Plant cell screen separates fast growth from staying power

A plant cell culture that grows quickly at first may not keep growing through repeated cultivation. By combining automated imaging with a large genetic screen, researchers found examples of that distinction—and evidence that bigger tissue cultures do not necessarily mean bigger plants.

Growing more plant tissue is not a single biological achievement. In a large Arabidopsis screen, researchers found that tissue size on solid medium, whole-plant growth, biomass accumulation in liquid culture and persistence through repeated transfers did not consistently move together. The findings suggest that selecting plant cells for production will require testing several aspects of performance rather than relying on one early growth measurement.

The study, led by Fuensanta Verdú-Navarro and colleagues, is a bioRxiv preprint that has not been certified by peer review. It presents a screening framework and candidate genetic targets, not a demonstrated improvement in commercial production.

Measuring thousands of tissue cultures

The team screened 2,880 Arabidopsis thaliana lines carrying T-DNA insertions, which can disrupt genes or alter their expression. Seeds were grown directly on hormone-containing medium to produce calli—masses of plant tissue maintained in culture. Each screening plate contained 15 mutant lines and a wild-type control, with nine seeds per genotype.

After 28 days, a standardized imaging chamber and automated analysis pipeline measured callus area. A deep-learning model detected calli and non-germinated seeds; subsequent image processing outlined the tissue and counted its pixels. This made area measurement scalable, although projected area remains a proxy for tissue size rather than a direct measurement of cell division or biomass.

The authors report analysing 27,648 calli. Germination was an important complication: 723 lines had reduced or absent germination, including 234 with none. Lines with fewer than four germinated seedlings were excluded from quantitative size comparisons, and 13 plates with abnormal wild-type growth were excluded from subsequent analyses.

Many smaller cultures, few larger candidates

Relative to the overall wild-type mean, 2,456 mutant lines produced smaller calli and 190 produced larger ones. Those descriptive counts should not be confused with statistically supported growth differences.

When mutants were compared with controls on the same plate, 19 lines showed increased callus size at an unadjusted significance threshold of P < 0.05. After correction for multiple testing, only one high-growth line remained significant at a false discovery rate below 0.05. The researchers advanced 15 candidates based on their plate-matched growth phenotype and confirmation of the insertion.

That distinction substantially tempers the screen's conclusions: most selected lines were candidates requiring follow-up, rather than discoveries supported by the corrected primary-screen statistics. Many of the smallest-callus lines also germinated poorly, complicating interpretations of reduced size as a specific proliferation defect.

Expression measurements added biological context without settling causality. Some selected lines lacked detectable transcripts for the affected gene, whereas others showed reduced or increased expression. One candidate had approximately 4.27 times the wild-type expression of NDB1, which encodes a mitochondrial enzyme. Two other candidates showed no significant transcript change. The authors acknowledge that additional mutations or insertion effects not captured by transcript abundance could explain those phenotypes. Confirming an insertion and an expression difference does not, by itself, prove that the named gene caused the growth change.

A fast start did not guarantee persistence

Selected high-callus-growth mutants produced larger, smaller or apparently unchanged leaf rosettes when grown as plants. These observations support a limited but useful conclusion: enhanced growth in the tested culture conditions did not reliably predict enhanced vegetative growth.

The team then tested five candidates in liquid culture. Over a seven-day cycle, two lines—SALK_036374C and SALK_086651C—showed approximately 1.27 and 2.47 times the wild-type net biomass gain, respectively, after adjustment for initial inoculum volume. Growth was assessed using sedimented packed cell volume, not direct dry-mass measurements.

Repeated transfers into fresh medium revealed a different ranking. SALK_086651C, the stronger short-term accumulator, lost proliferative capacity after four passages. Wild-type cultures declined markedly by the fifth and could not be maintained after the sixth. In contrast, SALK_036374C and SALK_023787C remained vigorously proliferative through seven passages, when the experiment ended.

Seven passages do not establish indefinite growth. Nevertheless, the contrast shows why early accumulation and sustained propagation need separate assessment under these conditions.

For plant cell biotechnology, the immediate contribution is a way to identify and compare promising lines. Whether their advantages persist in bioreactors, transfer to other species or coexist with high yields of a desired product remains untested. The practical lesson is narrower than a production breakthrough: larger cultures, faster accumulation and longer-lasting growth are different selection criteria.