#
The CMV-Specific CD8^{+} T Cell Response Is Dominated by Supra-Public Clonotypes with High Generation Probabilities

^{1}

^{2}

^{3}

^{*}

## Abstract

**:**

^{b}/SIINFEKL-recognizing TCR CDR3α or CDR3β sequences from 25 individual mice spanning seven different time points during acute infection and memory inflation. In-depth repertoire analysis revealed that from a pool of highly diverse, but overall limited sequences, T cell responses were dominated by public clonotypes, partly with unexpectedly extreme degrees of sharedness between individual mice (“supra-public clonotypes”). Public clonotypes were found exclusively in a fraction of TCRs with a high generation probability. Generation probability and degree of sharedness select for highly functional TCRs, possibly mediated through elevating intraindividual precursor frequencies of clonotypes.

## 1. Introduction

^{20}or in some estimates even 10

^{61}different solutions for TCRs [19,20]. The actual diversity in a single mouse [19] or human being [21] is considerably lower, but still believed to be in the range of about 10

^{7}–10

^{8}unique TCRs. It remains largely unanswered how many TCRs fit to a given antigen in principle (across different individuals) and how many TCRs contribute to an antigen-specific T cell response in a single individual.

## 2. Results

^{b}/SIINFEKL-specific CD8+ T cell population from peripheral blood by flow cytometry at day 7, 15, 30, 58, 86, 120 and 170 post infection with sort purities of at least 99.5% [29]. Following RNA extraction and rapid amplification of cDNA ends (RACE) PCR, CDR3α and CDR3β bulk sequencing was performed. Sequences were considered only if an intact unique molecular identifier (UMI) was present and found more than once in a given sample. Clonotypes were defined as chains with identical CDR3 amino acid sequences. For further details see Materials and Methods and references [29].

#### 2.1. TCR Repertoire Evolution Stabilizes during Memory Inflation

#### 2.2. Public Clonotypes Lead to Repertoire Saturation across Individual Mice

^{b}/SIINFEKL is highly diverse, but limited to an extent that makes reliable estimation of absolute numbers feasible for future investigation. Second, the number of unique SIINFEKL-specific CDR3α sequences is lower than the number of unique CDR3β sequences.

#### 2.3. (Supra-) Public Clonotypes Are Preferentially Expanded during the Immune Response

#### 2.4. TCR Generation Probability Is Linked with Degree of Sharedness and Thereby Clonal Expansion

^{−25}) were found exclusively for private clonotypes (Figure 4a). Since there are also more private than public clonotypes, we calculated the median generation probability. Irrespective of whether clonotypes were included for which no generation probability could be calculated (presumably, but not necessarily, because it was too low; gen prob = “0”), the median generation probability rose from private to public clonotypes by several orders of magnitude (Figure 4a,b). Nearly all of the clonotypes for which no generation probability could be calculated were private, which explains why the median shifted the most for private clonotypes (Figure 4b).

^{−17}(Figure 4d). Since mostly clones with a high generation probability expanded, the mean generation probability was as high as 10

^{−6}, which did not change over time (Figure 4e). This was further confirmed by focusing on the dominating clones at the peak of the acute immune response (day 15 post infection) and during memory inflation (day 120 post infection). The dominating clones (i.e., clones with a clone fraction > 0.1) had exclusively very high generation probabilities (Figure 4f). These high generation probabilities led in most cases to very high degrees of sharedness (median of 7 for day 15 and day 120 post infection). In turn, clones with a high degree of sharedness and/or generation probability represented higher median clone fractions (Figure 4g). We therefore finally explored whether generation probability and degree of sharedness could serve as predictors of TCRs that clonally expand within an antigen-specific immune response. In fact, both generation probability and degree of sharedness independently selected for clones with a higher clone fraction, with the combination of both parameters being the filter with the highest enrichment of expanding clones (Figure 4h). Among supra-public clonotypes with a generation probability of 10

^{−7}there was a 18.8% hit rate of clones with a clone fraction > 0.1. This hit rate is 0.2% without any filter, indicating a 94-fold enrichment of clones that will at some point contribute more than 10% (clone fraction of > 0.1) to an overall antigen-specific T cell response.

