (APAP). While donor-specific differences in APAP sensitivity were observed, no clear relationship emerged between albumin secretion and APAP EC₅₀ values. Instead, variation in drug response appeared more closely associated with donor-specific metabolic characteristics, including CYP3A4 activity. These findings suggest that albumin secretion alone may not predict hepatotoxicity outcomes in routine screening assays.⁴

Consistent with previous reports, hepatocytes cultured as 3D spheroids generally exhibited lower APAP EC₅₀ values than those maintained in 2D sandwich cultures, indicating enhanced sensitivity to compound-induced toxicity. Bell et al. demonstrated that 3D hepatocyte spheroids maintain liver-specific functions, including albumin secretion and xenobiotic metabolism, for extended periods and provide improved performance in long-term toxicity assessments relative to conventional 2D cultures.¹  Similarly, emerging microphysiological liver models have highlighted the importance of sustained hepatic function for disease modeling and drug evaluation applications.²

Taken together, these findings indicate that albumin secretion may serve as a useful functional descriptor when characterizing hepatocyte donor performance. However, albumin production should be interpreted alongside other donor attributes, including metabolic competence, inducible enzyme activity, and application-specific performance metrics. The absence of a direct relationship between albumin secretion and APAP sensitivity suggests that no single parameter is sufficient to predict suitability across all experimental contexts.^3,4

Lonza offers human hepatocytes (HUCPI, HUCPG). To facilitate more informed donor selection, Lonza will report albumin secretion data from hepatocytes cultured in 2D sandwich culture through Day 5. Such information may be particularly valuable when selecting hepatocyte lots for long-term studies, microphysiological systems, organ-chip platforms, and chronic toxicity investigations, where sustained cellular function is often desired. Conversely, donors exhibiting lower albumin secretion may remain suitable for acute toxicity screening, mechanistic studies, and exploratory ADME-Tox applications, depending on the experimental objectives.

Written by
Ming

Scientific Support Specialist

References

1. Bell, C.C., et al. 2018. Comparison of hepatic 2D sandwich cultures and 3D spheroids for long-term toxicity applications: A multicenter study. Toxicological Sciences 162(2): 655-666. DOI: 10.1093/toxsci/kfx289
    
2. Bonanini, F., Dinkelberg, R., Caro Torregrosa, M., et al. 2025. A microvascularized in vitro liver model for disease modeling and drug discovery. Biofabrication 17:015007. DOI: 10.1088/1758-5090/ad818a
    
3. Peng, Z., Wu, J., Zhang, X. et al. The validation of quality attributes in Primary Human Hepatocytes Standard. Cell Regen 14, 41 (2025). DOI: 10.1186/s13619-025-00258-6
    
4. Nautiyal M. et al., 2026. Preliminary evaluation of donor variability in albumin secretion for fit-for-purpose applications of primary human hepatocytes. https://bioscience.lonza.com/download/content/asset/40797