1 Modulation of let-7 miRNAs controls the differentiation of effector CD8 T cells
2 3 4 Alexandria C. Wells1, Keith A. Daniels2, Constance C. Angelou1, Eric Fagerberg1, Amy 5 S. Burnside1, Michele Markstein3, Dominique Alfandari1, Raymond M. Welsh2, Elena L. 6 Pobezinskaya1*, Leonid A. Pobezinsky1* 7 8 1Department of Veterinary and Animal Sciences, University of Massachusetts, Amherst,
9 MA 01003
10 2Department of Pathology, University of Massachusetts Medical School, MA 01605
11 3Department of Biology, University of Massachusetts, Amherst, MA 01003. 12 13 14 Running title: let-7 regulates CTL-mediated immune responses 15 16 Keywords: CTL; naive; Myc; Eomes; TCR; Metabolism 17 18 *Correspondence to: 19 Leonid A. Pobezinsky, Department of Veterinary and Animal Sciences, University
20 of Massachusetts, Amherst, MA 01003
21 Tel: 413-545-2393
22 Fax: 413-545-6326
23 Email: [email protected]
24 Elena L. Pobezinskaya, Department of Veterinary and Animal Sciences, 25 University of Massachusetts, Amherst, MA 01003 26 Tel: 413-545-2405 27 Email: [email protected]
28 29 Abstract
30 The differentiation of naïve CD8 T cells into effector cytotoxic T lymphocytes upon
31 antigen stimulation is necessary for successful anti-viral, and anti-tumor immune
32 responses. Here, using a mouse model, we describe a dual role for the let-7 microRNAs
33 in the regulation of CD8 T cell responses, where maintenance of the naïve phenotype in
34 CD8 T cells requires high levels of let-7 expression, while generation of cytotoxic T
35 lymphocytes depends upon T cell receptor mediated let-7 downregulation. Decrease of
36 let-7 expression in activated T cells enhances clonal expansion and the acquisition of
37 effector function through derepression of the let-7 targets, including Myc and
38 Eomesodermin. Ultimately, we have identified a novel let-7 mediated mechanism, which
39 acts as a molecular brake controlling the magnitude of CD8 T cell responses.
40 41
42 Introduction 43 44 CD8+ T cells are responsible for the rapid clearance of virally infected, and
45 cancerous cells in the organism. Prior to encounter with antigen, naïve CD8 T cells
46 exhibit a quiescent state that is characterized by very low rates of proliferation, and a
47 “quiet” transcriptional landscape with no expression of any effector molecules. After
48 antigen recognition, activated CD8 T cells undergo blastogenesis and rapid clonal
49 expansion, which is followed by differentiation into effector cytotoxic T lymphocytes
50 (CTLs), and memory T cells that are both capable of producing effector cytokines and
51 killing target cells. These changes in T cells are accompanied by metabolic
52 reprograming from oxidative phosphorylation to aerobic glycolysis that provides energy
53 and larger amounts of the biomacromolecular intermediates needed to support growth 1.
54 The productive differentiation of CD8 T cells requires three definitive signals; one
55 is from the T cell receptor (TCR) during antigen recognition, the second is from co-
56 stimulation, and the third is due to cytokines. It has been extensively studied how these
57 signals execute the CD8 T cell differentiation program by activation of a complex
58 network of transcription factors such as Myc, Notch-1, Tbet, Eomes, Blimp-1, Zbtb32,
59 Runx3, and Zeb2 2-5. However, the global regulation of these processes is not yet fully
60 understood.
61 Post-transcriptional regulatory mechanisms provide broad regulation of gene
62 expression. The most efficient post-transcriptional regulatory machinery involves RNA
63 interference mediated by microRNAs (miRNAs), small non-coding RNAs that target
64 mRNA in a sequence-specific manner to prevent protein synthesis, and the expression
65 of which is both inducible, and tissue- specific 6,7. The importance of this type of
1 66 regulation for T cells was first demonstrated through the ectopic expression of various
67 miRNAs during hematopoiesis 8. Later, it was found that miRNA depletion achieved by
68 a T cell- specific deletion of Dicer, a critical enzyme involved in the biogenesis of mature
69 miRNAs, resulted in severe defects, including aberrant proliferation, differentiation, and
70 function of T lymphocytes 9-11. Further research corroborated these earlier studies,
71 identifying the role of specific miRNAs in these processes. MiR-181 was shown to
72 modulate TCR sensitivity and signal strength in developing and mature T cells 12,13. In
73 CD4 T cells, miR-29 was shown to control the T helper response through the direct
74 targeting of IFN-γ mRNA 14, as well as the transcription factors T-bet and Eomes 15. In
75 CD8 T cells, the upregulation of miR-130/301 16 and the miR-17-92 cluster 15,17,18 was
76 demonstrated to be essential for the initiation of differentiation upon antigen stimulation.
77 Let-7 miRNAs are the largest and most abundant family of miRNAs in
78 lymphocytes, yet very little is know about their function. It has been shown that let-7
79 miRNAs are needed for the differentiation and function of natural killer T cells, an
80 innate-like subset of T cells 19,20. It has also been suggested that let-7 miRNAs may
81 regulate T helper responses 21,22, and play a role in the suppressive function of Tregs 23.
82 However, the extent to which let-7 miRNAs regulate the differentiation or function of
83 CD8 T cells has yet to be explored.
84 In the present study, we examine the role of let-7 miRNA expression in naïve and
85 differentiating CD8 T cells. We found that in CD8 T cells high levels of let-7 miRNAs are
86 necessary to maintain the naïve phenotype, while TCR-mediated down-regulation of let-
87 7 levels in activated cells is critical for the effective differentiation and function of CTLs.
88 In fact, experimentally forced let-7 expression severely impairs the proliferation and
2 89 differentiation of CD8 T cells, while let-7 deficiency significantly enhances the cytotoxic
90 function of CTLs, and consequently immune responses in vivo. Given these findings, we
91 propose a model in which let-7 acts as a molecular hub by converting the strength of
92 TCR signaling into the strength of CD8 T cell function.
93
94 Results 95 96 let-7 miRNA expression maintains the quiescent state in naïve CD8 T cells 97 98 To explore the potential regulatory role of let-7 miRNAs in CD8 T cells, the
99 expression levels of let-7 miRNA family members in naïve and activated CD8 T cells
100 were determined and normalized to two different house keeping RNAs. Surprisingly, the
101 initially very high expression of let-7 miRNAs in naïve CD8 T cells was reduced by TCR
102 signaling, and this downregulation was proportional to the strength and duration of TCR-
103 stimulation, regardless of house keeping RNA used (Figure 1A, B and Figure 1—figure
104 supplement 1 and Figure 1—figure supplement 2A). As a specificity control, we
105 confirmed the upregulation of miR-17 from the miR-17-92 locus (Figure 1—figure
106 supplement 2B) that is induced upon T cell activation17,18. Taken together, these results
107 suggest that TCR-mediated signaling inhibits the expression of let-7 miRNAs during T
108 cell activation.
109 To determine the functional significance of let-7 expression in naïve CD8
110 lymphocytes, we examined CD8 T cells from P14+Lin28TgRag2-/- mice where T cell-
111 specific expression of the Lin28 protein blocks let-7 biogenesis19,24, inhibiting let-7
112 expression, and P14 is a monoclonal T cell receptor specific to the lymphocytic
113 choriomeningitis virus (LCMV) peptide gp33-41, presented in the context of H-2Db
3 114 molecules (Figure 1—figure supplement 3A). In comparison to the P14+ wild type
115 counterparts, P14+Lin28Tg CD8 T cells were significantly larger in size, with a
116 dramatically increased proportion of Ki67 positive cells 25 (Figure 1C, D). Consistent
117 with the increased proportion of Ki67 positive cells, P14+Lin28Tg mice had a
118 significantly higher frequency of BrdU positive cells in both the spleen and lymph nodes,
119 as compared to wild type P14+ mice (Figure 1E). In addition, surface expression of T
120 cell activation markers, such as the IL-2 receptor beta-chain (CD122), and CD44 was
121 also increased, while cells remained CD25 negative (Figure 1—figure supplement 3B).
122 Thus, these results suggest that the expression of let-7 miRNAs may maintain the
123 quiescent state in naïve CD8 T cells.
124
125 let-7 miRNA expression in CTLs affects both the anti-viral and anti-tumor immune
126 responses
127 TCR stimulation of naive T cells leads to a rapid loss of the quiescent state and
128 differentiation into effector cells. Given that the expression of let-7 miRNAs, which is
129 critical for the maintenance of naïve CD8 T cells, is inhibited by TCR signaling (Figure
130 1A, B), we hypothesized that the downregulation of let-7 miRNAs in response to TCR
131 stimulation is necessary for the differentiation and function of effector CD8 T cells
132 (Figure 2A).
133 To determine whether TCR-mediated downregulation of let-7 miRNAs is required
134 for CD8 T cell differentiation in vivo, we analyzed the fate of P14+ CD8 T cells with
135 forced let-7 expression in response to acute viral infection with LCMV Armstrong. The
136 doxycycline- inducible let-7g transgene 26 maintains let-7g miRNA expression in
4 137 lymphocytes in the presence of doxycycline, even after TCR stimulation (Figure 2—
138 Figure supplement 1A). Donor CD45.2+CD8+ T cells from P14+ and P14+ let-7
139 transgenic (let-7Tg) mice were adoptively transferred into host congenic wild type
140 CD45.1+ mice that were concurrently infected with LCMV, and the differentiation state of
141 P14+ cells was assessed seven days post- injection. Interestingly, the recovery of donor
142 CD8 T cells at the peak of the immune response revealed that P14+let-7Tg CD8 T cells
143 failed to clonally expand (Figure 2B) and lacked KLRG1 expression, an established
144 marker of terminal effector CTLs 27-30 (Figure 2C). Furthermore, let-7Tg CTLs had a
145 reduced frequency of IFN-γ+TNF-α+ cytokine double-producing cells, a hallmark of an
146 effective CD8 T cell response 31-33, while the differentiation of endogenous host-derived
147 CTLs was normal, suggesting a cell-intrinsic mechanism (Figure 2C). Importantly,
148 mRNAs of the Klrg1, Ifng and Tnfa genes are not targets of let-7 miRNAs, therefore the
149 reduced frequencies of effector cells generated from let-7Tg CD8 T cells is not simply a
150 result of direct suppression of effector molecule expression. Thus, sustained let-7
151 expression following TCR activation severely impaired the clonal expansion and
152 differentiation of CTLs in response to viral infection in vivo.
153 As the allogeneic response to foreign MHC is one of the most robust responses
154 of the immune system, we further aimed to determine whether steady levels of let-7 in T
155 cells would suppress the allogeneic response in vivo 34,35. We used the P815
156 mastocytoma, an allogeneic (H-2d haplotype) tumor, that has been shown to elicit a
157 CD8 T cell-mediated allogeneic immune response 36. We confirmed this by
158 demonstrating that P14+Rag2-/- mice failed to reject the tumor without adoptive transfer
159 of CD8 T cells (Figure 2—figure supplement 1B). Interestingly, when the P815 cells
5 160 were injected into wild type or let-7Tg mice of H-2b haplotype, 60% of let-7Tg mice were
161 unable to effectively respond to, and clear the tumor (Figure 2D). Additionally, at seven
162 days post-injection, wild type mice retained on average 40 x106 cancer cells in the
163 peritoneal cavity, while let-7Tg mice contained 115 x106 cancer cells (Figure 2E). We
164 concluded that CD8 T cells that maintained let-7 expression upon stimulation and
165 differentiation had a compromised response to the alloantigen, and thus failed to reject
166 P815 tumor cells. Taken together, our results demonstrate that the decrease in let-7
167 expression upon TCR activation is necessary for the proper proliferation and
168 differentiation of cytotoxic CD8 T cells in vivo.
169
170 Expression of let-7 miRNAs in activated CD8 T cells inhibits proliferation and the gene
171 expression program responsible for the metabolic switch
172 To elucidate the underlying mechanisms of let-7 mediated suppression of CD8 T
173 cell responses (Figure 2), we first analyzed the impact of let-7 microRNAs on T cell
174 clonal expansion. Sorted naïve (CD44loCD25-) CD8 T cells with different levels of let-7
175 expression (Figure 3—figure supplement 1A, B) were activated with anti-CD3 mAbs in
176 vitro for 3 days. Interestingly, let-7Tg CD8 T cells proliferated less than their wild type
177 counterparts, while Lin28Tg CD8 T cells exhibited enhanced proliferation (Figure 3A).
178 These results indicate that let-7 miRNAs negatively regulate clonal expansion of
179 activated CD8 T cells, which is consistent with our previous results in vivo (Figure 2B).
180 Thus, we concluded that TCR-mediated downregulation of let-7 expression is needed
181 for successful proliferation of activated CD8 T cells.
6 182 Let-7 miRNAs are well-documented tumor suppressors. It has been shown that
183 let-7 inhibits proliferation in cancerous cells by directly targeting the mRNA of genes that
184 are involved in the regulation of the cell cycle 37. In fact, the expression of some let-7
185 targets such as phosphatase cdc25a (cdc25a), kinase cdk6 (cdk6), and cyclin D2
186 (ccnd2), was suppressed in activated let-7Tg CD8 T cells, as compared to Lin28Tg CD8
187 T cells where it was derepressed (Figure 3B). It has been shown that let-7 may also
188 regulate the transcription factor Myc, expression of which is upregulated upon T cell
189 activation and is essential for CD8 T cell proliferation 3,38-41. Interestingly, activated let-
190 7Tg CD8 T cells had reduced expression of Myc, whereas let-7- deficient (Lin28Tg)
191 CD8 T cells had increased Myc expression (Figure 3C). To demonstrate that Myc
192 activity was suppressed by let-7, we also analyzed the expression of a direct
193 transcriptional target of Myc, the transcription factor AP4 (Tfap4), which sustains Myc-
194 mediated effects in CD8 T cells during the later stages of differentiation 42. Although
195 mRNA of the Tfap4 gene is not a target of let-7, the expression of Tfap4 mRNA was
196 significantly reduced in let-7Tg CD8 T cells, and enhanced in Lin28Tg CD8 T cells
197 (Figure 3C), suggesting that let-7 regulates Myc activity in CD8 T cells.
198 Another important function of Myc in activated CD8 T cells is to support the
199 proliferative burst through the metabolic reprogramming of lymphocytes from primarily
200 oxidative phosphorylation (resting) to glycolysis (activated), as well as through an
201 increase in protein synthesis 43,44. To test whether let-7 miRNAs have an impact on the
202 metabolic switch in activated CD8 T cells through its regulation of Myc, we assessed the
203 expression of key glucose transporters, glycolytic enzymes, and protein synthesis
204 enzymes that have been established as direct targets of Myc in activated CD8 T cells 44.