#### 2.5. The Nucleotide-to-Amino Acid Ratio of Intraindividual TCR Sequences Is Associated with Interindividual Degrees of Sharedness via Generation Probability

^{−5}) and a maximum degree of sharedness (12/12). Clonotypes with a nt:aa ratio of 3, 2 or 1 had median generation probabilities and degrees of sharedness in descending respective orders. We finally investigated whether we could also directly observe that heightened nt:aa ratios would result in higher clone fractions. Indeed, median clone fractions rose with increasing nt:aa ratios from 1 to 3. For the two instances in which the clonotype CASSRTGEQYF was observed in the form of four different nucleotide sequence variants, it did not reach a high clone fraction. Overall, these data suggest that increased precursor frequencies, as potentially indicated by heightened nt:aa ratios, are not only a function of increased generation probabilities, but also result in higher degrees of sharedness. On average (as expressed by the median), higher nt:aa ratios result in higher clone fractions, but more and more precise data are needed to conclusively address this.

## 3. Discussion

^{−17}), but did not further predict the degree of sharedness. The latter stands in contrast to previous reported work, which however only investigated TCR sequences in bulk (and did not focus on certain antigen-specificities) and thereby also analyzed a much larger number of sequences [20]. We hypothesize that, apart from sampling bias, TCR affinity could be a relevant confounding factor in this regard [18]. Overall, dominating clones at the acute phase of the immune response and during memory inflation showed high degrees of sharedness as well as high generation probabilities and we could show that the combination of both parameters can serve as an enrichment filter for identifying clones that will contribute significantly to the antigen-specific T cell response (arguably representing highly functional TCRs). This may prove useful for in silico prediction of TCRs with clinical potential in the context of adoptive cell therapy.

## 4. Materials and Methods

#### 4.1. Experimental Data

^{b}/SIINFEKL-specific CD8+ T cell population by flow cytometry at day 7, 15, 30, 58, 86, 120 and 170 post infection with sort purities most often reaching 100% [23]. For CDR3 sequencing, RNA of sorted antigen-specific T cells was reverse-transcribed into cDNA using a biotinylated oligo dT primer. An adaptor sequence was added to the 3’ end of all cDNA, which contains the Illumina P7 universal priming site and a 17-nucleotide UMI. Products were purified using streptavidin-coated magnetic beads followed by a primary PCR reaction using a pool of primers targeting the TCRα and TCRβ regions, as well as a sample-indexed Illumina P7C7 primer. The TCR-specific primers contained tails corresponding to the Illumina P5 sequence. PCR products were then purified using AMPure XP beads. A secondary PCR was performed to add the Illumina C5 clustering sequence to the end of the molecule containing the constant region. The number of secondary PCR cycles was tailored to each sample to avoid entering plateau phase, as judged by a prior quantitative PCR analysis. Final products were purified, quantified with Agilent Tapestation and pooled in equimolar proportions, followed by high-throughput paired-end sequencing on the Illumina MiSeq platform. For sequencing, the Illumina 600 cycle kit was used with the modifications that 325 cycles was used for read 1, 6 cycles for the index reads, 300 cycles for read 2 and a 20% PhiX spike-in to increase sequence diversity.