7 205 Strikingly, the expression of all tested targets was suppressed in let-7 transgenic CD8 T
206 cells, and increased in Lin28Tg activated lymphocytes, suggesting that let-7 expression
207 may influence Myc-dependent metabolic reprogramming of activated CD8 T cells
208 (Figure 3D). Taken together, these results demonstrate that let-7 miRNAs control the
209 proliferation of activated CD8 T cells by modulating the expression and activity of genes
210 that are involved in the regulation of the cell cycle and metabolism.
211
212 let-7 expression regulates the function of differentiated CD8 T cells
213 To identify whether let-7 mediated suppression of CD8 T cell immune responses
214 is due to a failure to acquire effector function in addition to a proliferative defect, we
215 assessed the cytotoxic activity of CTLs generated from P14+ wild type, P14+let-7Tg, and
216 P14+Lin28Tg mice. In fact, the expression of the let-7 transgene in P14+ CTLs greatly
217 diminished their cytotoxic activity (Figure 4A). Alternatively, P14+Lin28Tg CTLs
218 exhibited enhanced cytotoxicity (Figure 4A), which could be reduced by restoring let-7
219 expression through the induction of the doxycycline-inducible let-7 transgene in
220 P14+Let-7TgLin28Tg (4 Tg) CTLs (Figure 4B), demonstrating that let-7- deficiency, and
221 not Lin28 overexpression, is responsible for increased cytotoxicity. Thus, we
222 demonstrated that TCR-mediated downregulation of let-7 microRNAs is necessary for
223 the acquisition of cytotoxic function in differentiating CD8 T cells.
224 To investigate the mechanism of how let-7 miRNA expression impacts CD8 T cell
225 function, we examined the phenotype of in vitro generated effector CTLs from wild type,
226 let-7Tg, and Lin28Tg mice. Let-7Tg CTLs had less internal complexity based on the
227 intensity of side scatter (SSC) than wild type cells, whereas Lin28Tg CD8 effector cells
8 228 had significantly greater complexity (Figure 4C), suggesting a change in the number of
229 cytotoxic granules. Indeed, let-7Tg CTLs contained fewer Granzyme A and Granzyme B
230 positive granules than wild type CTLs, while Lin28Tg CTLs had more (Figure 4D).
231 These results indicated that the expression of let-7 controls the quantity of cytotoxic
232 granules produced during the differentiation of CTLs.
233 Next, to determine whether let-7 expression in CTLs influences the number of
234 granules by controlling the expression of effector molecules, gene expression of
235 Granzyme A (Gzma), Granzyme B (Gzmb), and Perforin (Prf1), the key cytolytic factors
236 in cytotoxic granules 45,46, was measured. Let-7Tg CTLs, which contained fewer
237 cytotoxic granules, expressed less mRNA for Gzma, Gzmb, and Prf1, as compared to
238 wild type cells, while Lin28Tg CTLs had higher expression of these effector molecules
239 (Figure 4E). Moreover, the observed changes in mRNA expression of effector
240 molecules, including Interferon-gamma (IFN-γ), were consistent at the protein level
241 (Figure 4F,G). Importantly, the induction of let-7 expression in the presence of Lin28 in
242 P14+Let-7TgLin28Tg (4 Tg) CTLs reduced the expression of these effector molecules
243 (Figure 4—figure supplement 1), again demonstrating that enhanced effector molecule
244 expression is due to let-7- deficiency. Together, these data indicate that let-7 miRNAs
245 negatively regulate CTL function by preventing the expression of important cytotoxic
246 molecules.
247
248 let-7 miRNAs directly target the ‘master regulator’ transcription factor, Eomesodermin
249 during CTL differentiation
9 250 To determine how let-7 regulates the differentiation and function of CTLs, we
251 considered the possibility that let-7 miRNAs may directly regulate the expression of
252 effector molecules. Although typical miRNA binding sites are found in the 3’
253 untranslated regions (UTRs) of mRNA transcripts 47-49 we analyzed the full length
254 mRNA of Prf1, Gzma, Gzmb, and Ifng, yet no let-7 binding sites were found. This
255 suggested that let-7 may indirectly regulate the expression of these molecules by
256 controlling more global regulators, such as transcription factors. It is known that the
257 expression of effector molecules and cytotoxic function of CD8 T cells are tightly
258 regulated by a group of transcription factors, including Eomesodermin (Eomes), T-bet,
259 Zbtb32, Runx3, Notch-1 and Blimp-1 2,4,50-54. To determine if these factors are regulated
260 by let-7, relative gene expression in CTLs with different levels of let-7 was assessed for
261 Eomes, Tbx21, Zbtb32, Prdm1, Runx3d, and Notch-1 genes. The only transcription
262 factor whose expression on both the mRNA and protein levels was reduced in let-7
263 transgenic CTLs, and enhanced in Lin28Tg CTLs, was Eomes (Figure 5A, B and Figure
264 5—figure supplement 1A, B). Furthermore, the induction of let-7 expression in the
265 presence of Lin28 in P14+Let-7TgLin28Tg (4 Tg) CTLs reduced the expression of
266 Eomes (Figure 5C), demonstrating that the Lin-28-mediated knockdown of let-7
267 miRNAs results in increased expression of Eomes in Lin28Tg CTLs. Of note, we also
268 found that T-bet expression was inversely correlated with Eomes, suggesting that an
269 Eomes dependent mechanism may control T-bet expression (Figure 5A, B).Thus, these
270 results demonstrate that let-7 miRNAs negatively regulate Eomes expression.
271 Next, we wanted to determine whether let-7 miRNAs can target Eomes mRNA.
272 Interestingly, there is no let-7 binding site located in the 3’UTR of Eomes, but rather a
10 273 conserved binding motif within the open reading frame of Eomes mRNA was identified
274 (Figure 5D; Supplementary file 1). To determine whether this binding site is functional,
275 the 10-nucleotide mouse Eomes-let-7 binding motif was cloned into a dual luciferase
276 vector. The vector was then transfected into NIH 3T3 fibroblasts, which have high
277 endogenous expression of the let-7 family members (Figure 5—figure supplement 2A,
278 B). This resulted in a significant decrease in luciferase activity, indicating that let-7 can
279 directly bind Eomes (Figure 5E). When this sequence was mutated, disrupting the let-7
280 binding site, luciferase activity was restored, confirming that let-7 microRNAs directly
281 target Eomes mRNA (Figure 5E).
282
283 let-7 miRNAs suppress CD8 T cell function through targeting Eomes mRNA
284 To ascertain whether let-7 controlled CD8 T cell differentiation and function
285 through Eomes, we used a loss-of-function approach to knockout Eomes expression in
286 Lin28Tg CTLs, expecting the reduction of enhanced cytotoxicity in let-7-deficient cells.
287 T-cell specific deletion of either one or both alleles of Eomes in CTLs generated from
288 P14+CD4Cre+Eomesfl/wt Lin28Tg, and P14+CD4Cre+Eomesfl/flLin28Tg T cells resulted in
289 gradually decreased levels of Eomes (Figure 6A). Any residual Eomes expression in
290 CTLs derived from CD4Cre+Eomesfl/flLin28Tg mice, was attributed to “escapees” of the
291 conditional knockout. The expression of the effector molecules Gzma, and Prf1 was
292 reduced in a manner consistent with the expression of Eomes, where loss of Eomes
293 ameliorated the enhanced expression of these effector molecules in Lin28Tg effector
294 cells (Figure 6B, C). Interestingly, the expression of Granzyme B was not affected by
295 loss of Eomes expression, suggesting Granzyme B may be more complexly regulated
11 296 through other let-7- dependent, but Eomes- independent mechanisms (Figure 6—figure
297 supplement 1). Ultimately, the loss of Eomes expression resulted in a significant
298 reduction of cytotoxic function, even in the absence of let-7 microRNAs (Figure 6D)
299 while restored Eomes expression completely rescued both the cytotoxic function and
300 the expression of effector molecules in Eomes-deficient Lin28Tg CTLs (Figure 6—figure
301 supplement 2A,B). Furthermore, forced expression of the mutated form of Eomes
302 (lacking the let-7 binding motif) in let-7Tg CTLs also enhanced the effector phenotype
303 (Figure 6—figure supplement 2C,D). These results demonstrate that let-7-mediated
304 suppression of Eomes is in part responsible for the compromised differentiation and
305 cytotoxic function of let-7Tg CTLs in vitro.
306 Thus, our data suggest a model in which the let-7 miRNAs act as a molecular
307 control hub that drives CD8 T cell differentiation and function, in a manner dependent
308 on the magnitude of TCR stimulation (Figure 6—figure supplement 3). Furthermore, we
309 propose that modulation of let-7 miRNA expression in CD8 T cells provides an exciting,
310 novel therapeutic application to control CTL responses.
311
312 Discussion 313 314 Our study has identified a critical role for the let-7 miRNAs in regulating the
315 transition between naïve and effector stages of CD8 T cells. Specifically, let-7
316 expression is high in naïve T cells and, when absent, results in increased proliferation
317 and expression of activation markers, which may suggest a loss of the quiescent
318 phenotype in CD8 lymphocytes. Moreover, the entire family of let-7 miRNAs is
319 downregulated upon antigen stimulation through the TCR, which we demonstrated is
12 320 necessary for the successful progression of CD8 T cell differentiation into CTLs.
321 Therefore, our data suggest that let-7 expression keeps CD8 T cells in a naïve state
322 and prevents CTL differentiation, while the magnitude of TCR-mediated downregulation
323 of let-7 expression guides the proliferation, differentiation and the acquisition of effector
324 function of CD8 T cells.
325 The importance of miRNAs in the regulation of the immune system has been
326 demonstrated through the deletion of the miRNA processing enzyme, Dicer. In fact, it
327 has been shown that Dicer deficiency promotes the differentiation of CD8 T cells into
328 CTLs 55, suggesting the involvement of specific miRNAs.
329 Let-7 microRNAs, the largest and most abundantly expressed family of miRNAs
330 in CD8 T cells, have been shown to be important in early development, metabolism, and
331 cancer 26,56-59. Recent studies have also implicated the importance of the let-7 miRNAs
332 in the development, maintenance, and function of the immune system, including T cells
333 19,20,23. Using both gain-of-function experiments employing the let-7Tg mouse capable of
334 sustaining let-7 expression following TCR activation, and loss-of-function experiments
335 using the Lin28Tg mouse, where the Lin28Tg inhibits let-7 expression, our study has
336 corroborated these earlier reports, and has identified a novel role for let-7 in CD8 T cell
337 differentiation and function.
338 Prior to antigen encounter, naïve CD8 T cells are maintained in a quiescent
339 state, in which T cells have no effector function, are metabolically inactive and undergo
340 minimal proliferation. It has been shown that the homeostasis of naïve T cells depends
341 on the balance of two signals, the recognition of low affinity self-ligands by the TCR, and
342 cytokine stimulation 60-64. Although it has become clear that the weak recognition of self-
13 343 ligands is not enough for naïve T cells to lose the quiescent state, the molecular
344 mechanism that prevents the spontaneous activation of T cells under these conditions is
345 not fully understood. Here, we have demonstrated that the high expression of let-7
346 microRNAs is necessary to maintain CD8 T cells in a naïve state. In fact, we observed
347 that let-7 ablation in CD8 T cells leads to the loss of the quiescent phenotype, as
348 indicated by more active proliferation, an overall increase of cell size, and the
349 upregulation of activation markers such as CD122, and CD44 25,65. These experiments
350 were conducted using CD8 T cells from P14+Lin28TgRag2-/- mice in order to prevent
351 bystander effects from IL-4- producing PLZF+ cells present in Lin28Tg mice 19,20,66.
352 Although further investigation is needed to distinguish between the let-7-dependent and
353 let-7-independent effects of the Lin28 transgene in naïve CD8 T cells, we can speculate
354 that high expression of let-7 microRNAs prevents the spontaneous activation of naïve T
355 cells. Furthermore, it will be interesting to explore the precise let-7-mediated regulation
356 of transcription programs, such as Foxp1 67, that may contribute to the control of CD8 T
357 cell quiescence.
358 Furthermore, we have found that the let-7 mediated “molecular brake” is released
359 upon antigen stimulation, due to the profound downregulation of all members of the let-7
360 microRNA family in activated CD8 T cells in response to TCR stimulation, in a manner
361 proportional to its strength. Using in vivo models to assess both anti-viral and anti-tumor
362 immunity, we have demonstrated the importance of let-7 miRNAs in controlling CD8 T
363 cell- mediated immune responses. Let-7Tg CD8 T cells failed to proliferate in response
364 to acute LCMV infection, and the few let-7Tg cells that did respond exhibited a very
365 weak effector phenotype, suggesting that downregulation of let-7 is critical to both the
14 366 proliferative burst of antigen-specific CD8 T cells upon encounter with viral antigen, and
367 the differentiation of these cells in vivo. These results were further bolstered by the
368 failure of let-7Tg mice to reject an allogeneic tumor, the P815 mastocytoma. Altogether,
369 our in vivo studies demonstrate the significance of let-7 downregulation in effector CD8
370 T cells, and suggest a novel level of control of immune responses that can be
371 therapeutically targeted for treatment of infectious diseases, cancer and autoimmunity.
372 Antigen stimulation of T cells results in the increased biosynthesis that is needed
373 to support the clonal expansion of antigen specific lymphocytes, and ultimately the
374 acquisition of effector function. The hallmark of this process is a metabolic switch from
375 oxidative phosphorylation to aerobic glycolysis 68,69, as well as a concomitant increase
376 in cell size, both of which have been reported to be controlled by Myc 42,44,70, the
377 expression of which is rapidly induced upon antigen stimulation of CD8 T cells 33,44,70.
378 Presumably due to the pro-apoptotic activity of Myc 71-73, its expression is transient,
379 eventually receding during the later stages of CD8 T cell differentiation 3,40. In our study,
380 we have shown that let-7 suppresses the expression of Myc on the mRNA level, and
381 consequently modulates the function of Myc in CD8 T cells, based on our assessment
382 of established Myc targets, supporting the previous observation of Myc as a non-
383 canonical target of let-7 miRNAs in cancer cells 39,74. Our results suggest that let-7 likely
384 regulates the metabolic switch in activated CD8 T cells through Myc.