#### 4.2. Clone Identification

#### 4.3. Generation Probability

#### 4.4. Statistical Analyses

## Author Contributions

## Funding

## Acknowledgments

## Conflicts of Interest

## References

- Griffiths, P.; Baraniak, I.; Reeves, M. The pathogenesis of human cytomegalovirus. J. Pathol.
**2015**, 235, 288–297. [Google Scholar] [CrossRef] [PubMed] - Blyth, E.; Withers, B.; Clancy, L.; Gottlieb, D. CMV-specific immune reconstitution following allogeneic stem cell transplantation. Virulence
**2016**, 7, 967–980. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Starr, S.E.; Allison, A.C. Role of T lymphocytes in recovery from murine cytomegalovirus infection. Infect. Immun.
**1977**, 17, 458–462. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Reddehase, M.J.; Mutter, W.; Münch, K.; Bühring, H.J.; Koszinowski, U.H. CD8-positive T lymphocytes specific for murine cytomegalovirus immediate-early antigens mediate protective immunity. J. Virol.
**1987**, 61, 3102–3108. [Google Scholar] [CrossRef] [Green Version] - Riddell, S.R.; Watanabe, K.S.; Goodrich, J.M.; Li, C.R.; Agha, M.E.; Greenberg, P.D. Restoration of viral immunity in immunodeficient humans by the adoptive transfer of T cell clones. Science
**1992**, 257, 238–241. [Google Scholar] [CrossRef] - Neuenhahn, M.; Albrecht, J.; Odendahl, M.; Schlott, F.; Dössinger, G.; Schiemann, M.; Lakshmipathi, S.; Martin, K.; Bunjes, D.; Harsdorf, S.; et al. Transfer of minimally manipulated CMV-specific T cells from stem cell or third-party donors to treat CMV infection after allo-HSCT. Leukemia
**2017**, 31, 2161–2171. [Google Scholar] [CrossRef] - Karrer, U.; Sierro, S.; Wagner, M.; Oxenius, A.; Hengel, H.; Koszinowski, U.H.; Phillips, R.E.; Klenerman, P. Memory inflation: Continuous accumulation of antiviral CD8+ T cells over time. J. Immunol.
**2003**, 170, 2022–2029. [Google Scholar] [CrossRef] [Green Version] - Klenerman, P.; Oxenius, A. T cell responses to cytomegalovirus. Nat. Rev. Immunol.
**2016**, 16, 367–377. [Google Scholar] [CrossRef] - Ouyang, Q.; Wagner, W.M.; Wikby, A.; Walter, S.; Aubert, G.; Dodi, A.I.; Travers, P.; Pawelec, G. Large numbers of dysfunctional CD8+ T lymphocytes bearing receptors for a single dominant CMV epitope in the very old. J. Clin. Immunol.
**2003**, 23, 247–257. [Google Scholar] [CrossRef] - Mekker, A.; Tchang, V.S.; Haeberli, L.; Oxenius, A.; Trkola, A.; Karrer, U. Immune senescence: Relative contributions of age and cytomegalovirus infection. PLoS Pathog.
**2012**, 8, e1002850. [Google Scholar] [CrossRef] [Green Version] - Strandberg, T.E.; Pitkala, K.H.; Tilvis, R.S. Cytomegalovirus antibody level and mortality among community-dwelling older adults with stable cardiovascular disease. JAMA
**2009**, 301, 380–382. [Google Scholar] - Wikby, A.; Johansson, B.; Olsson, J.; Löfgren, S.; Nilsson, B.O.; Ferguson, F. Expansions of peripheral blood CD8 T-lymphocyte subpopulations and an association with cytomegalovirus seropositivity in the elderly: The Swedish NONA immune study. Exp. Gerontol.
**2002**, 37, 445–453. [Google Scholar] [CrossRef] - Olsson, J.; Wikby, A.; Johansson, B.; Löfgren, S.; Nilsson, B.O.; Ferguson, F.G. Age-related change in peripheral blood T-lymphocyte subpopulations and cytomegalovirus infection in the very old: The Swedish longitudinal OCTO immune study. Mech. Ageing Dev.
**2000**, 121, 187–201. [Google Scholar] [CrossRef] - Buchholz, V.R.; Neuenhahn, M.; Busch, D.H. CD8+ T cell differentiation in the aging immune system: Until the last clone standing. Curr. Opin. Immunol.
**2011**, 23, 549–554. [Google Scholar] [CrossRef] - Trautmann, L.; Rimbert, M.; Echasserieau, K.; Saulquin, X.; Neveu, B.; Dechanet, J.; Cerundolo, V.; Bonneville, M. Selection of T Cell Clones Expressing High-Affinity Public TCRs within Human Cytomegalovirus-Specific CD8 T Cell Responses. J. Immunol.
**2005**, 175, 6123–6132. [Google Scholar] [CrossRef] [PubMed] - Day, E.K.; Carmichael, A.J.; ten Berge, I.J.M.; Waller, E.C.P.; Sissons, J.G.P.; Wills, M.R. Rapid CD8+ T cell repertoire focusing and selection of high-affinity clones into memory following primary infection with a persistent human virus: Human cytomegalovirus. J. Immunol.
**2007**, 179, 3203–3213. [Google Scholar] [CrossRef] [Green Version] - Li, H.; Ye, C.; Ji, G.; Han, J. Determinants of public T cell responses. Cell Res.
**2012**, 22, 33–42. [Google Scholar] [CrossRef] - Schober, K.; Buchholz, V.R.; Busch, D.H. TCR repertoire evolution during maintenance of CMV-specific T-cell populations. Immunol. Rev.
**2018**, 283, 113–128. [Google Scholar] [CrossRef] - Zarnitsyna, V.I.; Evavold, B.D.; Schoettle, L.N.; Blattman, J.N.; Antia, R. Estimating the Diversity, Completeness, and Cross-Reactivity of the T Cell Repertoire. Front. Immunol.
**2013**, 4, 1–11. [Google Scholar] [CrossRef] [Green Version] - Elhanati, Y.; Sethna, Z.; Callan, C.G.; Mora, T.; Walczak, A.M. Predicting the spectrum of TCR repertoire sharing with a data-driven model of recombination. Immunol. Rev.
**2018**, 284, 167–179. [Google Scholar] [CrossRef] [Green Version] - Qi, Q.; Liu, Y.; Cheng, Y.; Glanville, J.; Zhang, D.; Lee, J.