385 We also found that the let-7 miRNAs may directly inhibit the proliferation of
386 activated CD8 T cells by suppressing the expression of the cell cycle regulators
387 Cdc25a, Ccnd2 and Cdk6, all of which are known let-7 targets 37. Since these factors
388 have also been described as transcriptional targets of Myc 75-77 we cannot rule out the
15 389 possibility of more complex regulation where let-7 is not solely responsible for
390 controlling their expression.
391 Moreover, we demonstrated that let-7 mediated suppression of CD8 T cell
392 immune responses is also due to modulation of effector function. We noticed that the
393 internal complexity of CTLs generated in vitro was diminished in the presence of forced
394 let-7 expression, and subsequently determined that this was due to a reduction in the
395 number of cytotoxic granules, as well as in the expression of effector molecules. We
396 thus concluded that let-7 functions as a molecular rheostat that quantifies TCR signaling
397 to direct the CTL response upon antigen stimulation, in a similar fashion to other
398 miRNAs, including miR-181 and the miR-17-92 cluster 12,13,18. These conclusions are
399 supported by a recent study demonstrating that in neonatal mice the residual
400 expression of Lin28B in T lymphocytes skews CD8 T cell differentiation towards an
401 effector phenotype 78.
402 We further wanted to determine the molecular mechanisms through which let-7
403 acts to inhibit CTL differentiation. We found that Eomes is directly regulated by let-7
404 microRNAs, which can in part explain the block in the differentiation of CD8 T cells with
405 forced let-7 expression. Yet, in contrast to previous studies on the differential roles of
406 Eomes and T-bet in governing CD8 T cell function, we have shown that heightened
407 expression of Eomes in effector T cells may be more important for effector function than
408 previously thought 4,65,79. In fact, let-7/ Eomes-double deficient CTLs had reduced
409 antigen-specific cytotoxicity in vitro. However, our data also suggest that Eomes is only
410 a part of this “cytotoxic program”, as deletion of Eomes only reduced cytotoxicity to the
411 levels exhibited by wild type CTLs, and re-expression of Eomes in let-7Tg CTLs only
16 412 marginally increased cytotoxicity. These results indicate that other let-7- dependent, but
413 Eomes- independent mechanisms are at play. More experiments are required to
414 elucidate these Eomes- independent mechanisms, and should be enlightening given the
415 suggested redundancy of Eomes activity. Additionally, we have identified Granzyme A
416 as a probable target of Eomes, while also confirming that Perforin, and IFN-γ
417 expression is controlled by Eomes 4. These results are consistent with the enhanced
418 cytotoxicity exhibited by Lin28Tg CD8 T cells with increased Eomes expression. We
419 also observed previously reported reciprocal expression between T-bet and Eomes
420 65,79,80.
421 These results clearly demonstrate that downregulation of let-7 upon TCR
422 stimulation is a critical process in the determination of the magnitude of the CD8 T cell
423 response in vivo, as let-7 miRNAs inhibit proliferation and differentiation by targeting cell
424 cycle regulators, affecting metabolic reprogramming through the suppression of Myc,
425 and repression of effector function through Eomes dependent and independent
426 mechanisms. Thus, naïve CD8 T cells require let-7 miRNAs to remain quiescent, and
427 only upon antigen stimulation through the TCR can this molecular “brake” be released.
428 Based on these results, we propose that let-7 miRNAs act as a molecular control hub
429 that translates TCR signaling to control CD8 T cell differentiation and function.
430
431 Methods
432 Animals
433 C57BL/6J (CD45.2+ wild type, stock no. 000664), B6.SJL- PtprcaPepcb/ BoyJ (CD45.1+
434 wild type, stock no. 002014), B6(Cg)- Rag2Tm1.1Cgn/J (Rag2-/-, stock no. 008449), B6
17 435 Tg(CD4-cre)1Cwi/BfluJ (CD4-Cre, stock no. 017336), and B6.129S1 (Cg)-
436 Eomestm1.1Bflu/J (Eomesfl/fl, stock no. 017293) were obtained from the Jackson
437 Laboratory. B6.Cg- Col1a1tm3(tetO-Mirlet7g/Mir21)Gqda/J (let-7g, stock no. 023912) and B6.Cg-
438 Gt(ROSA)26 Sortm1(rtTA*M2)Jae/J (M2rtTA, stock no. 006965) were also obtained from the
439 Jackson Laboratory and subsequently crossed to generate let-7Tg mice. Mice with the
440 Lin28B transgene (Lin28Tg) driven under the control of the human CD2 promoter 19,
441 and B6 Tg(TcrLCMV)327Sdz/JDvs/J (P14) mice were a generous gift from Alfred Singer
442 (NCI, NIH). P14+ mice, and wild type C57Bl/6J mice on a Rag2-/- background were
443 crossed to generate wild type P14 mice. let-7Tg mice, and P14+ mice were crossed on
444 a Rag2-/- background to generate P14+ doxycycline- inducible let-7 transgenic mice.
445 Lin28B Tg mice were crossed with P14+ mice on a Rag2-/- background to generate P14+
446 Lin28Tg mice. let-7Tg, Lin28B Tg, P14+ mice were crossed on a Rag2-/- background to
447 generate 4Tg mice. CD4-Cre and Eomesfl/fl mice were crossed to generate mice with a
448 T cell- specific conditional knockout of Eomes. CD4-Cre, Eomesfl/fl Lin28B Tg, P14+
449 mice were crossed to generate Lin28Tg mice with a T cell- specific deletion of Eomes.
450 Littermates or age and sex-matched mice were used as controls. All breedings were
451 maintained at the University of Massachusetts, Amherst. This study was performed in
452 accordance with the recommendations in the Guide for the Care and Use of Laboratory
453 Animals of the National Institutes of Health. All of the animals were handled according
454 to approved institutional animal care and use committee (IACUC) protocols (#2014-
455 0045, 2014-0065, 2015-0035) of the University of Massachusetts.
456
457
18 458 Doxycycline-mediated induction of let-7 transgene expression
459 Experimental mice including control animals (unless specifically stated otherwise) were
460 fed with 2 mg/mL doxycycline in drinking water supplemented with 10 mg/mL sucrose
461 for four days prior to the initiation of experimental procedures to ensure maximal
462 induction of let-7g expression. Doxycycline drinking water was replaced every other
463 day. In vitro, lymphocytes were cultured with 2 μg/mL doxycycline in CTL culture media
464 (see cell sorting and in vitro culture below).
465
466 In vivo BrdU labeling
467 Mice were injected i.p. with 1 mg BrdU in PBS, and subsequently fed with 0.8 mg/mL
468 BrdU in drinking water supplemented with 2% sucrose for four days. BrdU water was
469 kept in the dark to eliminate light-sensitivity effects of BrdU and was replaced daily.
470 Incorporation of BrdU in CD8 T cells from the spleen and lymph nodes were analyzed
471 by flow cytometry (see flow cytometry analysis below).
472
473 Flow cytometry analysis
474 Flow cytometry data were acquired on a BD Fortessa or a MilliPore Amnis
475 ImageStream. The following monoclonal antibodies were used: CD8α (53-6.7,
476 eBioscience, RRID:AB_962672; 5H10, Invitrogen, RRID:AB_10372364), CD8β
477 (YTS156.7.7, BioLegend, RRID:AB_961293), CD4 (RM4-5, BioLegend,
478 RRID:AB_312718), BrdU (3D4, BioLegend, RRID:AB_2564481), CD44 (IM7, BD
479 Pharmingen, RRID:AB_2076224), CD25 (PC61, BioLegend, RRID:AB_312857; PC61.5
480 eBioscience, RRID:AB_465606), CD122 (TM-β1, BioLegend, RRID:AB_940607),
19 481 Granzyme A (3G8.5 BioLegend, RRID:AB_2565309; eBioscience, RRID:AB_2665373),
482 Granzyme B (GB11, BioLegend, RRID:AB_2562195), IFN- γ (XMG1.2, eBioscience,
483 RRID:AB_469503), Eomes (Dan11mag, eBioscience, RRID:AB_1603275), Ki67
484 (SolA15, eBioscience, RRID:AB_11149672), T-bet (O4-46, BD Pharmingen,
485 RRID:AB_10564093), PE-Streptavadin (BioLegend, RRID:AB_2571914).
486 For restimulation in vitro, 2x106 cells were stimulated with phorbol myristate acetate
487 (PMA) and Ionomycin for 4 h at 37°C and Monensin A for 2 h at 37°C. Cells were first
488 stained for surface proteins then fixed, permeablized, and stained for intracellular
489 proteins according to the manufacturer’s instructions (BD Pharmingen, eBio).
6 490 For LCMV studies, 2x10 cells were stimulated for 4 hr at 37°C with 2 μg/ mL of GP33-41
491 or NP396-404 peptides, and 1 μl/ mL GolgiPlug (BD Pharmingen). Cells were first stained
492 for surface proteins then fixed, permeablized, and stained for intracellular proteins
493 according to the manufacturer’s instructions (BD Pharmingen). The following
494 monoclonal antibodies were used: CD8β (YTS156.7.7; BioLegend; RRID:AB_961293),
495 CD45.1 (A20; BD Pharmingen; RRID:AB_1727488), CD45.2 (104; BD Pharmingen;
496 RRID:AB_1727491), KLRG1 (2F1; BD Pharmingen; RRID:AB_2686921), CD44 (IM7;
497 BD Pharmingen; RRID:AB_2076224), TNF-α (MP6-XT22; BD Pharmingen;
498 RRID:AB_395380), IFN-γ (XMG1.2; BD Pharmingen; RRID:AB_2686922). Flow
499 cytometry data were acquired on a BD LSR II. All flow cytometry data was analyzed
500 with FlowJo software (TreeStar; RRID:SCR_008520). MilliPore Amnis ImageStream
501 data was analyzed with IDEAS software (EMD Millipore).
502
503 Western blot analysis
20 504 Cells were collected and lysed in M2 lysis buffer (20mM Tris, pH7.0, 0.5% NP40,
505 250mM NaCl, 3mM EDTA, 3mM EGTA, 2mM DTT, 05mM PMSF, 20mM β-glycerol
506 phosphate, 1mM sodium vanadate, 1μg/mL Leupeptin), then resolved by SDS-PAGE.
507 Blots were probed with anti-Perforin (CB5.4, Abcam; RRID:AB_302236) and anti-actin
508 (AC40, Sigma; RRID:AB_2686923), and visualized using enhanced chemiluminescence
509 (ThermoScientific) with horse-radish peroxidase conjugated anti-rat IgG (712-035-150,
510 Jackson ImmunoResearch; RRID:AB_2340638) or anti-mouse IgG (401215,
511 Calbiochem; RRID_AB:2686924).
512
513 Cell sorting and in vitro culture
514 Lymph nodes were harvested and gently tweezed to remove lymphocytes. CD8 lymph
515 node T cells were enriched for via antibody-mediated depletion of B cells using anti-
516 mouse IgG magnetic beads (BioMag, Qiagen). CD4 T cells were removed via anti-rat
517 IgG magnetic beads (BioMag, Qiagen) following incubation with anti-mouse CD4
518 antibodies conjugated with rat IgG (GK1.5). Lymphocytes were electronically sorted for
519 the further purification of naïve CD8 T cells (CD44lo CD25loCD8+CD4-) (Figure 3—figure
520 supplement 1A).
521 Cells were stimulated either with irradiated splenocytes loaded with anti-CD3 mAbs (10
522 μg/mL), or plate-bound anti-TCRβ mAbs (10 μg/mL) and anti-CD28 mAbs (5 μg/mL),
523 then differentiated for five days in RPMI supplemented with 10% fetal bovine serum, 1%
524 HEPES, 1% sodium pyruvate, 1% penicillin/ streptomycin, 1% L-glutamine, 1% non-
525 essential amino acids, 0.3% β-mercaptoethanol, 100 U/mL IL-2, 100 mg/mL gentamicin,
526 and 2μg/mL doxycycline when necessary.
21 527
528 Eomes site-directed mutagenesis and retroviral transduction
529 Site-directed mutagenesis of the let-7 binding site in the Eomes ORF was performed
530 using the Agilent QuikChange II Site-Directed Muatagenesis kit according to the
531 manufacturer’s protocol.
532 Retrovirus expressing Eomes or EomesMUTANT cDNA with a GFP reporter were
533 produced from the transfection of PlatE cells using Lipofectamine plus (Invitrogen). For
534 retroviral transduction, naïve lymphocytes were stimulated with irradiated splenocytes in
535 the presence of anti-CD3 mAbs (10 μg/mL) for 24 hours, then spin-fected (2000 RPM,
536 90 minutes, 30°C) with virus and polybrene (4 μg/mL). Retrovirally transduced cells
537 were obtained by sorting on the GFP+ population.
538
539 Prediction of miRNA targets
540 Eomes was independently identified in an unbiased search of all ORFs in the mouse
541 and humans genomes, for matches to an extended 9 bp let-7 seed, “TACTACCTC“.
542 This search utilized a hashing algorithm as described in 81 and identified 119 genes in
543 the mouse genome, and 159 genes in the human genome that have one or more
544 matches to the 9 bp let-7 seed in their ORF sequences. Interestingly, humans have
545 three splice variants of Eomes, one of which lacks the exon containing the match to let-
546 7, thus opening the possibility that Eomes may escape let-7 repression in some cells by
547 alternative splicing of the target sequence. This may require further investigation.
548
549 Luciferase assay
22 550 NIH 3T3 cells (ATCC) were transfected with the pmirGLO vector (Promega) containing
551 either the intact let-7 binding motif from Eomes, or a mutated version of this binding
552 motif, or either intact antisense or mutated antisense seed regions of let-7b, let-7g, or
553 let-7i using Lipofectamine and Plus reagent (Invitrogen). Luciferase activity was
554 measured 48 hours later on a POLARstar Omega 96-well plate reader (BMG Labtech),
555 using the Dual- Luciferase Reporter Assay System (Promega).
556
557 CellTrace Violet proliferation assay
558 Electronically sorted naïve CD8 T cells were labeled with CellTrace Violet (Invitrogen)
559 for 15 minutes at 37°C. Cells were stimulated using plate-bound anti-TCRβ mAbs (10
560 μg/mL) and anti-CD28 mAbs (5 μg/mL), cultured for 72 hours, and analyzed by flow
561 cytometry.