-Y.; Olshen, R.A.; Weyand, C.M.; Boyd, S.D.; Goronzy, J.J. Diversity and clonal selection in the human T-cell repertoire. Proc. Natl. Acad. Sci. USA
**2014**, 111, 13139–13144. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Venturi, V.; Price, D.A.; Douek, D.C.; Davenport, M.P. The molecular basis for public T-cell responses? Nat. Rev. Immunol.
**2008**, 8, 231–238. [Google Scholar] [CrossRef] [PubMed] - Venturi, V.; Kedzierska, K.; Price, D.A.; Doherty, P.C.; Douek, D.C.; Turner, S.J.; Davenport, M.P. Sharing of T cell receptors in antigen-specific responses is driven by convergent recombination. Proc. Natl. Acad. Sci. USA
**2006**, 103, 18691–18696. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Quigley, M.F.; Greenaway, H.Y.; Venturi, V.; Lindsay, R.; Quinn, K.M.; Seder, R.A.; Douek, D.C.; Davenport, M.P.; Price, D.A. Convergent recombination shapes the clonotypic landscape of the naive T-cell repertoire. Proc. Natl. Acad. Sci. USA
**2010**, 107, 19414–19419. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Smith, C.J.; Venturi, V.; Quigley, M.F.; Turula, H.; Gostick, E.; Ladell, K.; Hill, B.J.; Himelfarb, D.; Quinn, K.M.; Greenaway, H.Y.; et al. Stochastic Expansions Maintain the Clonal Stability of CD8+ T Cell Populations Undergoing Memory Inflation Driven by Murine Cytomegalovirus. J. Immunol.
**2020**, 204, 112–121. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Sethna, Z.; Elhanati, Y.; Callan, C.G.; Walczak, A.M.; Mora, T. OLGA: Fast computation of generation probabilities of B- and T-cell receptor amino acid sequences and motifs. Bioinformatics
**2019**, 35, 2974–2981. [Google Scholar] [CrossRef] [Green Version] - Elhanati, Y.; Murugan, A.; Callan, C.G.; Mora, T.; Walczak, A.M. Quantifying selection in immune receptor repertoires. Proc. Natl. Acad. Sci. USA
**2014**, 111, 9875–9880. [Google Scholar] [CrossRef] [Green Version] - Murugan, A.; Mora, T.; Walczak, A.M.; Callan, C.G. Statistical inference of the generation probability of T-cell receptors from sequence repertoires. Proc. Natl. Acad. Sci. USA
**2012**, 109, 16161–16166. [Google Scholar] [CrossRef] [Green Version] - Schober, K.; Voit, F.; Grassmann, S.; Müller, T.R.; Eggert, J.; Jarosch, S.; Weißbrich, B.; Hoffmann, P.; Borkner, L.; Nio, E.; et al. Reverse TCR repertoire evolution toward dominant low-affinity clones during chronic CMV infection. Nat. Immunol.
**2020**, 21, 434–441. [Google Scholar] [CrossRef] - Dekhtiarenko, I.; Jarvis, M.A.; Ruzsics, Z.; Čičin-Šain, L. The context of gene expression defines the immunodominance hierarchy of cytomegalovirus antigens. J. Immunol.
**2013**, 190, 3399–3409. [Google Scholar] [CrossRef] - Snyder, C.M.; Cho, K.S.; Bonnett, E.L.; van Dommelen, S.; Shellam, G.R.; Hill, A.B. Memory inflation during chronic viral infection is maintained by continuous production of short-lived, functional T cells. Immunity
**2008**, 29, 650–659. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Laydon, D.J.; Bangham, C.R.M.; Asquith, B. Estimating T-cell repertoire diversity: Limitations of classical estimators and a new approach. Phil. Trans. R. Soc. B
**2015**, 370, 20140291. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Lythe, G.; Callard, R.E.; Hoare, R.; Molina-París, C. How many TCR clonotypes does a body maintain? J. Theor. Biol.
**2015**, 389, 214–224. [Google Scholar] [CrossRef] [PubMed] [Green Version] - Buchholz, V.R.; Flossdorf, M.; Hensel, I.; Kretschmer, L.; Weissbrich, B.; Gräf, P.; Verschoor, A.; Schiemann, M.; Höfer, T.; Busch, D.H. Disparate individual fates compose robust CD8+ T cell immunity. Science
**2013**, 340, 630–635. [Google Scholar] [CrossRef] [PubMed] - Warren, R.L.; Freeman, J.D.; Zeng, T.; Choe, G.; Munro, S.; Moore, R.; Webb, J.R.; Holt, R.A. Exhaustive T-cell repertoire sequencing of human peripheral blood samples reveals signatures of antigen selection and a directly measured repertoire size of at least 1 million clonotypes. Genome Res.
**2011**, 21, 790–797. [Google Scholar] [CrossRef] [Green Version] - Robins, H.S.; Campregher, P.V.; Srivastava, S.K.; Wacher, A.; Turtle, C.J.; Kahsai, O.; Riddell, S.R.; Warren, E.H.; Carlson, C.S. Comprehensive assessment of T-cell receptor beta-chain diversity in alphabeta T cells. Blood
**2009**, 114, 4099–4107. [Google Scholar] [CrossRef] - Obar, J.J.; Khanna, K.M.; Lefrançois, L. Endogenous Naive CD8+ T Cell Precursor Frequency Regulates Primary and Memory Responses to Infection. Immunity
**2008**, 28, 859–869. [Google Scholar] [CrossRef] [Green Version] - Van Heijst, J.W.J.; Gerlach, C.; Swart, E.; Sie, D.; Nunes-Alves, C.; Kerkhoven, R.M.; Arens, R.; Correia-Neves, M.; Schepers, K.; Schumacher, T.N.M. Recruitment of antigen-specific CD8+ T cells in response to infection is markedly efficient. Science
**2009**, 325, 1265–1269. [Google Scholar] [CrossRef] [Green Version] - Hansen, S.G.; Sacha, J.B.; Hughes, C.M.; Ford, J.C.; Burwitz, B.J.; Scholz, I.; Gilbride, R.M.; Lewis, M.S.; Gilliam, A.N.; Ventura, A.B.; et al. Cytomegalovirus vectors violate CD8+ T cell epitope recognition paradigms. Science
**2013**, 340, 1237874. [Google Scholar] [CrossRef] [Green Version]