562
563 CellTrace Violet cytotoxicity assay
564 P14+ CD8 T cells were stimulated with anti-TCRβ mAbs (10 μg/mL) and anti-CD28
565 mAbs (5 μg/mL) plate-bound antibodies, differentiated into CTLs for 5 days in the
566 presence of IL-2, gentamicin, and 2 μg/mL doxycycline when appropriate. On day 5, live
567 splenocytes were warmed for 10 minutes at 37°C, then labeled with CellTrace Violet
568 (Invitrogen) at two different concentrations (CTVhigh or CTVlow) for 15 minutes at 37°C.
569 CTVlow splenocytes were then loaded with either LCMV gp33-41 peptide (1 μM,
570 GenScript) or LCMV np396-404 peptide (1 μM, GenScript) for 1 hour at 37°C, and are
571 referred to below as “experimental splenocytes”. CTVhigh splenocytes remained peptide-
572 free, and were used as a reference control, referred to below as “control splenocytes”.
23 573 Equal amounts of both experimental and control splenocytes were co-cultured with
574 CTLs at different ratios for four to five hours. Cytotoxicity was assessed by flow
575 cytometry using propidium iodide.
576 Measured as the percent target lysis of live experimental splenocytes loaded with either
577 target (gp33-41) or control (np396-404) peptide from the lymphocytic choriomeningitis
578 virus. The following formula was used to calculate the percent target lysis:
퐴 퐶 579 (1 − ( 푋 )) X 100 퐵 퐷
580 A- frequency of live experimental splenocytes co-cultured with CTLs; B- frequency of
581 live control splenocytes co-cultured with CTLs; C- frequency of live control splenocytes
582 incubated in the absence of CTLs; D- frequency of live experimental splenocytes
583 incubated in the absence of CTLs.
584
585 Lymphocytic choriomeningitis virus infection and T cell adoptive transfer
586 10X103 CD45.2+ P14+ donor cells from the indicated mice were transferred i.v. into
587 CD45.1+ congenic hosts. Mice were inoculated with 5X104 p.f.u. of LCMV Armstrong
588 i.p.. Spleens were harvested and processed 7 days post- infection.
589
590 Isolation of RNA and quantitative PCR
591 RNA was isolated according to the manufacturer’s instructions (QIAGEN miRNeasy),
592 and genomic DNA removed using the DNA-free DNA removal kit (Ambion). mRNA-
593 encoding cDNA was synthesized using the SuperScript III Reverse Transcriptase kit
594 (Invitrogen), while miRNA- encoding cDNA was synthesized using the Taqman
595 MicroRNA Reverse Transcription kit (Applied Biosystems). SYBR Green quantitative
24 596 PCR was performed using the Bioline SensiFAST SYBR Lo-Rox kit and Taqman
597 quantitative PCR was performed using the Bioline SensiFAST Lo-Rox kit. Both SYBR
598 Green and Taqman amplification primers (Integrated DNA Technologies, or Applied
599 Biosystems) are listed in Supplementary File 2.
600
601 Tumor transplantation
602 For P815 (ATCC; RRID:CVCL_2124) survival studies, 30X106 tumor cells were injected
603 i.p. into host mice. For P815 tumor burden studies, 20X106 tumor cells were injected i.p.
604 into host mice. Seven days post-injection, tumor burden was assessed by washing the
605 peritoneum with cold PBS and counting the collected cells. For studies involving let-7Tg
606 mice, all mice were fed with doxycycline for the duration of the study. For studies
607 involving adoptive CD8 T cell transfer, 10X106 naïve CD8 T cells were injected i.v. into
608 P14+Rag2-/- mice, concurrently with i.p. tumor injection. The P815 cell line was
609 authenticated by analyzing H-2Kd expression via flow cytometry.
610
611 Statistical analysis
612 P-values were determined using a two-tailed Student’s t-test, or one-tailed Student’s t-
613 test where indicated.
614
615 Acknowledgements
616 We thank A. Singer, X. Tai, T. Kadakia, F. Van Laethem, M. Y. Kimura, and B. A.
617 Osborne for critical reading of the manuscript; A. Singer for providing reagents, and
618 Lin28Tg and P14 mice; V. Lazarevic for providing the plasmid encoding Eomes-GFP
25 619 and control empty retrovirus; P. Markstein and E. D’Souza for assistance with the
620 computational analysis of predicted miRNA targets; and G. Abbruzzesse for technical
621 help.
622
623 Competing financial interests
624 The authors declare no competing financial interests.
625
626 Figure Legends
627 Figure 1. let-7 expression is necessary and sufficient to maintain the naïve phenotype
628 of CD8 T cells prior to TCR stimulation.
629 (a) Quantitative RT-PCR analysis of individual let-7 miRNAs in naïve CD8 T cells
630 stimulated with plate-bound anti-TCR (μg as indicated) and anti-CD28 (5 μg) for 24
631 hours, presented relative to results obtained for the small nuclear RNA (U6 control) and
632 normalized to the unstimulated (0 μg) sample. (b) Quantitative RT-PCR analysis of
633 individual let-7 miRNAs in naïve CD8 T cells stimulated with plate-bound anti-TCR (5μg)
634 and anti-CD28 (5 μg) over increasing periods of time as indicated, presented relative to
635 results obtained for the small nuclear RNA (U6 control) and normalized to the
636 unstimulated (0h) sample. (c) Size analysis based on FSC (forward scatter) of naïve
637 CD8 T cells from the spleens and lymph nodes of P14+ wild type and P14+Lin28Tg
638 mice, both on Rag2-/- background, normalized to wild type. (d) Expression of Ki67 in
639 naïve CD8 T cells from spleens and lymph nodes of P14+ wild type and P14+Lin28Tg
640 mice, both on Rag2-/- background (left). Quantification of the frequency of Ki67+ cells in
641 these populations (right). (e) Frequency of BrdU+ cells in naïve CD8 T cells from
26 642 spleens and lymph nodes of P14+ wild type (n=6) and P14+Lin28Tg (n=5) mice, both on
643 Rag2-/- background, labeled with BrdU in vivo. *** P < 0.001, compared with wild type
644 using two-tailed Student’s t-test. Data are one experiment representative of three
645 independent experiments (a, b; mean and s.e.m. of technical triplicates; c, d; mean and
646 s.e.m. of three experiments).
647
648 Figure 2. Inhibitory role of let-7 miRNA expression in CD8 T cell- mediated responses
649 in vivo
650 (a) Schematic representation of the hypothesis that let-7 expression inhibits the
651 differentiation of CD8 T cells. (b) Quantification of the number of donor (CD45.2+) CD8
652 T cells from P14+ wild type (n=3) or P14+let-7Tg mice (n=3) in the spleens of congenic
653 (CD45.1+) host mice 7 days after cell transfer and LCMV Armstrong infection. (c)
654 Surface expression of the activation marker KLRG1 on wild type host, and indicated
655 donor cells (top). Expression of IFN-γ, and TNF-α in wild type host, and donor LCMV-
656 specific CD8 T cells from the indicated mice, as determined by re-stimulation with the
657 gp33 peptide, and subsequent intracellular staining (middle). Quantification of the
658 frequency of KLRG1+, and IFN-γ+, TNF-α+ populations in wild type host, and donor cells
659 from the indicated mice (bottom). (d) Analysis of the survival of wild type (n=5) or let-
660 7Tg (n=5) mice injected i.p. with 30X106 P815 cells. (e) Quantification of the number of
661 P815 tumor cells remaining in the peritoneal cavity 7 days after i.p. injection of 20X106
662 cells into either wild type (n=6), or let-7Tg mice (n=5). n.s., not significant (P > 0.05), * P
663 < 0.05, ** P < 0.01, and *** P < 0.001, compared with wild type using two-tailed
664 Student’s t-test (b, c, d) or one-tailed Student’s t-test (e). Data are from one experiment
27 665 representative of three independent experiments (b, c; mean and s.e.m. of technical
666 triplicates) two experiments (d,e).
667
668 Figure 3. let-7 miRNAs suppress the proliferation and metabolism of activated CD8 T
669 cells
670 (a) Analysis of the proliferation of CellTrace Violet- labeled naïve CD8 T cells from the
671 indicated mice 72 hours after activation with anti- CD3 mAbs. (b) Quantitative RT-PCR
672 analysis of cell cycle regulators: Cdc25a (Cell division cycle 25A phosphatase), Ccnd2
673 (Cyclin D2), Ccnf (Cyclin F), Cdk6 (Cyclin dependent kinase 6) in naïve and activated
674 wild type, let-7Tg, and Lin28Tg CD8 T cells 3 days after anti-CD3 mAb stimulation,
675 presented relative to the ribosomal protein Rpl13a. (c) Quantitative RT-PCR analysis of
676 Myc and Tfap4 (Transcription factor AP-4) in CD8 T cells after stimulation with anti-CD3
677 mAbs, presented relative to the ribosomal protein Rpl13a. Wild type, let-7Tg, and
678 Lin28Tg CD8 T cells were stimulated with anti-CD3 mAbs and differentiated for the
679 indicated time. (d) Quantitative RT-PCR analysis of the expression of genes involved in
680 the metabolic switch: Glut1 (Glucose transporter 1), Glut3 (Glucose transporter 3),
681 Gpd2 (Glycerol-3-phosphate dehydrogenase 2), Pfk1 (Phosphofructokinase 1), Hk2
682 (Hexokinase 2), Tpi (Triose phosphate isomerase), Pkm (Pyruvate kinase muscle
683 isozyme), Ldha (Lactate dehydrogenase A), Yars (Tyrosyl-tRNA synthetase) in wild
684 type, let-7Tg, and Lin28Tg CD8 T cells three days after stimulation with anti-CD3 mAbs,
685 presented relative to the ribosomal protein, Rpl13a. n.s., not significant (P > 0.05), * P <
686 0.05, ** P < 0.01, and *** P < 0.001, compared with wild type using two-tailed Student’s
28 687 t-test. (a, b, c, d; one experiment representative of three independent experiments (a) or
688 mean and s.e.m. of technical triplicates (b, c, d)).
689
690 Figure 4. let-7 miRNAs negatively regulate differentiation and acquisition of effector
691 function in CTLs
692 (a, b) Cytotoxicity assay of differentiated CTLs from P14+ wild type, P14+let-7Tg, and
693 P14+Lin28Tg lymph nodes (a) or from P14+ wild type (WT), or P14+let-
694 7Tg+Lin28Tg+Rag2-/- (4Tg) lymphocytes co-cultured with either LCMV gp33 or LCMV
695 np396 peptide-pulsed splenocytes for 4-5 hours, either in the presence or absence of
696 doxycycline (b). (c) Analysis of the internal complexity (FSC, forward scatter; SSC, side
697 scatter) of effector (5 days after anti-CD3 mAb stimulation) CD8 T cells generated from
698 wild type, let-7Tg, and Lin28Tg lymphocytes (left) and quantification of SSC of CD8 T
699 cells normalized to wild type. (d) Quantification of Granzyme A and Granzyme B-
700 positive granules in wild type, let-7Tg, and Lin28Tg CTLs via MilliPore Amnis
701 ImageStream. (e) Quantitative RT-PCR analysis of effector molecule mRNA expression:
702 Gzma (Granzyme A), Gzmb (Granzyme B), Prf1 (Perforin) in naïve and effector CD8 T
703 cells from wild type, let-7Tg, and Lin28Tg lymph nodes, presented relative to expression
704 of the ribosomal protein Rpl13a. (f) Staining (top, middle) and MFI (bottom) of
705 Granzyme B, Interferon-γ, and Granzyme A in wild type, let-7Tg, and Lin28Tg effector
706 CD8 T cells normalized to wild type. (g) Western blot analysis of lysates of wild type, let-
707 7Tg, and Lin28Tg effector CD8 T cells, probed with monoclonal antibodies against
708 Perforin and actin. n.s., not significant (P > 0.05), * P < 0.05, ** P < 0.01, and *** P <
709 0.001, compared with wild type using two-tailed Student’s t-test. Data are from one
29 710 experiment representative of three experiments (a, b, e; mean and s.e.m. of technical
711 triplicates, f; mean and s.e.m of three experiments).
712
713 Figure 5. let-7 miRNAs directly target the mRNA of Eomes in activated CD8 T cells
714 (a) Quantitative RT-PCR analysis of Eomes (Eomesodermin) and Tbx21 (T-bet) mRNA
715 in naïve and effector CD8 T cells (5 days after anti-CD3 mAb stimulation) generated
716 from wild type, let-7Tg, and Lin28Tg lymphocytes, presented relative to the ribosomal
717 protein Rpl13a. (b) Staining of Eomes and T-bet (left) and MFI of Eomes and T-bet
718 (right) in wild type, let-7Tg, and Lin28Tg effector CD8 T cells, normalized to wild type.
719 (c) Quantitative RT-PCR analysis of Eomes (Eomesodermin) mRNA in P14+ wild type,
720 or P14+let-7Tg+Lin28Tg+Rag2-/- (4Tg) lymph nodes cultured either in the presence or
721 absence of Dox, presented as the fold change in expression, normalized to wild type.
722 (d) Eomes mRNA, including the 5’ and 3’ untranslated regions (UTR) and the open
723 reading frame (ORF) within which one let-7 binding site was identified (top), sequence
724 conservation between the mouse and human let-7 binding sites of Eomes is shown
725 (middle). Mutated sequence of the let-7 binding site in the cDNA of mouse Eomes
726 (bottom) used in the luciferase reporter assay. (e) Luciferase reporter assay of let-7
727 targeting of the Eomes ORF in NIH 3T3 fibroblasts transfected with a luciferase reporter
728 containing either the intact or mutated sequence of the let-7 seed region from the
729 mouse Eomes ORF; the activity of firefly luciferase was normalized to the Renilla
730 luciferase control. ** P < 0.01, and *** P < 0.001, compared with wild type using two-
731 tailed Student’s t-test. Data are from one experiment representative of three
30 732 independent experiments (a, b, c, e; mean and s.e.m. of technical triplicates (a,c,e) or
733 three experiments (b)).
734
735 Figure 6. let-7 miRNAs control the differentiation of CTLs through Eomes-dependent
736 and independent mechanisms
737 (a) Quantitative RT-PCR analysis of Eomes mRNA, presented as the fold change in
738 expression, normalized to wild type (left) and MFI of Eomes protein expression,
739 normalized to wild type (right), from CTLs generated from P14+CD4-Cre-Eomesfl/wt,
740 P14+CD4-Cre-EomeswtLin28Tg, P14+CD4Cre+Eomesfl/wtLin28Tg, and
741 P14+CD4Cre+Eomesfl/flLin28Tg CD8 T cells. (b) Quantitative RT-PCR analysis of GzmA
742 (Granzyme A) and Prf1 (Perforin) mRNA in effector CTLs generated from the indicated
743 mice, presented as the fold change in expression, normalized to wild type. (c) Staining
744 of Granzyme A in CTLs generated from the indicated mice. (d) Cytotoxicity assay
745 demonstrating specific target lysis of differentiated P14+ CTLs from the indicated mice,
746 co-cultured with either LCMV gp33 or LCMV np396 peptide-pulsed splenocytes for 4-5
747 hours. ** P < 0.01, and *** P < 0.001, compared with Lin28Tg using two-tailed Student’s
748 t-test. Data are from one experiment representative of two independent experiments (a;
749 mean and s.e.m. of two experiments; b, d; mean and s.e.m. or technical triplicates).
750
751 Figure 1—figure supplement 1. let-7 expression is downregulated upon TCR stimulation
752 in CD8 T cells
753 (a) Quantitative RT-PCR analysis of individual let-7 miRNAs in naïve CD8 T cells
754 stimulated with plate-bound anti-TCRβ (μg as indicated) and anti-CD28 (5 μg) for 24
31 755 hours (from figure 1A), presented relative to results obtained for the small nuclear RNA
756 (U6 control) and normalized to the unstimulated (0 μg) sample, and displayed as
757 individual miRNAs. (b) Quantitative RT-PCR analysis of individual let-7 miRNAs in naïve
758 CD8 T cells stimulated with plate-bound anti-TCRβ (5μg) and anti-CD28 (5 μg) over
759 increasing periods of time as indicated (from figure 1B), presented relative to U6 control
760 and normalized to the unstimulated (0h) sample, and displayed as individual miRNAs.
761 n.s., not significant (P > 0.05), * P < 0.05, ** P < 0.01, and *** P < 0.001, compared with
762 0 μg (a) or 0 hours (b) using two-tailed Student’s t-test (a, b). Data are one experiment
763 representative of three independent experiments (b; mean and s.e.m. of technical
764 triplicates).
765
766 Figure 1—figure supplement 2. TCR-mediated regulation of miRNA expression
767 (a) Quantitative RT-PCR analysis of individual let-7 miRNAs in naïve CD8 T cells
768 stimulated with plate-bound anti-TCRβ (10μg) and anti-CD28 (5 μg) over increasing
769 periods of time as indicated, presented relative to results obtained for the small
770 nucleolar RNA (sno135) and normalized to the unstimulated (0h) sample. (b)
771 Quantitative RT-PCR analysis of miR-17 in naïve CD8 T cells stimulated with plate-
772 bound anti-TCRβ (10μg) and anti-CD28 (5 μg) over increasing periods of time as
773 indicated, presented relative to results obtained for the small nuclear RNA (U6) (left) or
774 the small nucleolar RNA (sno135) (right) and normalized to the unstimulated (0h)
775 sample. *** P < 0.001, compared with 0 hours using two-tailed Student’s t-test (b). Data
776 are from one experiment representative of two independent experiments (a, b; mean
777 and s.e.m. of technical triplicates).
32 778
779 Figure 1—figure supplement 3. Expression of let-7 and activation markers in P14+ CD8
780 T cells
781 (a) Quantitative RT-PCR analysis of individual let-7 miRNAs in naïve and effector (day
782 5) TCR transgenic CD8 T cells from P14+Rag2-/-, and P14+Lin28TgRag2-/- mice. (b)
783 Surface expression (right) and normalized MFI (left) of CD25, CD44, and CD122 on
784 naïve CD8 T cells from both spleens and lymph nodes of P14+ wild type and P14+
785 Lin28Tg mice, both on Rag2-/- background. Gray indicates an isotype control for
786 staining. n.s., not significant (P > 0.05), * P < 0.05, ** P < 0.01, and *** P < 0.001,
787 compared with wild type using two-tailed Student’s t-test (b). Data are from one
788 experiment representative of three independent experiments (a; mean and s.e.m. of
789 technical triplicates; b, c; mean and s.e.m. of three experiments).
790
791 Figure 2—figure supplement 1. P815 (H-2d) tumor rejection is mediated by CD8 T cells
792 in H-2b mice
793 (a) Quantitative RT-PCR analysis of individual let-7 miRNAs in naïve and effector (day
794 5) TCR transgenic CD8 T cells from P14+Rag2-/-, and P14+let-7TgRag2-/- mice (b)
795 Analysis of the survival of Rag2-/- mice injected i.p. with 30X106 P815 cells, which
796 received i.v. adoptive transfer of 10X106 naïve purified CD8 T cells (n=6) or no T cells at
797 all (n=8). Data are from one experiment representative of three independent
798 experiments (a; mean and s.e.m. of technical triplicates) or one experiment (b).
799
33 800 Figure 3—figure supplement 1. Sorting strategy and let-7 expression in polyclonal CD8
801 T cells
802 (a) Pre- sort and post- sort purities for naïve CD8 T cells from wild type, let-7Tg, and
803 Lin28Tg mice. Cells were isolated from the lymph nodes of the indicated mice, and
804 enriched for T cells using anti-mouse IgG magnetic beads. Next, to acquire naïve CD8
805 T cells, lymph node T cells were then stained for CD25, CD44, CD4, and CD8, and
806 sorted as the CD25-CD44loCD4-CD8+ population. (b) Quantitative RT-PCR analysis of
807 individual let-7 miRNAs in naïve and activated (day 5) polyclonal CD8 T cells from wild
808 type, let-7Tg, and Lin28Tg mice.
809
810 Figure 4—figure supplement 1. Effector molecule mRNA expression in 4Tg mice
811 Quantitative RT-PCR analysis of effector molecule mRNA expression: Prf1 (Perforin),
812 Gzma (Granzyme A), Gzmb (Granzyme B) in effector CD8 T cells from P14+ wild type,
813 or P14+let-7Tg+Lin28Tg+Rag2-/- (4Tg) lymph nodes cultured either in the presence or
814 absence of Dox, presented as the fold change in expression, normalized to wild type.
815 *** P < 0.001, 4Tg without Dox compared with 4Tg with Dox using two-tailed Student’s t-
816 test. Data are from one experiment (mean and s.e.m. of technical triplicates).
817
818 Figure 5—figure supplement 1. Effect of let-7 miRNAs on expression of transcription
819 factors in effector CD8 T cells.
820 (a, b) Quantitative RT-PCR analysis of Notch target genes: Hes1 (Hes family BHLH
821 transcription factor 1), Hey1 (Hes related family BHLH transcription factor with YRPW
822 motif 1), Heyl (Hes related family BHLH transcription factor with YRPW motif- like), Dtx1
34 823 (Deltex 1) (a) and transcription factors that control CD8 T cell differentiation: Zbtb32
824 (Zinc finger and BTB domain containing 32), Prdm1 (Blimp-1), Runx3d (Runt related
825 transcription factor 3 distal) (b) in naïve and effector (5 days after anti-CD3 mAb
826 stimulation) wild type, let-7Tg, and Lin28Tg CD8 T cells, presented relative to the
827 expression of the ribosomal protein Rpl13a. Data are from one experiment
828 representative of three independent experiments (a, b; mean and s.e.m. of technical
829 triplicates). N.D., not determined.
830
831 Figure 5—figure supplement 2. Let-7 expression and luciferase detection in NIH 3T3
832 fibroblasts
833 (a) Quantitative RT-PCR analysis of individual let-7 miRNAs in NIH 3T3 fibroblasts,
834 presented relative to results obtained for the small nuclear RNA (U6) (b) Expression
835 analysis of three let-7 family members by luciferase reporter assay. A firefly luciferase
836 reporter containing either intact or mutated let-7b, let-7g, or let-7i antisense seed
837 regions were transfected into NIH 3T3 cells. The activity of firefly luciferase was
838 normalized to the Renilla luciferase control. Data are one experiment representative of
839 three independent experiments (a, b; mean and s.e.m. of technical triplicates).
840
841 Figure 6—figure supplement 1. Granzyme B expression is regulated independent of
842 Eomes
843 Staining of Granzyme B in CTLs generated from the indicated mice. n.s., not significant
844 (P > 0.05), 4Tg without Dox compared with 4Tg with Dox using two-tailed Student’s t-
35 845 test.Data are from one experiment representative of two independent experiments
846 (mean and s.e.m. of two experiments).
847
848 Figure 6—figure supplement 2. Re-expression of Eomes enhances cytotoxic function of
849 CTLs
850 (a) Cytotoxicity assay demonstrating specific target lysis of LCMV gp33 or LCMV np396
851 peptide-pulsed splenocytes by differentiated P14+CD4Cre+Eomesfl/flLin28Tg CTLs
852 transduced with the indicated virus, cells were co-cultured for 4-5 hours. (b) Quantitative
853 RT-PCR analysis of Eomes (Eomesodermin), Gzma (Granzyme A), Gzmb (Granzyme
854 B), and Prf1 (Perforin) expression in CTLs described in a, presented as the fold change
855 in expression, normalized to RV-Empty. (c) Visualization of the mutations made to the
856 let-7 binding motif in mouse Eomes ORF to generate the RV-EomesMUTANT (top), and
857 cytotoxicity assay demonstrating specific target lysis of differentiated P14+let-7Tg CTLs
858 transduced with the indicated virus, co-cultured with either LCMV gp33 or LCMV np396
859 peptide-pulsed splenocytes for 4-5 hours (bottom). (d) Quantitative RT-PCR analysis of
860 Eomes (Eomesodermin), Gzma (Granzyme A), Gzmb (Granzyme B), and Prf1
861 (Perforin) in effector CTLs from c, presented as the fold change in expression,
862 normalized to RV-Empty. n.s., not significant (P > 0.05), ** P <0.01, *** P < 0.001, RV-
863 Eomes or RV-EomesMUTANT compared with RV-Empty using two-tailed Student’s t-test
864 (a, b, c, d). Data are from one experiment representative of two independent
865 experiments (a, b, c, d; mean and s.e.m. of technical triplicates).
866
36 867 Figure 6—figure supplement 3. Model of let-7- mediated regulation of CD8 T cell
868 differentiation
869 (a) Upon TCR stimulation, let-7 miRNAs are rapidly down-regulated, in order to release
870 the ‘molecular brake’ let-7 places on CD8 T cell differentiation in naïve cells. Our study
871 suggests that let-7 miRNAs inhibit clonal expansion and metabolic reprogramming of
872 activated CD8 T cells through inhibition of Myc and other cell cycle controlling factors.
873 Further, our data indicates let-7 also suppresses the acquisition of effector function of
874 CTLs through the direct targeting of Eomes mRNA.
875
876 Supplementary File 1. A list of all genes containing a 9-base pair let-7 binding motif in
877 either the 3’ untranslanted region (UTR) or open reading frame (ORF) from both the
878 mouse and human genomes, where a (+) or (-) demonstrates the orientation of the gene
879 in the genome. The ending models are an illustration of the location of the let-7 binding
880 motifs within the human and mouse Eomes gene, where red represents exons and pink
881 represents introns.
882
883 Supplementary File 2. Primer sequences and Taqman Assay catalog numbers
884
885 Figure 1- source data 1. Quantification of the expression of miRNAs by qPCR, and flow
886 cytometry analysis of the phenotype of naïve CD8 T cells.
887
888 Figure 2- source data1. Quantification of let-7 expression by qPCR, and flow cytometry
889 analysis of effector CD8 T cells during the response to LCMV infection.
37 890
891 Figure 3- source data 1. Quantification of the expression of cell cycle regulators, Myc
892 and Tfap4, metabolic enzymes, and let-7 miRNAs by qPCR.
893
894 Figure 4- source data 1. Quantification of cytotoxic assays and expression of effector
895 molecules assessed by qPCR and flow cytometry in CTLs.
896
897 Figure 5- source data 1. Quantification of luciferase activity, and the expression of
898 transcription factors in CTLs assessed by flow cytometry and qPCR.
899
900 Figure 6- source data 1. Quantification of cytotoxic assays and expression of effector
901 molecules assessed by qPCR and flow cytometry in CTLs.
902
903 References
904 1. Ward, P.S. & Thompson, C.B. Signaling in control of cell growth and metabolism. Cold 905 Spring Harb Perspect Biol 4, a006783 (2012). 906 2. Backer, R.A. et al. A central role for Notch in effector CD8(+) T cell differentiation. Nat 907 Immunol 15, 1143-51 (2014). 908 3. Best, J.A. et al. Transcriptional insights into the CD8(+) T cell response to infection and 909 memory T cell formation. Nat Immunol 14, 404-12 (2013). 910 4. Pearce, E.L. et al. Control of effector CD8+ T cell function by the transcription factor 911 Eomesodermin. Science 302, 1041-3 (2003). 912 5. Szabo, S.J. et al. A novel transcription factor, T-bet, directs Th1 lineage commitment. 913 Cell 100, 655-69 (2000). 914 6. Bartel, D.P. MicroRNAs: target recognition and regulatory functions. Cell 136, 215-33 915 (2009). 916 7. Ha, M. & Kim, V.N. Regulation of microRNA biogenesis. Nat Rev Mol Cell Biol 15, 917 509-24 (2014). 918 8. Chen, C.Z., Li, L., Lodish, H.F. & Bartel, D.P. MicroRNAs modulate hematopoietic 919 lineage differentiation. Science 303, 83-6 (2004). 920 9. Cobb, B.S. et al. A role for Dicer in immune regulation. J Exp Med 203, 2519-27 (2006).