**Figure 1.**TCR repertoire evolution of H2k

^{b}/SIINFEKL-specific T cells during CMV infection. (

**a**) Experimental setup; (

**b**) evenness of TCR repertoires based on CDR3β sequences (analogous results were obtained for CDR3α). Higher values indicate homogenously distributed repertoires whereas lower values indicate skewed repertoires. Shown is the mean +/− SEM for n = 5–7 mice per treatment (thymectomy vs. no thymectomy). Representative of two independent experiments. Statistical testing by mixed-effects model (REML) analysis for fixed effects (type Ⅲ), * p value < 0.05, ** p value < 0.01; (

**c**) cumulative unique CDR3β sequences of TCR repertoires per mouse (analogous results were obtained for CDR3α). Shown is the mean +/− SEM for n = 5–7 mice per treatment. Representative of two independent experiments. Statistical testing by two-way repeated measures (RM) ANOVA analysis. * p value < 0.05, **** p value < 0.0001; (

**d**) TCR repertoire evolution in two representative mice (both representative mice are from the thymectomized groups, but show overall repertoire evolution patterns that are representative of thymectomized as well as non-thymectomized mice). Different colors denote unique CDR3β sequences (analogous results were obtained for CDR3α).

**Figure 2.**Cumulative absolute size of H2k

^{b}/SIINFEKL-specific TCR repertoires (

**a**) Share of public clonotypes; (

**b**) unique CDR3 sequences for different degrees of sharedness (maximum degree of sharedness 12/12); (

**c**) cumulative unique CDR3β sequences vs. cumulative mouse repertoires; orange denotes experimental data including public clones; gray denotes simulated outcomes if all public clones are treated as private clones; (

**d**) function of fitted data from (

**c**); black lines surrounding orange line indicates 95% confidence interval; Kd denotes number of mice at which half of the maximum cumulative CDR3 sequences would be observed; (

**e**) as in (

**c**), but for CDR3α; (

**f)**as in (

**d**), but for CDR3α. (

**g**) Bmax (number of unique CDR3 sequences when X equals an infinite number of cumulative mouse repertoires) for CDR3β and CDR3α in dependence of clone fraction filters. All clones had a filter of UMI > 1. (