38 921 10. Cobb, B.S. et al. T cell lineage choice and differentiation in the absence of the RNase III 922 enzyme Dicer. J Exp Med 201, 1367-73 (2005). 923 11. Muljo, S.A. et al. Aberrant T cell differentiation in the absence of Dicer. J Exp Med 202, 924 261-9 (2005). 925 12. Henao-Mejia, J. et al. The microRNA miR-181 is a critical cellular metabolic rheostat 926 essential for NKT cell ontogenesis and lymphocyte development and homeostasis. 927 Immunity 38, 984-97 (2013). 928 13. Li, Q.J. et al. miR-181a is an intrinsic modulator of T cell sensitivity and selection. Cell 929 129, 147-61 (2007). 930 14. Ma, F. et al. The microRNA miR-29 controls innate and adaptive immune responses to 931 intracellular bacterial infection by targeting interferon-gamma. Nat Immunol 12, 861-9 932 (2011). 933 15. Steiner, D.F. et al. MicroRNA-29 regulates T-box transcription factors and interferon- 934 gamma production in helper T cells. Immunity 35, 169-81 (2011). 935 16. Zhang, N. & Bevan, M.J. Dicer controls CD8+ T-cell activation, migration, and survival. 936 Proc Natl Acad Sci U S A 107, 21629-34 (2010). 937 17. Katz, G. et al. T cell receptor stimulation impairs IL-7 receptor signaling by inducing 938 expression of the microRNA miR-17 to target Janus kinase 1. Sci Signal 7, ra83 (2014). 939 18. Wu, T. et al. Temporal expression of microRNA cluster miR-17-92 regulates effector and 940 memory CD8+ T-cell differentiation. Proc Natl Acad Sci U S A 109, 9965-70 (2012). 941 19. Pobezinsky, L.A. et al. Let-7 microRNAs target the lineage-specific transcription factor 942 PLZF to regulate terminal NKT cell differentiation and effector function. Nat Immunol 943 16, 517-24 (2015). 944 20. Yuan, J., Nguyen, C.K., Liu, X., Kanellopoulou, C. & Muljo, S.A. Lin28b reprograms 945 adult bone marrow hematopoietic progenitors to mediate fetal-like lymphopoiesis. 946 Science 335, 1195-200 (2012). 947 21. Polikepahad, S. et al. Proinflammatory role for let-7 microRNAS in experimental asthma. 948 J Biol Chem 285, 30139-49 (2010). 949 22. Swaminathan, S. et al. Differential regulation of the Let-7 family of microRNAs in CD4+ 950 T cells alters IL-10 expression. J Immunol 188, 6238-46 (2012). 951 23. Okoye, I.S. et al. MicroRNA-containing T-regulatory-cell-derived exosomes suppress 952 pathogenic T helper 1 cells. Immunity 41, 89-103 (2014). 953 24. Piskounova, E. et al. Lin28A and Lin28B inhibit let-7 microRNA biogenesis by distinct 954 mechanisms. Cell 147, 1066-79 (2011). 955 25. Cuylen, S. et al. Ki-67 acts as a biological surfactant to disperse mitotic chromosomes. 956 Nature 535, 308-12 (2016). 957 26. Zhu, H. et al. The Lin28/let-7 axis regulates glucose metabolism. Cell 147, 81-94 (2011). 958 27. Dominguez, C.X. et al. The transcription factors ZEB2 and T-bet cooperate to program 959 cytotoxic T cell terminal differentiation in response to LCMV viral infection. J Exp Med 960 212, 2041-56 (2015). 961 28. Joshi, N.S. et al. Inflammation directs memory precursor and short-lived effector CD8(+) 962 T cell fates via the graded expression of T-bet transcription factor. Immunity 27, 281-95 963 (2007). 964 29. Thimme, R. et al. Increased expression of the NK cell receptor KLRG1 by virus-specific 965 CD8 T cells during persistent antigen stimulation. J Virol 79, 12112-6 (2005).
39 966 30. Voehringer, D. et al. Viral infections induce abundant numbers of senescent CD8 T cells. 967 J Immunol 167, 4838-43 (2001). 968 31. Kaech, S.M., Hemby, S., Kersh, E. & Ahmed, R. Molecular and functional profiling of 969 memory CD8 T cell differentiation. Cell 111, 837-51 (2002). 970 32. Wherry, E.J. & Ahmed, R. Memory CD8 T-cell differentiation during viral infection. J 971 Virol 78, 5535-45 (2004). 972 33. Williams, M.A. & Bevan, M.J. Effector and memory CTL differentiation. Annu Rev 973 Immunol 25, 171-92 (2007). 974 34. Felix, N.J. & Allen, P.M. Specificity of T-cell alloreactivity. Nat Rev Immunol 7, 942-53 975 (2007). 976 35. Jankovic, V., Remus, K., Molano, A. & Nikolich-Zugich, J. T cell recognition of an 977 engineered MHC class I molecule: implications for peptide-independent alloreactivity. J 978 Immunol 169, 1887-92 (2002). 979 36. Zhan, Y., Corbett, A.J., Brady, J.L., Sutherland, R.M. & Lew, A.M. CD4 help- 980 independent induction of cytotoxic CD8 cells to allogeneic P815 tumor cells is absolutely 981 dependent on costimulation. J Immunol 165, 3612-9 (2000). 982 37. Johnson, C.D. et al. The let-7 microRNA represses cell proliferation pathways in human 983 cells. Cancer Res 67, 7713-22 (2007). 984 38. Iritani, B.M. et al. Modulation of T-lymphocyte development, growth and cell size by the 985 Myc antagonist and transcriptional repressor Mad1. EMBO J 21, 4820-30 (2002). 986 39. Kim, H.H. et al. HuR recruits let-7/RISC to repress c-Myc expression. Genes Dev 23, 987 1743-8 (2009). 988 40. Nie, Z. et al. c-Myc is a universal amplifier of expressed genes in lymphocytes and 989 embryonic stem cells. Cell 151, 68-79 (2012). 990 41. Verbist, K.C. et al. Metabolic maintenance of cell asymmetry following division in 991 activated T lymphocytes. Nature 532, 389-93 (2016). 992 42. Chou, C. et al. c-Myc-induced transcription factor AP4 is required for host protection 993 mediated by CD8+ T cells. Nat Immunol 15, 884-93 (2014). 994 43. Cham, C.M., Driessens, G., O'Keefe, J.P. & Gajewski, T.F. Glucose deprivation inhibits 995 multiple key gene expression events and effector functions in CD8+ T cells. Eur J 996 Immunol 38, 2438-50 (2008). 997 44. Wang, R. et al. The transcription factor Myc controls metabolic reprogramming upon T 998 lymphocyte activation. Immunity 35, 871-82 (2011). 999 45. Hayes, M.P., Berrebi, G.A. & Henkart, P.A. Induction of target cell DNA release by the 1000 cytotoxic T lymphocyte granule protease granzyme A. J Exp Med 170, 933-46 (1989). 1001 46. Lancki, D.W., Hsieh, C.S. & Fitch, F.W. Mechanisms of lysis by cytotoxic T lymphocyte 1002 clones. Lytic activity and gene expression in cloned antigen-specific CD4+ and CD8+ T 1003 lymphocytes. J Immunol 146, 3242-9 (1991). 1004 47. Lai, E.C. Micro RNAs are complementary to 3' UTR sequence motifs that mediate 1005 negative post-transcriptional regulation. Nat Genet 30, 363-4 (2002). 1006 48. Reinhart, B.J. et al. The 21-nucleotide let-7 RNA regulates developmental timing in 1007 Caenorhabditis elegans. Nature 403, 901-6 (2000). 1008 49. Wightman, B., Ha, I. & Ruvkun, G. Posttranscriptional regulation of the heterochronic 1009 gene lin-14 by lin-4 mediates temporal pattern formation in C. elegans. Cell 75, 855-62 1010 (1993).
40 1011 50. Cho, O.H. et al. Notch regulates cytolytic effector function in CD8+ T cells. J Immunol 1012 182, 3380-9 (2009). 1013 51. Kallies, A., Xin, A., Belz, G.T. & Nutt, S.L. Blimp-1 transcription factor is required for 1014 the differentiation of effector CD8(+) T cells and memory responses. Immunity 31, 283- 1015 95 (2009). 1016 52. Rutishauser, R.L. et al. Transcriptional repressor Blimp-1 promotes CD8(+) T cell 1017 terminal differentiation and represses the acquisition of central memory T cell properties. 1018 Immunity 31, 296-308 (2009). 1019 53. Szabo, S.J. et al. Distinct effects of T-bet in TH1 lineage commitment and IFN-gamma 1020 production in CD4 and CD8 T cells. Science 295, 338-42 (2002). 1021 54. Shin, H.M., Kapoor, V., Welsh, R. & Berg, L. The transcription factor ZBTB32 1022 negatively regulates CD8+T cell responses during acute and chronic viral infection. 1023 Journal of Immunology 192(2014). 1024 55. Trifari, S. et al. MicroRNA-directed program of cytotoxic CD8+ T-cell differentiation. 1025 Proc Natl Acad Sci U S A 110, 18608-13 (2013). 1026 56. Abbott, A.L. et al. The let-7 MicroRNA family members mir-48, mir-84, and mir-241 1027 function together to regulate developmental timing in Caenorhabditis elegans. Dev Cell 9, 1028 403-14 (2005). 1029 57. Bussing, I., Slack, F.J. & Grosshans, H. let-7 microRNAs in development, stem cells and 1030 cancer. Trends Mol Med 14, 400-9 (2008). 1031 58. Roush, S. & Slack, F.J. The let-7 family of microRNAs. Trends Cell Biol 18, 505-16 1032 (2008). 1033 59. Yu, F. et al. let-7 regulates self renewal and tumorigenicity of breast cancer cells. Cell 1034 131, 1109-23 (2007). 1035 60. Goldrath, A.W. et al. Cytokine requirements for acute and Basal homeostatic 1036 proliferation of naive and memory CD8+ T cells. J Exp Med 195, 1515-22 (2002). 1037 61. Kamimura, D. & Bevan, M.J. Naive CD8+ T cells differentiate into protective memory- 1038 like cells after IL-2 anti IL-2 complex treatment in vivo. J Exp Med 204, 1803-12 (2007). 1039 62. Kieper, W.C., Burghardt, J.T. & Surh, C.D. A role for TCR affinity in regulating naive T 1040 cell homeostasis. J Immunol 172, 40-4 (2004). 1041 63. Kieper, W.C. et al. Overexpression of interleukin (IL)-7 leads to IL-15-independent 1042 generation of memory phenotype CD8+ T cells. J Exp Med 195, 1533-9 (2002). 1043 64. Kimura, M.Y. et al. IL-7 signaling must be intermittent, not continuous, during CD8(+) T 1044 cell homeostasis to promote cell survival instead of cell death. Nat Immunol 14, 143-51 1045 (2013). 1046 65. Intlekofer, A.M. et al. Effector and memory CD8+ T cell fate coupled by T-bet and 1047 eomesodermin. Nat Immunol 6, 1236-44 (2005). 1048 66. Weinreich, M.A., Odumade, O.A., Jameson, S.C. & Hogquist, K.A. T cells expressing 1049 the transcription factor PLZF regulate the development of memory-like CD8+ T cells. 1050 Nat Immunol 11, 709-16 (2010). 1051 67. Feng, X. et al. Transcription factor Foxp1 exerts essential cell-intrinsic regulation of the 1052 quiescence of naive T cells. Nat Immunol 12, 544-50 (2011). 1053 68. Frauwirth, K.A. et al. The CD28 signaling pathway regulates glucose metabolism. 1054 Immunity 16, 769-77 (2002). 1055 69. Frauwirth, K.A. & Thompson, C.B. Regulation of T lymphocyte metabolism. J Immunol 1056 172, 4661-5 (2004).
41 1057 70. Grumont, R. et al. The mitogen-induced increase in T cell size involves PKC and NFAT 1058 activation of Rel/NF-kappaB-dependent c-myc expression. Immunity 21, 19-30 (2004). 1059 71. Askew, D.S., Ihle, J.N. & Cleveland, J.L. Activation of apoptosis associated with 1060 enforced myc expression in myeloid progenitor cells is dominant to the suppression of 1061 apoptosis by interleukin-3 or erythropoietin. Blood 82, 2079-87 (1993). 1062 72. Evan, G.I. et al. Induction of apoptosis in fibroblasts by c-myc protein. Cell 69, 119-28 1063 (1992). 1064 73. Fanidi, A., Harrington, E.A. & Evan, G.I. Cooperative interaction between c-myc and 1065 bcl-2 proto-oncogenes. Nature 359, 554-6 (1992). 1066 74. Sampson, V.B. et al. MicroRNA let-7a down-regulates MYC and reverts MYC-induced 1067 growth in Burkitt lymphoma cells. Cancer Res 67, 9762-70 (2007). 1068 75. Bouchard, C. et al. Direct induction of cyclin D2 by Myc contributes to cell cycle 1069 progression and sequestration of p27. EMBO J 18, 5321-33 (1999). 1070 76. Galaktionov, K., Chen, X. & Beach, D. Cdc25 cell-cycle phosphatase as a target of c- 1071 myc. Nature 382, 511-7 (1996). 1072 77. Mateyak, M.K., Obaya, A.J. & Sedivy, J.M. c-Myc regulates cyclin D-Cdk4 and -Cdk6 1073 activity but affects cell cycle progression at multiple independent points. Mol Cell Biol 1074 19, 4672-83 (1999). 1075 78. Wang, J. et al. Fetal and adult progenitors give rise to unique populations of CD8+ T 1076 cells. Blood 128, 3073-3082 (2016). 1077 79. Chang, J.T., Wherry, E.J. & Goldrath, A.W. Molecular regulation of effector and 1078 memory T cell differentiation. Nat Immunol 15, 1104-15 (2014). 1079 80. Nayar, R. et al. TCR signaling via Tec kinase ITK and interferon regulatory factor 4 1080 (IRF4) regulates CD8+ T-cell differentiation. Proc Natl Acad Sci U S A 109, E2794-802 1081 (2012). 1082 81. Markstein, M., Markstein, P., Markstein, V. & Levine, M.S. Genome-wide analysis of 1083 clustered Dorsal binding sites identifies putative target genes in the Drosophila embryo. 1084 Proc Natl Acad Sci U S A 99, 763-8 (2002). 1085
42 FSC
Normalized A let-7 expression in CD8 T cells B let-7 expressionFigure 1A in CD8 T cells C Naïve CD8 T cells * 1.0 1.5 1.5 1.0 1.0 let-7a 1.0 let-7a ** let-7b let-7b B let-7a let-7a + 1.0 let-7clet-7b let-7clet-7b 1.0 W let-7c let-7c T 0.5 let-7d 0.5 let-7d 0.5 let-7d 0.5 let-7d let-7elet-7e let-7elet-7e 0.5 0.5 miRNA expression miRNA expression let-7flet-7f let-7flet-7f let-7g let-7g let-7glet-7i let-7glet-7i 0.0 0.0 0.0 Normalized miRNA NormalizedmiRNA 0.0 NormalizedmiRNA 0.0 0.0 Normalized Area(FSC) Normalized let-7i 0 h let-7i 0µg 0.1µg 0.5µg 1µg 0h 15h15 h 24h24 h 48h48 h 72h72 h Spleen Lymph node expression(relative U6)to expression(relative U6)to %Ki67+ spleen 24 hr stimulation Time Timeof ofTCR TCR stimulation stimulation WT Lin28Tg 1 ug Naive 0.1 ug 0.5 ug
LNT D Time of TCR stimulation Naïve CD8 T cells E %BrdU+ spleenNaïve CD8Spleen T cells%BrdU+ LNT
40 *** WT Lin28Tg 40 Spleen Lymph node B 30 30 + *** *** 1.02% 32.6% W 15T 1.5 20 15 1.5 20 %Ki67+ LNT B
Spleen + 10 10 10 cells 10 1.0 1.0 W
+ T 0 0 5 0.5 4 5 0.5 10 *** BrdU 40
40 % 103 2.66% 33.4% 0 0 0.0 0.0 30 30 Spleen
2 cells 10 + WT Lin28Tg 20 20 1 LNT 10 Spleen 10 10 %Ki67 Lymph node Lymph 100 CD8 100 101 102 103 104 0 0 Ki67 WT Lin28Tg
LNT
Figure 1 B C D A A let-7a let-7b let-7c let-7d 1.5 1.5 1.5 1.5 1.5 1.5 1.5 b 1.5 c *** d ** n.s. n.s. 1.0 1.0 1.0 1.0 1.0 1.0 1.0 1.0 *** *** *** *** 0.5 0.5 0.5 0.5 *** 0.5 0.5 *** 0.5 0.5 *** *** a 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0µg 0.1µg 0.5µg 1µg 0µg 0.1µg 0.5µg 1µg 0µg 0.1µg 0.5µg 1µg 0µg 0.1µg 0.5µg 1µg expression 24 hr stimulation 24 hr stimulation 24 hr stimulation 24 hr stimulation 0 ug 1 ug 0 ug 1 ug 0 ug 1 ug 0 ug 1 ug 01. ug E05. ug 01. ug 05.F ug 01. ug 05.G ug 01. ug 05.I ug
miRNA let-7e let-7f let-7g let-7i Hours Hours Hours 1.5 Hours 1.5 1.5 1.5 1.5 1.5 e 1.5 f 1.5 g i
n.s. 1.0 1.0 *** 1.0 1.0 1.0 1.0 1.0 1.0
Normalized *** ** *** *** *** *** *** *** *** 0.5 0.5 0.5 0.5 0.5 0.5 *** 0.5 0.5
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0µg 0.1µg 0.5µg 1µg 0µg 0.1µg 0.5µg 1µg 0µg 0.1µg 0.5µg 1µg 0µg 0.1µg 0.5µg 1µg 24 hr stimulation 24 hr stimulation 24 hr stimulation 24 hr stimulation 0 ug 1 ug 0 ug 1 ug 0 ug 1 ug 0 ug 1 ug 01. ug 05. ug 01. ug 05. ug 01. ug 05. ug 01. ug 05. ug
B C D B HoursA Hours Hours Hours let-7a let-7b let-7c let-7d 1.5 1.5 1.5 1.5 1.5 1.5 1.5 b 1.5 c d n.s. 1.01.0 1.0 1.0 1.0 1.0 1.0 1.0 ** *** *** *** 0.5 *** *** 0.5 *** ***0.5 0.5 0.5 0.5 0.5 0.5 *** *** *** *** a *** *** *** 0.00.0 0.0 0.0 0.0 0.0 0.0 0.0 0h 15h 24h 48h 72h 0h 15h 24h 48h 72h
expression 0h 15h 24h 48h 72h 0h 15h 24h 48h 72h
0 0 0 0 Time of TCR15 24 stimulation48 72 Time of TCR15 24 stimulation48 72 Time of 15TCR24 stimulation48 72 Time of 15TCR24 stimulation48 72 E F G I
miRNA let-7eHours let-7fHours let-7gHours let-7iHours 1.5 1.5 1.5 1.5 1.5 1.5 e 1.5 f 1.5 g i
1.0 1.0 ** 1.0 1.0 * 1.0 1.0 1.0 1.0 Normalized *** *** *** *** *** *** 0.5 0.5 *** 0.5 0.5 0.5 0.5 *** 0.5 0.5 *** *** *** *** *** ***
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0h 15h 24h 48h 72h 0h 15h 24h 48h 72h 0h 15h 24h 48h 72h 0h 15h 24h 48h 72h 0 0 0 0 Time of TCR15 24 stimulation48 72 Time of TCR15 24 stimulation48 72 Time of 15TCR24 stimulation48 72 Time of 15TCR24 stimulation48 72
Hours Hours Hours Hours
Figure 1- figure supplement 1 Normalized to sno135
A let-7 expression in CD8 T cells
1.0 1.0 let-7a
let-7blet-7a let-7clet-7b let-7c 0.5 0.5 let-7d let-7d let-7elet-7e miRNA expression let-7flet-7f let-7glet-7g
Normalized miRNA NormalizedmiRNA let-7i 0.0 0.0 let-7i 0h 12h 24h 48h expression (relative sno135)to Time of TCR stimulation 0h 12h 24h 48h B miR17-miR-17 U6 expression in CD8 T cellsmiR17- sno135 Time of TCR stimulation *** *** *** 6 6 8 8 *** *** 5 5 7 7 *** 6 6 4 4 5 5 3 3 4 4 3 2 2 3 miRNA expression miRNA expression 2 2 1 1 1
Normalized miRNA NormalizedmiRNA 1 Normalized miRNA NormalizedmiRNA 0 0 0 0 expression(relative U6)to 0h 12h 24h 48h 0h 12h 24h 48h Time of TCR stimulation expression(relative sno135)to Time of TCR stimulation 0h 0h 12h 24h 48h 12h 24h 48h
Time of TCR stimulation Time of TCR stimulation
Figure 1- figure supplement 2 Let-7 in P14+/ B+ P14+
1.5 CD25 A B let-7 expression in P14+ CD8 T cells Naïve CD8 T cells CD25 let-7a let-7a Spleen Lymph node 1.01.0 1.5 1.5 n.s. let-7b 100let-7b 100 n.s. B let-7c80 80 + let-7c 1.0 1.0 W let-7d let-7d60 60 T let-7e 0.50.5 let-7e 40 40 0.5 0.5 let-7f CD122 miRNA expression let-7f 20 20 let-7g0 0 0.0 0.0 let-7g let-7i 100 101 102 103 104 100 101 102 103 104 Spleen Lymph node
Normalized miRNA NormalizedmiRNA let-7i CD25 CD25 0.00.0 CD122 LNT
expression(relative U6)to 2.5 100 100 2.5 Spleen* * WT WT WT 2.0 2.0 B 80 80 + W 1.5 WT WT 60 60 1.5 T Lin28Tg Lin28 Lin28 Lin28Tg 40 40 1.0 1.0 Naïve Effector 20 20 0.5 0.5 CD44 0 0 0.0 0.0 100 101 102 103 104 100 101 102 103 104 Spleen Lymph node CD122 CD122 CD44 2.5 LNT 100 100 2.5 Spleen* * 2.0 2.0 B 80 80 + W 1.5 60 60 1.5 T 40 40 1.0 1.0 20 20 0.5 0.5 0.0 Events (% of max) Events(%of 0 0 Normalized MFI 0.0 100 101 102 103 104 100 101 102 103 104 Spleen Lymph node CD44 CD44
WT Lin28Tg WT Lin28TgLNT Spleen
Figure 1- figure supplement 3 P14+ (WT) WT P14+ (let-7Tg) WT A C host donor CD8 T cell host donor (CD45.1+) (CD45.2+) (CD45.1+) (CD45.2+) Function 100 let-7 80 expression 60 52.5% 33.6% 40 58.1% 2.34% B 20
Total CD45.2+ CTLs max %of 0 CD45.2+ count 3 3 4 5
) -10 0 10 10 10 400006 ** 40 KLRG1 30000 30 105 5.48 16.2 3.44 60.8 6.46 11.5 4.48 32.5 20000 20 104 10000 10 CD8 T cells from: 103 0 0 0 donor host 69.6 8.74 13.4 22.4 75.2 6.90 26.8 36.2 + -103 P14 (WT) WT Numbercellsof (x10 WT let-7Tg TNF-a -103 0 103 104 105 P14+ (let-7Tg) WT WT ilet-7 IFN-g
host donor host donor AW-59 Host % KLRG1 AW-59 Donor % KLRG1 AW-59 Host % IFNg+TNFa+ AW-59 Donor % IFNg+TNFa+ (CD45.1+) (CD45.2+) (CD45.1+) (CD45.2+)
n.s. + *** n.s. + ** 60 50 50 20 20 80 80
60
+ + 40 40 15 15 60 60 40 40 30 α TNF- , 30 α TNF- , + 10 + 40 γ 10 γ 40 20 20 20 20 5 20
%KLRG1 5 %KLRG1 10 10 20
0 % IFN- 0 Survival proportions: Survival of Two groups0 0 0 0 % IFN- 0 0 D E P815 tumor burden P815 tumor rejection P815 tumor burden (day 7) WT WT let-7 WT WT let-7 let-7 let-7
) * 2506 250 100 Wild Type WT WT let-7 Tg 200 200 let-7Tg let-7Tg 150 150 5050 100 100 x X10^6 50 50 %survival Percent survival Percent 0 0 0 0 -50 -50 0 5 10 15 20 Numbercellsof (x10 Days 0 5 10 15 20
WT ilet-7 Days elapsed
Figure 2 Let-7 in P14+/ ilet-7 P14+ Survival proportions: Survival of P815 CD8
4 A let-7 expression in P14+ CD8 T cells B P815 tumor rejection 3 3 let-7a let-7a 100 Injected 100 Rag2-/- + CD8 T cells let-7b let-7b Uninjected -/- let-7c let-7c Rag2 2 2 let-7d let-7d let-7e let-7e 50 50 let-7f miRNA expression 1 1 let-7f let-7g %survival Percent survival Percent let-7g let-7i Normalized miRNA NormalizedmiRNA let-7i 0 0 0 0 expression(relative U6)to 0 4 8 12 16 20 24 WT WT WT WT let-7Tg WT let-7Tg WT 0 4 8 12 16 20 24 ilet-7 ilet-7 Naïve Effector Days elapsed
Figure 2- figure supplement 1 A Proliferation of CD8 LNT B CD8 LNT
) Cdc25a RNA Ccnd2 RNA Ccnf RNA Cdk6 RNA Ccnd2 Ccnf Cdk6 100 Cdc25a n.s. 30 250250 *** 0.040.04 30 *** WT WT n.s. WT *** WT 80 Rpl13a 200 ilet-7 *** ilet-7 ilet-7 ilet-7 *** 200 0.030.03 ** 20 Lin28B Lin28B 20 Lin28B Lin28B 60 150150 0.020.02 40 100100 10 10 0.010.01 20 50 0 0 0.00 0 Events(%) Relativeexpression 0 0 0 0
0 (normalizedto Naive Activated Naive Activated Naive Activated Naive Activated 10010 1 102 103 104 CTV WT let-7Tg Lin28Tg
WT let-7Tg Lin28Tg Day 0 Day 2 Day 0 Day 2 Day 0 Day 2 Day 0 Day 2
C CD8Myc LNT D CD8 LNT Myc Glut1 RNA Glut3 RNA Gpd2 RNA Glut1 Glut3 Gpd2 150 WT 60 *** 20 2.52.5 ** *** *** WT * WT n.s. WT ) let-7 Tg ilet-7 2.0ilet-7 ilet-7 *** 15 *** 2 Lin28B Tg 40 Lin28B Lin28B Lin28B 100 1.51.5 10 ) 1.0 1 Rpl13a 20 5 5 5050 *** 0.50.5 0 0 0 0 0.0 0 *** RPL13a Naive Activated Naive Activated Naive Activated 0 0 Hk2 RNA 0 2 3 Pfk1 RNA Tpi RNA Tfap4 Pfk1 Hk2 Tpi Time (days) 0.8 20 Day 3 500 *** 0.8 Day 0 Day* 3 20 Day 0 ** 500 Day 0 Day 3 Day 0 Day 2 Day 3 WT * WT ** WT Tfap4 * ilet-7 400ilet-7 ilet-7 0.6 15 400 0.6 Lin28B 2.02.0 Lin28B Lin28B WT 300300 0.40.4 10 n.s. let-7 Tg 200200 1.5 1.5 0.2 5 5 Lin28B Tg 0.2 100100 1.01.0 0.0 0 0 0 0 0 ** Naive Activated Naive Activated Naive Activated