**h**) as in (

**g**), but for Kd (number of mice to reach 50% of Bmax). All data come from non-thymectomized animals.

**Figure 3.**Clone fractions in dependence of degree of sharedness. (

**a**) Clone fraction of CDR3β sequences for different degrees of sharedness (maximum degree of sharedness 12/12; data from all mice and time points pooled); left: log scale; right: linear scale. (

**b**) fraction of unique CDR3β sequences (left column) or sum of clone fractions (middle and right columns) for all mice (left and middle column) or one representative mouse (right column) vs. the degree of sharedness for day 15 (top) or day 120 (bottom) after infection; sum of clone fraction means all clone fractions with a specific degree of sharedness were added up; each dot indicates data for one degree of sharedness in one mouse; (

**c**) mean degree of sharedness over time post infection; each dot represents one mouse repertoire for a given time point; all data come from non-thymectomized animals.

**Figure 4.**Association of generation probabilities with degree of sharedness and clone fraction. (

**a**) Generation probabilities of CDR3β sequences for different degrees of sharedness (maximum degree of sharedness 12/12; data from all mice and time points pooled); clonotypes for which no generation probability could be calculated were not included in this depiction; medians are shown in red. (

**b**) medians from (

**a**) without (black) or with (gray) clonotypes for which no generation probability could be calculated (gen prob = “0”); (

**c**) as in (

**a**), but over time post infection, statistical testing by Kruskal-Wallis test (****) followed by Dunn’s multiple comparisons test (shown are results for comparisons of each day from day 86 onwards tested against each day until day 58); **** p value < 0.0001; (

**d**) as in (

**a**), but against clone fraction at a given time point in a given mouse; (

**e**) mean generation probability over time post infection; each dot represents one mouse repertoire for a given time point; (

**f**) generation probabilities for dominators (i.e., clones with a clone fraction >0.1) at day 15 and day 120 post infection, in comparison to all clones; medians in red, boxes indicate 95% confidence interval; statistical testing by Kruskal-Wallis test (****) followed by Dunn’s multiple comparisons test; ns non-significant, * p value < 0.05, **** p value < 0.0001; (

**g**) Clone fraction of clones with specified filters from all mice and time points pooled; (

**h**) data from (

**g**), but percentage of clones with a clone fraction above the indicated values (below the bars) depending on the filters (shown on the right); all data come from non-thymectomized animals.

**Figure 5.**Association of nucleotide-to-amino acid ratio of TCR sequences to generation probability and degree of sharedness. (

**a**) Generation probability of CDR3β sequences (all time points and mice pooled) vs. the nucleotide-to-amino acid (nt:aa) ratio found in individual mice; (

**b**) as in (

**a**), but for degree of sharedness and with boxes indicating 95% confidence interval and individual data points displayed outside this interval; (

**c**) as in (

**a**,

**b**), but with all data in one graph; (

**d**) as in (

**a**), but against clone fraction; all data come from non-thymectomized animals; red line indicates median; statistical testing for all data by Kruskal-Wallis test (****) followed by Dunn’s multiple comparisons test; ** p value < 0.01, **** p value < 0.0001.

© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## Share and Cite

**MDPI and ACS Style**

Schober, K.; Fuchs, P.; Mir, J.; Hammel, M.; Fanchi, L.; Flossdorf, M.; Busch, D.H.
The CMV-Specific CD8^{+} T Cell Response Is Dominated by Supra-Public Clonotypes with High Generation Probabilities. *Pathogens* **2020**, *9*, 650.
https://doi.org/10.3390/pathogens9080650

**AMA Style**

Schober K, Fuchs P, Mir J, Hammel M, Fanchi L, Flossdorf M, Busch DH.
The CMV-Specific CD8^{+} T Cell Response Is Dominated by Supra-Public Clonotypes with High Generation Probabilities. *Pathogens*. 2020; 9(8):650.
https://doi.org/10.3390/pathogens9080650

**Chicago/Turabian Style**

Schober, Kilian, Pim Fuchs, Jonas Mir, Monika Hammel, Lorenzo Fanchi, Michael Flossdorf, and Dirk H. Busch.
2020. "The CMV-Specific CD8^{+} T Cell Response Is Dominated by Supra-Public Clonotypes with High Generation Probabilities" *Pathogens* 9, no. 8: 650.
https://doi.org/10.3390/pathogens9080650