Relativeexpression (normalized to 0.50.5 ** Pkm RNA Ldha RNA Yars RNA Pkm Ldha Yars
8000 Day 0 Day 3 Day 3 10001000 Day 0 Day 3 8000 300300 Day 0 * 0.00.0 ** * WT n.s. WT WT
Relativeexpression (normalized to *** 0 2 3 800 *** ilet-7 *** ilet-7 ilet-7 800 60006000 Lin28B 200200Lin28B Lin28B Time (days) 600600 Day 0 Day 2 Day 3 40004000 400400 WT let-7Tg Lin28Tg 100100 2000 200200 2000 0 0 0 0 0 0 Naive Activated Naive Activated Naive Activated
WT let-7Tg Lin28Tg Day 3 Day 3 Day 0 Day 0 Day 3 Day 0
Figure 3 B A Lin28Tg Let-7Tg Wild type CD4 Normalized -10 miRNA expression 10 10 10 10 let-7 expression in polyclonal CD8 CD8 polyclonal in expression let-7 T cells CD8 -10 2 2 0 1 2 3 0 miRNA expression 3 4 5
0 1 2 3 Gated on live cells cells live Gated on 2
0
WT 10 Pre-sort: gated on on Pre-sort: gated 2
WT Naïve Naïve 10
ilet-7 polyclonal 1C Figure 24.6% 37.7% 33.7% 3
let-7Tg 10 Lin28 4
Lin28Tg 10
WT 5
Effector WT ilet-7 singlets let-7Tg CD25 -10
Lin28 10 10 10 10 CD44 -10 2 2 0 3 4 5 and live cells cells live and
Lin28Tg Gated onCD8 2 50.0% 23.6% 48.5%
0 10 let-7i let-7i let-7g let-7g let-7f let-7f let-7e let-7e let-7d let-7c let-7c let-7a let-7a let-7b let-7b 2
10 let-7i let-7g let-7f let-7e let-7d let-7c let-7b let-7a 3
10 + CD4 4
10 5 -
Figure 3- figure supplement 1
CD4 -10 10 10 10 10 CD8 2 3 4 5 2 0
10 Gated on live cells cells live Gated on 0
10 Post-sort: gated on Post-sort:on gated 1
10 2
99.2% 10 98.3% 99.0% 3
10 4
10 5
singlets CD25 -10 10 10 10 10 CD44 -10 2 2 0 3 4 5
and live cells cells live and Gated onCD8 2 99.5% 99.7% 99.8%
0 10 2
10 3
10 + 4 CD4
10 5
-
WT (GP33) WT no Dox (GP33) let-7 TG (GP33) WT+Dox (GP33) let-7 KD (GP33) 4Tg TG no Dox (GP33) WT (NP396) 4Tg+Dox (GP33) let-7 TG (NP396) WT no Dox (NP396) A Cytotoxicity assay let-7 KD (NP396) B WT+Dox (NP396) Cytotoxicity assay 4Tg TG no Dox (NP396) 4Tg TG + Dox (NP396) 100100 WT 100100 WT *** let-7Tg gp33 ** WT + Dox 80 80 *** 80 80 ** gp33 Lin28Tg 4Tg 60 60 60 60 *** ** 4Tg+ Dox WT n.s. n.s. 40 40 *** 40 40 n.s. *** let-7Tg np396 WT % Target lysis 20 20
% Target lysis n.s. 20 *** 20 * n.s. WT + Dox %Target lysis %Target Lin28Tg 0 ** 0 4Tg np396 0 * lysis %Target 0 -20 4Tg+ Dox -20-20 -20 5:1 1:1 10:1 2:1 2:5 5:1 1:1 1:51:5 1:251:10 20:1 2:1 2:5 Effector:Target Effector:Target Effector: Target Effector: Target C CD8 effector cells (5 days) GranulationSSC-A MFI (SSC-A) 10 ** 8 1.51.5 n.s. 6 4 1.01.0 WT 2 let-7Tg 0.50.5 0
SSC Lin28Tg 0 2 4 6 8 10 Area Normalized 0.0 0 FSC
WT D CTL granules E ilet-7 Lin28 CD8 T cells Granzyme B RNA Granzyme A RNA Prf1 RNA
Granzyme A Granzyme B ) Gzmb Gzma Prf1 n.s. 2500 150150 *** *** 15001500 * 2500 *** WT *** n.s. WT WT *** *** ilet-7 1010 * 3030 20002000 ilet-7 ilet-7 9 RPL13a Lin28B 100 Lin28B Lin28B 8 10001000 100 8 15001500 7 20 6 20 6 10001000 50 5 500500 50 4 4 500 3 1010 500 22 0
Number of granules of Number 0 0 0
Number of granules of Number 1 0 0 0 0 Naive Effector Relativeexpression Numbergranulesof 0 0 Naive Effector Naive Effector WT iLET7 LIN28B WT iLET7 LIN28B (normalizedto
WT let-7Tg Lin28Tg Day 0 Day 5 let-7Tg Lin28Tg Day 0 Day 5WT Day 0 Day 5
F Effector CD8 T cells Effector CD8 T cells IFN-g protein WT let-7Tg Lin28Tg GranzymeGranzyme A protein A Granzyme B protein B IFN-γ 104 18.3 80.2 37.4 52.7 8.10 91.8 n.s. * 4040 3 3 * 3 3 n.s. 3 10 * B * 2 3030 10 2 2 2 2 20 101 20 1 1 1 1 10 0 0.74 0.77 6.65 3.24 0.06 0.04 10
Granzyme 10 0 1 2 3 4 10 10 10 10 10 Normalized MFI 0 0 0 0 0 0 IFN-γ WT let-7Tg Lin28Tg WT WT 4 ilet-7 ilet-7 10 Lin28 Lin28 85.3 13.8 97.0 1.25 WT 1.01 98.3 ilet-7 Lin28 103 Effector CD8 T cells B G 102 WT Lin28Tg let-7Tg 101 Perforin 0 0.95 0 1.72 0 0.30 0.43
Granzyme 10 100 101 102 103 104 Granzyme A Actin Figure 4 Effector CD8 T cells
Prf1Prf1 RNA GzmaGzma GzmbGzmb *** *** *** 30 30 30 30 4 4 3 3 WT 20 20 20 20 WT + Dox 2 2 10 10 10 10 4Tg 1 1
Foldchange 4Tg+ Dox 0 0 0 0 0 0
WT 4Tg WT 4Tg WT 4Tg
WT +Dox WT +Dox WT +Dox 4Tg + Dox 4Tg + Dox 4Tg + Dox
Figure 4- figure supplement 1 A CD8 T cells B Effector CD8 T cells Effector CD8 T cells Eomes RNA Eomes EomesEomes protein 10.1 87.7 ) *** * 4040 3 3 *** WT ** ilet-7 30 Rpl13a 30
Lin28B WT 2 2 2020 0.74 1.39 1 1 1010 20.0 78.4 Normalized MFI
Relativeexpression 0 0 0 0 (normalizedto Tbx21 RNA T-betT-bet protein Tbx21 WT ** ilet-7n.s. Lin28 ) Day 0 Day 5 let-7Tg n.s. 20 20 *** 1.5 1.5 WT ilet-7 1.13 0.43 15 Rpl13a 15 Lin28B 104 1.0 1.0 0.98 96.2 10 10 103 0.5 0.5 5 5 102 Normalized MFI
Relativeexpression 0 0 0.0 Lin28Tg 1 0.0 (normalizedto Naive Effector 10 0.19 100 2.66 WT let-7Tg Lin28Tg WTWT let-7Tg Lin28Tg T-bet T-bet 0 1 2 3 4 Day 0 Day 5 10 10 10 10 10 ilet-7 Lin28 Eomes C D E Effector CD8 T cells Eomes mRNA: let-7 seed region NIH 3T3 Eomes Eomes eomes luc assay *** ) 3 3 *** 55 5’UTR ORF 3’UTR 44 2 2 mouse Renilla CGCTACTACCTCCC 3 human CGGTACTACCTCCA 3 22 1 1 mouse CGCTTAACCTCCC Firfly/Renilla Foldchange (mutated) 11 0 0 00 (normalizedto Vector WT Mut Firefly-Luciferaseactivity Vectoralone alone WT let-7 seedMut let-7 seed WT let-7 seed region WT 4Tg in eomes mRNA WT +Dox WT + Dox 4Tg + Dox 4Tg 4Tg+ Dox
Figure 5 A Effector CD8 T cells
Hes1 RNA Hey1 RNA Heyl RNA Dtx1 RNA ) Hes1 Hey1 Heyl Dtx1 0.06 0.5 1.01.0 150 0.06 0.5 WT WT 150 WT WT ilet-7 Rpl13a 0.40.4 ilet-7 0.80.8 ilet-7 WTilet-7 0.04 Lin28B Lin28B 100100 Lin28B Lin28B 0.04 0.3 0.60.6 0.3 let-7Tg 0.20.2 0.40.4 0.020.02 5050 0.2 Lin28Tg 0.10.1 0.2 N.D. N.D. 0.000 0.00 0.00.0 0 Relativeexpression
(normalizedto Naive Effector Naive Effector Naive Effector Naive Effector B Day 0 Day 5 EffectorDay 0 CD8 DayT 5 cells Day 0 Day 5 Day 0 Day 5
Zbtb32 RNA Blimp-1 RNA Runx3d RNA ) Zbtb32 Prdm1 Runx3d 80 1.51.5 40 80 WT 40 WT WT
Rpl13a ilet-7 ilet-7 ilet-7 6060 3030 WT 1.01.0 Lin28B Lin28B Lin28B 4040 2020 let-7Tg 0.50.5 2020 1010 Lin28Tg 0 0.00.0 0 Relativeexpression
(normalizedto Naive Effector Naive Effector Naive Effector
Day 0 Day 5 Day 0 Day 5 Day 0 Day 5
Figure 5- figure supplement 1 NIH 3T3 NIH 3T3 let-7 expression
1500 NIH 3T3 A NIH 3T3 B 4 4
10001000 let-7a let-7a ) let-7b let-7b 3 3 let-7c let-7c let-7d let-7d Renilla Legend 500 500 let-7e let-7e
miRNA expression 2 let-7f 2 let-7f let-7g let-7g let-7i (normalizedU6)to Relativeexpression 0 0 let-7i 1 1 (normalizedto Firefly-luciferaseactivity
0 h Time of TCR stimulation 0 0 Empty vector + ------let-7b - + - - - - - let-7b (mutated) - - + - - - - Relative luminescence (normalized to Renilla) Vector alonelet-7blet-7b mutatedlet-7glet-7g mutatedlet-7ilet-7i mutated let-7g - - - + - - - let-7g (mutated) - - - - + - - let-7i - - - - - + - let-7i (mutated) ------+
Figure 5- figure supplement 2 A Effector CD8 T cells B Effector CD8 T cells
Eomes RNAEomes Day 5 (normalized) EomesEomes protein Gzma RNAGzma Day 5 (normalized) Prf1 RNAPrf1 Day 5 *** *** *** *** *** 2 2 ** 22 2020 8 8 *** *** Cre-B-P14+Efl/- Cre-B-P14+Efl/- Cre-B-P14+Efl/- Cre-B+P14+Ewt Cre-B+P14+Ewt Cre-B+P14+Ewt 1515 6 6 Cre+B+P14+Efl/- Cre+B+P14+Efl/- Cre+B+P14+Efl/- Cre+B+P14+E-/- Cre+B+P14+E-/- Cre+B+P14+E-/- 1 1 1 1 1010 4 4 5 5 2 2 Foldchange Foldchange 0 0 Normalized MFI 0 0 0 0 0 0
P14+ CTLs: CD4Cre-Eomesfl/wt P14+ CTLs: CD4Cre-Eomesfl/wt - wt - wt Day 5 CD4Cre Eomes Lin28Tg CD4CreDay 5 Eomes Lin28TgDay 5 Cre- B- P14+Cre- Efl/- B+ P14+Cre+ Ewt B+ P14+Cre+ Efl/- B+ P14+ E-/- CD4Cre+Eomesfl/wtLin28Tg CD4Cre+Eomesfl/wtLin28Tg + fl/fl Cre- B- P14+ Efl/-+ (GP33) fl/fl CD4Cre Eomes Lin28Tg CD4CreCre- B+ P14+ Ewt Eomes(GP33) Lin28Tg Cre+ B+ P14+ Efl/- (GP33) Cre+ B+ P14+ E-/- (GP33) Cre- B- P14+ Efl/- (NP396) C Granzyme A D Cytotoxicity assayCre- B+ P14+ Ewt (NP396) Cre+ B+ P14+ Efl/- (NP396) Cre+ B+ P14+ E-/- (NP396) 100 P14+ CTLs: 4.21% 8080 CD4Cre-Eomesfl/wt 6060 CD4Cre-Eomeswt Lin28Tg *** gp33 50.8% 4040 ** CD4Cre+Eomesfl/wt Lin28Tg + fl/fl % Target lysis 20 *** CD4Cre Eomes Lin28Tg 33.7% 20 ** *** ** % Target lysis Target % 0 0 CD4Cre-Eomesfl/wt 1:1 1:5 1:25 25.6% -20-20 CD4Cre-EomeswtLin28Tg 1:1 1:5 1:25 + fl/wt np396 100 101 102 103 104 Effector:Target CD4Cre Eomes Lin28Tg Effector: Target CD4Cre+Eomesfl/flLin28Tg P14+ CTLs: CD4Cre-Eomesfl/wt CD4Cre-EomeswtLin28Tg CD4Cre+Eomesfl/wtLin28Tg CD4Cre+Eomesfl/flLin28Tg
Figure 6 Effector CD8 T cells
Granzyme B
Granzyme n.s.B protein n.s. 33
22
11
Normalized MFI 00 P14+ CTLs: CD4Cre-Eomesfl/wt - wt Cre-CD4Cre B- P14+Cre- Efl/- B+ P14+Cre+ Ewt B+Eomes P14+Cre+ Efl/- B+ P14+ E-/- Lin28Tg CD4Cre+Eomesfl/wtLin28Tg CD4Cre+Eomesfl/flLin28Tg
Figure 6- figure supplement 1 Cre+ B+ P14+ E-/- +empty virus (GP33) Cre+ B+ P14+ E-/- +eomes virus(GP33) Cre+ B+ P14+ E-/- +empty virus (NP396) Cre+ B+ P14+ E-/- +eomes virus(NP33)
A Cytotoxicity assay 100100 *** P14+CD4Cre+Eomesfl/flLin28Tg CTLs: 8080 *** 6060 RV- Empty gp33 4040 RV- Eomes
% Target lysis 2020 RV- Empty np396 % Target lysis Target % 0 0 RV- Eomes -20 1:1 1:5 1:1 1:5 Effector:Target Effector: Target
B Effector CD8 T cells
Eomes RV-->EomesEomes KO (Eomes EomesRNA) (normalized) RV-->Eomes GzmaKO (Gzma Eomes RNA) (normalized)RV-->EomesGzmb KO (Gzmb RNA)Eomes (normalized) RV-->EomesPrf1 KO (Prf1 RNA) (normalized) *** *** n.s. *** 15001500 2.52.5 1.51.5 150 2.02.0 1000 1.01.0 100 1000 1.51.5 100 1.01.0 500 0.50.5 5050 0.50.5 Foldchange 0 0 0.0 0 0.0 0 0 0 P14+CD4Cre+Eomesfl/flLin28Tg CTLs: RV- Empty RV- Eomes
Day 5 Day 5 Day 5 C Day 5 Eomes (449-459): Tyr Tyr Leu
let-7 +empty virus (GP33) Wild type let-7CTACTACCTC +eomes mut virus(GP33) let-7 +empty virus (NP396) Mutant let-7CTACTA +eomes mut virus(NP33)TTTA
Cytotoxicity assay 100 8080 P14+let-7Tg CTLs: 6060 ** *** RV- Empty 4040 gp33 RV- EomesMUTANT % Target lysis 2020
% Target lysis Target % 0 0 RV- Empty MUTANT np396 -20-20 1:1 1:5 RV- Eomes Effector:Target1:1 1:5
Effector: Target D Effector CD8 T cells Eomes Gzmb EomesMUT RV-->let-7Tg (EomesEomesMUT RNA) (normalized) RV-->let-7TgGzma (GzmaEomesMUT RNA) (normalized) RV-->let-7Tg (GzmbEomesMUT RNA) (normalized) RV-->let-7TgPrf1 (Prf1 RNA) (normalized) *** *** ** 6 6 2.52.5 2 2 2 n.s. 2 2.02.0 4 4 1.51.5 1 1 1 1 1.01.0 2 2 0.50.5 Foldchange 0 0 0.00.0 0 0 0 0
P14+let-7Tg CTLs: RV- Empty
Day 5 Day 5 MUTANT RV- Eomes Day 5 FigureDay 5 6- figure supplement 2 TCR
LET-7
ACQUISITION PROLIFERATION OF FUNCTION
NAÏVE BLAST CTL
Figure 6- figure supplement 3